University of Southern California Dissertations and Theses

Age, sex, and task difficulty as predictors of social conformity: A search for general tendencies of conformity behavior

(USC Thesis Other)

Age, sex, and task difficulty as predictors of social conformity: A search for general tendencies of conformity behavior

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content ÂGE, SEX, ME TASK EIEEICULÏY AS PHEEIGTOBS OF i i SOCIAL OOEFOBMIIY: A SBAEGE FOR GENERAL TENDENCIES OF. CONFORMITY BEHAVIOR Ronald Lawrence Klein I M A Eiseertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (PsYchology) August 1971 UMI Number: DP30477 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. Dissertation Pubi sknq UMI DP30477 Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106- 1346 • Copyright by RONALD LAWRENCE KLEIN 1971 UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES, CALIFORNIA 90007 This dissertationf w ritte n by .......... under the direction of i s . . . . Dissertation Com mittee, and approved by a ll its members, has been presented to and accepted by The G rad u ate School, in p a rtia l fu lfillm e n t of require ments of the degree of D O C T O R O F P H I L O S O P H Y ... Dean Da t e . . . 3 . ept.emb.er. . . 1 . 9 7 . 1 . ;SERTATION COMMITTEE I would like to express my appreciation to the mem bers of my dissertation committee, Dr, James Birren, Dr. Norman Cliff and Dr. Tern Bengtson for their valuable assistance in the preparation of this dissertation and for their advice and guidance during my academic career at the University of Southern California. I am especially indebted to Dr, James Birren, under whom this investigation was begun. He provided that indefinable support indispensable at the outset of one's research endeavors. I also wish to thank Dr. Jaeek Szafran^ who along with Dr. Birren, has influenced my thinking, research interest and the carrying through of the present undertak ing. I wish to extend my gratitude to Dr. Nick Pappas, and to my cousin, Gary Klein, for assisting in the prepara tion of some of the stimulus material. To Dean Leroy Hixson and Mr. Joseph Gunn, of the Institute of Lifetime Learhjrig, a special thanks for allow ing me to use their facilities and for providing subjects. This research was conducted while I was a N.I.C.H.D, (HD 001 $7-0%) trainee in Psychology at the Gerontology Center of the University of Southern California, I would like to extend my appreciation to the Gerontology Center and its personnel, especially Mrs. Eleanor James, Miss Gloria Haether, and Mrs* Eileen Salter, for the excep tional support and services provided. I would also like to acknowledge the National Science Foundation who awarded me a grant to support my dissertation research. % very special appreciation goes to my wife for her assistance throu^i editorial criticism and especially for her understanding and patience during my graduate career. Finally, I would like to thank my parents, Mr, and Mrs. Alex Klein, for their encouragement, under standing and assistance throughout my graduate training and life. ii TABLE OE CONTENTS ! Page j ILIST CE TABLES............................... vi I |LIST OF FIGUEES.......................... xxi j DISSERTATION ABSTRACT.......................... xxx I Chapter I. GENERAL INTRODUCTION.................. 1 Movement vs. Congruence Conformity . . 2 Normative vs. Informational Social Influence ................ 4 Desirability vs. Undesirability of Conformity ...., ........... 6 Historical Landmarks in the Area of Social Conformity ............. 8 The Present Investigation......... 1$ The Generality of Conformity Behavior .................. 19 Age and Conformity................ 25 Sex and Conformity................ 51 Task Difficulty and Conformity .... 56 Nature of the Experimental Task ... 62 Hypotheses .............. 64- Compliance vs. Private Acceptance . . 65 II. GENERAL METHOD........................ 73 Subjects................ 73 Conformity Apparatus.............. 74- General Procedure................ 78 General Data Analysis............. 83 III. EXPERIMENT I .... .................. 86 PROCEDURES......................... 87 Method .................... 88 Data Analysis .............. 91 RESULTS........................... 92 111 Chapter Page IV. EXPERIMENT I I ........................ 109 PROCEDURES......................... 110 Method........................... Ill Data Analysis.................... 114 RESULTS........................... 115 V. EXPERIMENT III........................ 152 PROCEDURES......................... 254 Method........................... 155 Data Analysis.................... 139 RESULTS........................... 140 VI. EXPERIMENT I V ........................ 157 PROCEDURES......................... 158 Method........................... 159 Data Analysis.................... 163 RESULTS........................... 164 VII. EXPERIMENT V .......................... 182 PROCEDURES......................... 184 Method........................... 185 Data Analysis.................... 188 RESULTS........................... 189 VIII. GENERAL RESULTS...................... 206 Age ........................... 206 S e x ............................. 218 Task Difficulty.................. 226 Nature of the Task................ 240 Other Significant Effects ....... 248 PostexpGriment Questionnaire ....... 249 IX. DISCUSSION............................ 256 A g e ............................. 256 iv Chapter Page S e x ............................. 270 Task Difficulty.................. 272 Nature of the Task .......... 275 Extremeness of the Norm........... 275 General Tendency vs. Trait......... 277 Future Research.................. 279 Ethical Concerns . . . . . . . . . 284 X. SUMMARY AND CONCLUSION................ 286 ; APPENDIX.................................... 295 I BIBLIOGRAPHY................................. 524- V LIST OF TABLES Table Page 1• Complete Analysis of Variance Summary Table for Compliance Experiment I . . . • 93 2. Complete Analysis of Variance Summary Table for the First Test of Private Acceptance Experiment I .............. 94 3. Complete Analysis of Variance Summary Table for the Second Test of Private Acceptance Experiment I .............. 93 4. Experiment I Mean Conformity Scores (Compliance) of Old and Young Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 1 ) .................. ^6 5. Experiment I Mean Conformity Scores (1st Private Acceptance) of Old and Young Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 2 ) ....... 97 6. Experiment I Mean Conformity Scores (2nd Private Acceptance) of Old and Young Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 3) ..... 98 7. Experiment I Mean Conformity Scores (Compliance) of Male and Female Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 1 ) ....... 100 8. Experiment I Mean Conformity Scores (ist Private Acceptance) of Male and Female Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 2 ) ....... 101 __________________________vi_________________________ Table Page 9. Experiment I Mean Conformity Scores (2d Private Acceptance) of Male and Female Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 3) ....... 102 10. Experiment I Mean Conformity Scores (Compliance) of Subjects for the Three Levels of Task Difficulty (Combining Age and Sex). F Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 1 ) ....... .................... 104 11. Experiment I Mean Conformity Scores (1st Private Acceptance) of Subjects for the Three Levels of Task Difficulty (Combining Age and Sex). F Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 2) . . . 105 12. Experiment I Mean Conformity Scores (2d Private Acceptance) of Subjects for the Three Levels of Task Difficulty (Combining Age and Sex). F Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 5) * . . 106 15. Complete Analysis of Variance Summary Table for Compliance Experiment II ... 116 14. Complete Analysis of Variance Summary Table for the First Test of Private Acceptance Experiment I I ............. 117 15. Complete Analysis of Variance Summary Table for the Second Test of Private Acceptance Experiment II ............. 118 16. Experiment II Mean Conformity Scores (Compliance) of Old and Young Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 15) .................. vii 119 Table Page 17- Experiment II Mean Conformity Scores (1st Private Acceptance) of Old and Young Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 14).......... 120 18. Experiment II Mean Conformity Scores (2d Private Acceptance) of Old and Young Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 15)....... 121 1 9. Experiment II Mean Conformity Scores (Compliance) of Male and Female Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 15).................. 125 20. Experiment II Mean Conformity Scores (1st Private Acceptance) of Male and Female Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 14)....... 124 21. Experiment II Mean Conformity Scores (2d Private Acceptance) of Male and Female Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 15)....... 125 22. Experiment II Mean Conformity Scores (Compliance) of Subjects for the Three Levels of Task Difficulty (Combining Age and Sex). F Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 15) ........................... 127 viii Table Page 25. Experiment II Mean Conformity Scores (1st Private Acceptance) of Subjects for the Three Levels of Task Difficulty (Combining Age and Sex). P Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 14)....... 128 24. Experiment II Mean Conformity Scores %2d Private Acceptance) of Subjects for the Three Levels of Task Difficulty (Combining Age and Sex). P Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 15) ....... 129 25. Complete Analysis of Variance Summary Table for Compliance Experiment III ... 141 26. Complete Analysis of Variance Summary Table for the First Test of Private Acceptance Experiment III ............. 142 27. Complete Analysis of Variance Summary Table for the Second Test of Private Acceptance Experiment III .... 145 28. Experiment III Mean Conformity Scores (Compliance) of Old and Young Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 25) .................. 144 29. Experiment III Mean Conformity Scores (1st Private Acceptance) of Old and Young Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 26) ....... 145 50. Experiment III Mean Conformity Scores (2d Private Acceptance) of Old and Young Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 27) ....... ... .. ix.. 146 Table Page 51. Experiment III Mean Conformity Scores (Compliance) of Male and Female Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 25) ....... 148 52. Experiment III Mean Conformity Scores %1 st Private Acceptance) of Male and Female Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 26) ........................... 149 55. Experiment III Mean Conformity Scores (2d Private Acceptance) of Male and Female Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 27) ........................... 150 54. Experiment III Mean Conformity Scores (Compliance) of Subjects for the Three Levels of Task Difficulty (Combining Age and Sex). F Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 25) ........................... 152 55. Experiment III Mean Conformity Scores (1st Private Acceptance) of Subjects for the Three Levels of Task Difficulty (Combining Age and Sex). F Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 26) . . 155 56. Experiment III Mean Conformity Scores (2d Private Acceptance) of Subjects for the Three Levels of Task Difficulty (Combining Age and Sex). F Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 27) . . 154 57. Complete Analysis of Variance Summary Table for Compliance Experiment IV ... X 165 Table Page 58. Complete Analysis of Variance Summary Table for the First Test of Private Acceptance Experiment I V ............... 156 59- Complete Analysis of Variance Summary Table for the Second Test of Private Acceptance Experiment IV ... . 167 40. Experiment IV Mean Conformity Scores (Compliance) of Old and Young Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 57)..................... 168 41. Experiment IV Mean Conformity Scores (1st Private Acceptance) of Old and Young Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 58)....... 169 42. Experiment IV Mean Conformity Scores (2d Private Acceptance) of Old and Young Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 59)........ 170 4 5. Experiment IV Mean Conformity Scores (Compliance) of Male and Female Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 57)........ 175 44. Experiment IV Mean Conformity Scores (ist Private Acceptance) of Male and Female Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 58) . . ...................... 174 XI Table Page 4 5. Experiment IV Mean Conformity Scores (2d Private Acceptance) of Male and Female Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 39) ....... 175 46. Experiment IV Mean Conformity Scores (Compliance) of Subjects for the Three Levels of Task Difficulty (Combining Age and Sex). F Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 37) ........................... 176 4 7. Experiment IV Mean Conformity Scores ( 1st Private Acceptance) of Subjects for the Three Levels of Task Difficulty (Combining Age and Sex). F Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 38) ............... 177 48. Experiment IV Mean Conformity Scores (2d Private Acceptance) of Subject for the Three Levels of Task Difficulty (Combining Age and Sex). F Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 59) .................. 178 4 9. Complete Analysis of Variance Summary Table for Compliance Experiment V . . . . 190 50. Complete Analysis of Variance Summary Table for the First Test of Private Acceptance Experiment V .............. 191 51. Complete Analysis of Variance Summary Table for the Second Test of Private Acceptance Experiment V .............. 191 52. Experiment V Mean Conformity Scores (Compliance) of Old and Young Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 49) = « * . » xii 195 Table Page 55. Experiment V Mean Conformity Scores (1st Private Acceptance) of Old and Young Subjects (Combining the Three Levels of Task Difficulty). P Value for Main Effect of Age (Taken from Analysis of Variance— Table 50)........ 194 54. Experiment V Mean Conformity Scores (2d Private Acceptance) of Old and Young Subjects (Combining the Three Levels of Task Difficulty). E Value for Main Effect of Age (Taken from Analysis of Variance— Table 51)........ 195 55. Experiment V Mean Conformity Scores (Compliance) of Male and Female Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 49)........ 197 56. Experiment V Mean Conformity Scores (1st Private Acceptance) of Male and Female Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 50)........................... 198 57* Experiment V Mean Conformity Scores (2d Private Acceptance) of Male and Female Subjects (Combining the Three Levels of Task Difficulty). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 51) .... 199 58. Experiment V Mean Conformity Scores (Compliance) of Subjects for the Three Levels of Task Difficulty (Combining Age and Sex). F Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 49) . . ...................... 201 Xlll Table Page 59* Experiment V Mean Conformity Scores (1st Private Acceptance) of Subjects for the Three Levels of Task Difficulty (Combining Age and Sex). P Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 5 0 ) ............... 202 60. Experiment V Mean Conformity Scores (2d Private Acceptance) of Subjects for the Three Levels of Task Difficulty (Combining Age and Sex). E Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 5 1 ) ............... 205 61. Complete Analysis of Variance Summary Table for Compliance Combined Analysis........................... 207 62. Complete Analysis of Variance Summary Table for the First Test of Phivate Acceptance Combined Analysis ......... 208 65. Complete Analysis of Variance Summary Table for the Second Test of Private Acceptance Combined Analysis ....... 209 64. Mean Conformity Scores (Compliance) of Old and Young Subjects (Averaged over the Five Experimental Tasks). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 61)........................... 211 65. Mean Conformity Scores (1st Private Acceptance) of Old and Young Subjects (Averaged over the Five Experimental Tasks). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 62).................. 212 66. Mean Conformity Scores (2d Private Acceptance) of Old and Young Subjects (Averaged over the Five Experimental Tasks). F Value for Main Effect of Age (Taken from Analysis of Variance— Table 65).................. 215 XIV Table Page 67. Mean Conformity Scores (Compliance) of Young and Old Subjects for Each of the Five Experiments. F Values for Main Effect of Age (Taken from Separate Analysis of Variance Analyses) ....... 214 68. Mean Conformity Scores (1st Private Acceptance) of Young and Old Subjects for Each of the Five Experiments. F Values for Main Effect of Age (Taken from Separate Analysis of Variance Analyses) .................. 215 69. Mean Conformity Scores (2d Private Acceptance) of Young and Old Subjects for Each of the Five Experiments. F Values for Main Effect of Age (Taken from Separate Analysis of Variance Analyses) .................. 216 70. Mean Conformity Scores (Compliance) of Male and Female Subjects (Averaged Over the Five Experimental Tasks). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 61) ........................... 220 71. Mean Conformity Scores (1st Private Acceptance) of Male and Female Subjects (Averaged Over the Five Experimental Tasks). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 62) ....... 221 72. Mean Conformity Scores (2d Private Acceptance) of Male and Female Subjects (Averaged Over the Five Experimental Tasks). F Value for Main Effect of Sex (Taken from Analysis of Variance— Table 65) ....... 222 75. Mean Conformity Scores (Compliance) of Male and Female Subjects for Each of the Five Experiments. F Values for Main Effect of Sex (Taken from Separate Analysis of Variance Analyses) .................. X V 225 Table Page 74. Mean Conformity Scores (1st Private Acceptance) of Male and Female Subjects for Each, of the Five Experiments. F Values for Main Effect of Sex (Taken from Separate Analysis of Variance Analyses)....... 224 75. Mean Conformity Scores (2d Private Acceptance) of Male and Female Subjects for Each of the Five Experiments. F Values for Main Effect of Sex (Taken from Separate Analysis of Variance Analyses)....... 225 76. Mean Conformity Scores (Compliance) of the Three Levels of Task Difficulty (Averaged Over the Five Experimental Tasks). F Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 61)..................... 228 77. Mean Conformity Scores (1st Private Acceptance) of the Three Levels of Task Difficulty (Averaged Over the Five Experimental Tasks). F Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 62) ........................... 229 78. Mean Conformity Scores (2d Private Acceptance) of the Three Levels of Task Difficulty (Averaged Over the Five Experimental Tasks). F Value for Main Effect of Task Difficulty (Taken from Analysis of Variance— Table 65) ........................... 250 79. Comparison between Levels of Task Difficulty on the Combined Data of the Five Experiments ............... 252 80. Mean Conformity Scores (Compliance) of the Three Levels of Task Difficulty for Each of the Five Experiments. F Values for Main Effect of Task Difficulty (Taken from Separate Analysis of Variance Analyses) ........................... xvi 254 Table Page 81. Mean Conformity Scores (1st Private Acceptance) of the Three Levels of Task Difficulty for Each of the Five Experiments. F Values for Main Effect of Task Difficulty (Taken from the Separate Analysis of Variance Analyses) .................. 256 82. Mean Conformity Scores (2d,Private Acceptance) of the Three Levels of Task Difficulty for Each of the Five Experiments. F Values for Main Effect of Task Difficulty (Taken from the Separate Analysis of Variance Analyses).................. 257 85. Summary Table of the Comparisons between the Levels of Task Difficulty Made in Experiments I— V ............. 259 84. Mean Conformity Scores (Compliance) of the Five Experimental Tasks (Averaged over Age, Sex and Task Difficulty). F Value for Main Effect of Experimental Task (Taken from Analysis of Variance— Table 61) ....... 241 85. Mean Conformity Scores (1st Private Acceptance) of the Five Experimental Tasks (Averaged over Age, Sex and Task Difficulty). F Value for Main Effect of Experimental Task (Taken from Analysis of Variance— Table 62) . . 242 86. Mean Conformity Scores (2d Private Acceptance) of the Five Experimental Tasks (Averaged over Age, Sex and Task Difficulty). F Value for Main Effect of Experimental Task (Taken from Analysis of Variance— Table 65) . . 245 87. Comparisons between Experimental Tasks ............................... 246 88. Responses Given in the Postexperiment Questionnaire Data Combined from the Five Experiments .................... xvii 250 Table Page 89. Conditions of Visual Perceptual Judgment Task— Experiment I ........... 298 90. Conditions of the Auditory Judgment Task— Experiment II ......... 500 91. Conditions of the Auditory Detection Task— Experiment III ....... 501 92. Conditions of the Problem Solving Task (for the Young Subjects) — Experiment IV ...................... 502 95. Conditions of the Problem Solving Task (for the Old Subjects) — Experiment IV ...................... 505 94. Conditions of the Social Attitude Task (for the Young Subjects) — Experiment V ...................... 507 95. Conditions of the Social Attitude Task (for the Old Subjects) — Experiment V ...................... 508 96. Experiment I Mean Conformity Scores (Compliance) for the Young, Old, Male, and Female Subjects at Each of the Three Levels of Task Difficulty. (Scores at Each Level of Difficulty Could Range from 0 to 8. ) ................ .......... 509 97. Experiment I Mean Conformity Scores (1st Private Acceptance) for the Young, Old, Male and Female Subjects at Each of the Three Levels of Task Difficulty. (Scores at Each Level of Difficulty Could Range from 0 to 8.) ........................... 510 98. Experiment I Mean Conformity Scores (2d Private Acceptance) for the Young, Old, Male and Female Subjects at Each of the Three Levels of Task Difficulty. (Scores at Each Level of Difficulty Could Range from 0 to 8. ) .......................... xviii 511 Table Page 99. Experiment II Mean Conformity Scores (Compliance) for the Young, Old, Male and Female Subjects at Each of the Three Levels of Task Difficulty. (Scores at Each Level of Difficulty Could Range from 0 to 8. ) ............ 312 100. Experiment II Mean Conformity Scores (1st Private Acceptance) for the Young, Old, Male and Female Subjects at Each of the Three Levels of Task Difficulty. (Scores at Each Level of Difficulty Could Range from 0 to 8 . ) ........................... 313 101. Experiment II Mean Conformity Scores (2d Private Acceptance) for the Young, Old, Male and Female Subjects at Each of the Three Levels of Task Difficulty. (Scores at Each Level of Difficulty Could Range from 0 to 8. ) ........................... 314 102. Experiment III Mean Conformity Scores (Compliance) for the Young, Old, Male and Female Subjects at Each of the Three Levels of Task Difficulty. (Scores at Each Level of Difficulty Could Range from 0 to 8. ) ............. 315 103. Experiment III Mean Conformity Scores (1st Private Acceptance) for the Young, Old, Male and Female Subjects at Each of the Three Levels of Task Difficulty. (Scores at Each Level of Difficulty Could Range from 0 to 8.) . . 516 104. Experiment III Mean Conformity Scores (2d Private Acceptance) for the Young, Old, Male and Female Subjects at Each of the Three Levels of Task Difficulty. (Scores at Each Level of Difficulty Could Range from 0 to 8 . ) ............ 517 XIX Table Page 105. Experiment IV Mean Conformity Scores (Compliance) for the Young, Old, Male and Female Subjects at Each of the Three Levels of Task Difficulty. (Scores at Each Level of Difficulty Could Range from 0 to 8. ) ............. 518 106. Experiment IV Mean Conformity Scores ( 1st Private Acceptance) for the Young, Old, Male and Female Subjects at Each of the Three Levels of Task Difficulty. (Scores at Each Level of Difficulty Could Range from 0 to 8 . ) ........................... 519 107. Experiment IV Mean Conformity Scores (2d Private Acceptance) for the Young, Old, Male and Female Subjects at Each of the Three Levels of Task Difficulty. (Scores at Each Level of Difficulty Could Range from 0 to 8 . ) ........................... 520 108. Experiment V Mean Conformity Scores (Compliance) for the Young, Old, Male and Female Subjects at Each of the Three Levels of Task Difficulty. (Scores at Each Level of Difficulty Could Range from 0 to 8. ) ............ 521 109. Experiment V Mean Conformity Scores (1st Private Acceptance) for the Young, Old, Male and Female Subjects at Each of the Three Levels of Task Difficulty. (Scores at Each Level of Difficulty Could Range from 0 to 8.) ........................... 110. Experiment V Mean Conformity Scores (2d Private Acceptance) for the Young, Old, Male and Female Subjects at Each of the Three Levels of Task Difficulty. (Scores at Each Level of Difficulty Could Range from 0 to 8 . ) ........................... 525 XX LIST OF FIGURES Figure Page 1. Mean conformity scores (compliance) of young and old subjects (combining the three levels of task difficulty) for Experiment I. Range of scores also indicated..................... . 96 2. Mean conformity scores (1st private acceptance) of young and old subjects (combining the three levels of task difficulty) for Experiment I. Range of scores also indicated........ 97 5. Mean conformity scores (2nd private acceptance) of young and old subjects (combining the three levels of task difficulty) for Experiment I. Range of scores also indicated.............. 98 4. Mean conformity scores (compliance) of male and female subjects (combining the three levels of task difficulty) for Experiment I. Range of scores also indicated....................... 100 5. Mean conformity scores (1st private acceptance) of male and female subjects (combining the three levels of task difficulty) for Experiment I. Range of scores also indicated........ 101 5. Mean conformity scores (2d private acceptance) of male and female subjects (combining the three levels of task difficulty) for Experiment I. Range of scores also indicated........ 102 7. Mean conformity scores (compliance) of subjects for the three levels of task difficulty_(combining age and sex) for Experiment I. Range of scores also indicated.............. 104 XXI I ----------------------------------------------------------------------------------- j Figure Page ! 8. Mean conformity scores (1st private I acceptance) of subjects for the j three levels of task difficulty I (combining age and sex) for I Experiment I. Range of scores j also indicated........................ 105 I 9 * Mean conformity scores (2d private I acceptance) of subjects for the ( three levels of task difficulty j (combining age and sex) for I Experiment I. Range of scores j also indicated......................... 106 ! 10. Mean conformity scores (compliance) I of young and old subjects i (combining the three levels of ! task difficulty) for Experiment II. I Range of scores also indicated......... 119 I 11. Mean conformity scores (1st private acceptance) of young and old subjects (combining the three levels of task I difficulty) for Experiment II. j Range of scores also indicated......... 120 I 12. Mean conformity scores (2d private ' acceptance) of young and old subjects I (combining the three levels of task difficulty) for Experiment II. Range of scores also indicated........ 121 13. Mean conformity scores (compliance) of male and female subjects (combining the three levels of task difficulty) for Experiment II. Range of scores also indicated........ 123 14. Mean conformity scores (1st private acceptance) of male and female subjects (combining the three levels of task difficulty) for Experiment II. Range of scores also indicated....................... 124 15. Mean conformity scores (2d private acceptance) of male and female subjects (combining the three levels of task difficulty; for Experiment II. Range of scores also indicated........ 125 _________ xxii Figure Page 16. Mean conformity scores (compliance of subjects for the three levels of task difficulty (combining age and sex) for Experiment II. Range of scores also indicated.............. 127 17- Mean conformity scores (1st private acceptance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment II. Range of scores also indicated....................... 128 18. Mean conformity scores (2d private acceptance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment II. Range of scores also indicated........ 129 19* Mean conformity scores (compliance) of young and old subjects (combining the three levels of task difficulty; for Experiment III. Range of scores also indicated....................... 14-4 20. Mean conformity scores (1st private acceptance) of young and old subjects (combining the three levels of task difficulty) for Experiment III. Range of scores also indicated......... 14-5 21. Mean conformity scores (2d private acceptance) of young and old subjects (combining the three levels of task difficulty) for Experiment III. Range of scores also indicated......... 14-6 22. Mean conformity scores (compliance) of male and female subjects (combining the three levels of task difficulty) for Experiment III. Range of scores also indicated....................... 14-8 23. Mean conformity scores (1st private acceptance) of male and female subjects (combining the three levels of task difficulty) for Experiment III. Range of scores also indicated......... 14-9 XXlll Figure Page 24. Mean conformity scores (2d private acceptance) of male and female subjects (combining the three levels of task difficulty) for Experiment III. Range of scores also indicated........ I50 Mean conformity scores (compliance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment III. Range of scores also indicated........ 152 26. Mean conformity scores (1st private acceptance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment III. Range of scores also indicated....................... 153 27. Mean conformity scores (2d private acceptance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment III. Range of scores also indicated........ 154 28. Mean conformity scores (compliance) of young and old subjects (combining the three levels of task difficulty) for Experiment IV. Range of scores also indicated....................... 168 29. Mean conformity scores (1st private acceptance) of young and old subjects (combining the three levels of task difficulty) for Experiment IV. Range of scores also indicated......... 169 50. Mean conformity scores (2d private acceptance) of young and old subjects (combining the three levels of task difficulty) for Experiment IV. Range of scores also indicated......... 170 51. Mean conformity scores (compliance) of male and female subjects (combining the three levels of task difficulty) for Experiment IV. Range of scores also indicated......... 173 XXXV Figure Page 32. Mean conformity scores (1st private acceptance) of male and female subjects (combining the three levels of task difficulty) for Experiment IV. Range of scores also indicated......... 174 33* Mean conformity scores (2d private acceptance) of male and female subjects (combining the three levels of task difficulty) for Experiment IV. ^ Range of scores also indicated......... 173 | 34. Mean conformity scores (compliance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment IV. Range of scores also indicated................ 176 33* Mean conformity scores (1st private acceptance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment IV. Range of scores also indicated....................... 177 36. Mean conformity scores (2d private acceptance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment IV. Range of scores also indicated....................... 178 37* Mean conformity scores (compliance) of young and old subjects (combining the three levels of task difficulty; for Experiment V. Range of scores also indicated....................... 193 38. Mean conformity scores (1st private acceptance) of young and old subjects (combining the three levels of task difficulty) for Experiment V. Range of scores also indicated........ 194 39* Mean conformity scores (2d private acceptance) of young and old subjects (combining the three levels of task difficulty) for Experiment V. Range of scores also indicated........ 193 XXV Figure Page 40. Mean conformity scores (compliance) of male and female subjects (combining the three levels of task difficulty) for Experiment V. Range of scores also indicated....................... 197 41. Meeui conformity scores (1st private acceptance) of male and female subjects (combining the three levels of task difficulty) for Experiment V. Range of scores also indicated......... 198 42. Mean conformity scores (2d private acceptance) of male and female subjects (combining the three levels of task difficulty) for Experiment V. Range of scores also indicated........ 199 43. Mean conformity scores (compliance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment V. Range of scores also indicated................ 201 44. Mean conformity scores (1st private acceptance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment V. Range of scores also indicated....................... 202 4 5. Mean conformity scores (2d private acceptance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment V. Range of scores also indicated....................... 203 46. Mean conformity scores (compliance) of young and old subjects (averaged over the five experimental tasks;. Range of scores also indicated........ 211 4 7. Mean conformity scores (1st private acceptance) of young and old subjects (averaged over the five experimental tasks). Range of scores also indicated........................... 212 XXVI I Figure Page t 48. Mean conformity scores (2d private acceptance) of young and old subjects (averaged over the five experimental tasks). Range of scores also indicated................ 213 49# Mean conformity scores (compliance) of young and old subjects for each of the five experiments.......... 214 : 1 50. Mean conformity scores (1st private acceptance) of young and old : subjects for each of the five : experiments.......................... 214 | 51. Mean conformity scores (2d private ! acceptance) of young and old i subjects for each of the five ! experiments.......................... 215 j 52. Mean conformity scores (compliance) ! of male and female subjects (averaged , over the five experimental tasks). j Range of scores also indicated........ 220 I 53# Mean conformity scores ( 1st private ' acceptance) of male and female subjects (averaged over the five experimental tasks). Range of scores also indicated................ 221 54. Mean conformity scores (2d private acceptance) of male and female subjects (averaged over the five experimental tasks). Range of scores also indicated................ 222 55# Mean conformity scores (compliance) of male and female subjects for each of the five experiments.......... 223 56. Mean conformity scores (1st private acceptance) of male and female subjects for each of the five experiments. ................ 224 57# Mean conformity scores (2d private acceptance) of male and female subjects for each of the five experiments.......................... 225 xxvii Figure Page 58. Mean conformity scores (compliance) of the three levels of task difficulty (averaged over the five experimental tasks). Range of scores also indicated.............................. 228 59# Mean conformity scores (1st private acceptance) of the three levels of task difficulty (averaged over the five experimental tasks). Range of scores also indicated.................. 229 60. Mean conformity scores (2d private acceptance) of the three levels of task difficulty (averaged over the five experimental tasks). Range of scores also indicated.................. 250 61. Mean conformity scores (compliance) of the three levels of task difficulty for each of the five experiments. ................. 254 62. Mean conformity scores (1st private acceptance) of the three levels of task difficulty for each of the five experiments...................... 256 65. Mean conformity scores (2d private acceptance) of the three levels of task difficulty for each of the five experiments...................... 257 64. Mean conformity scores (compliance) of the five experimental tasks (averaged over age, sex, and task difficulty). Range of scores also indicated....................... 241 65# Mean conformity scores (1st private acceptance) of the five experimental tasks (averaged over age, sex, and task difficulty). Range of scores also indicated.....................'.. 242 66. Mean conformity scores (2d private acceptance) of the five experimental tasks (averaged over age, sex, and task difficulty). Range of scores also indicated......................... 243 _____________________ xxviii________________________ Figure Page 67. Equipment Set-Up Used in Making Auditory Stimulus— Experiment III .... 299 XXIX DISSERTATION ABSÎRAOf AGE. SEX, AND TASK DIFFICULTY AS PREDICTORS OF SOCIAL CONFORMITY: A SEARCH FOR GENERAL TENDENCIES OF CONFORMITY BEHAVIOR Many studies within the past several decades have considered various aspects of social conformity. However, there has been a dearth of research dealing specifically with age differences in social conformity. The relation ship between age and conformity is an important social issue that deserves empirical investigation. One recent investigation indicated that older individuals were more conformii^ than younger individuals on a task of visual perceptual judgment. The present investigation was pri marily concerned with determining if there is a "general tendency" for older individuals to be more conforming than younger individuals regardless of the experimental task. As a secondary concern, this investigation attempted to determine if the degree of expressed conformity varies due to the nature of the experimental task. In order to answer these questions, five different conformity experiments were conducted on a variety of experimental tasks. Sixty young subjects (Ss) (17-^) were compared to sixty old Ss (60-81) regarding susceptibility to social XXX influence. There were twenty-four Ss in each of the five conformity experiments. The Ss in each of the experiments were subjected to contrived group pressures toward erro neous judgments or opinions. Experiments I through V were all concerned with age, sex and task difficulty as predic tors of social conformity. The following represent the five conformity experiments and the task performed in each: Experiment (I) Visual perceptual judgment task— S*s task was to judge which of two circular discs was largest; Experiment (II) Audi to ly perceptual judgment task— S’s task was to judge the number of metronome clicks heard; Experiment (III) Auditory signal detection task— S’s task was to report whether he heard a pure tone plus noise or noise alone in his right ear, while sometimes receiving information in his left ear; Experiment (IV) Problem solv ing task— S’s task was to solve one-operation arithmetic problems; Experiment (V) Social attitudes task— S’s task was to rate statements on nationalism as to whether he agreed or disagreed with them. The five experiments followed an identical research design, which provided measures of both compliance and pri vate acceptance as two distinct measures of social conform ity, The Ss were first tested alone, then in the conform ity situation, and then again alone. The Ss were retested alone one week later. xxxi Analysis of variance of the frequency of conformity revealed the following primary results: (a) Old Ss con formed significantly more often than young Ss— Experiment I and III (compliance), Experiment IV and V (compliance and private acceptance); (b) there was not a significant sex difference in conformity in any of the five experiments; (c) conformity occurred significantly more frequently as the difficulty of the experimental task increased— Experi ments I, II, III, IV and V (compliance and private accep tance ); (d) older Ss were found to conform significantly more often than younger Ss when the data from the five experiments were combined (compliance and private accep tance); (e) there was not a significant sex difference in conformity based on the combined data; (f) conformity was found to occur significantly more often as task difficulty increased when the data of the five experiments were com bined (compliance and private acceptance); (g) conformity occurred more frequently as the nature of the task pro ceeded from an objective frame of reference to a social frame of reference (i.e., conformity increased from Experiment I through Experiment V, compliance and private acceptance)• This investigation demonstrated that old individuals conformed more often than young individuals on a variety of experimental tasks. Therefore, the results of this xxxii investigation lend support to the position that there is a "general tendency" for older individuals to be more con forming than younger individuals regardless of the experi mental task. XXXlll CBAPTEH I GENERAL INTRODUCTION The present investigation focused on determining if conformity behavior is task specific. That is, the present investigation attempted to ascertain whether the greater degree of conformity expressed by older subjects in a visual perceptual judgment situation (Klein 1970) is unique to that situation, or whether there is a general tendency for the older person to be more conforming regardless of the experimental task. It was also the purpose of this study to investigate sex differences in conformity behavior on various experi mental tasks and to investigate variation in conformity behavior due to ambiguity of the stimulus on various experimental tasks. The major hypothesis of this investigation is that older subjects will be more conforming than younger sub jects on the various experimental tasks employed. This investigation was designed to answer the questions raised above by conducting a series of conformity experiments on a variety of tasks. Movement vs. Congruence Conformity Conformity is a social phenomenon that lends itself to many definitions. Contemporary treatments of social conformity, however, often neglect to consider the type of distinction illustrated in Beloff*s (1958) contradistinc tion of "conventionality" and "acquiescence" as modes of social response. Acquiescence is operationally defined as high "agreement" between an individual’s response and the mean or mo<|al response of his group or class. Conven tionality is operationally defined as the amount of shift" from private to public opinion. This distinction between the two basic descriptive criteria of conformity have been formally analyzed by Willis (196^) and labeled "congru ence" and "movement. " In descriptive te ms, the congruence criterion requires that conformity be measured on the basis of the extent of agreement between a given response and the normative ideal. The criterion of movement measures con formity on the basis of a change in response resulting in a greater or lesser degree of congruence (Hollander & Willis 1967). Congruence conformity and movement conformity both entail an acceptance of influence which demonstrates "dependence" on opinions and attitudes of other persons. However, in the case of congruence, this dependence has origins in a past influence (which have some continuity in 3 the present such as art preference and aversions).. In the case of movement, the individual responds to a present influence by changing his behavior to coincide with new demands ; such as complying with the political opinions of others. If these two aspects of conformity are mixed, or if there is a failure to recognize one or the other, confusion will be encountered when attempting to understand the operations of conformity. Therefore, for purposes of understanding, it is essential to maintain a rigorous differentiation between these aspects of conformity. Experimentation in the laboratory has directed a great deal of scrutiny toward "movement conformity." This tendency to be swayed by outside influence is the concern of the present investigation. This tendency is well illus trated in the pioneering studies of Sherif (1935) and Asch (1951). Numerous studies have subsequently varied the Sherif and Asch procedure, utilizing a range of stimuli, demonstrating repeatedly that knowledge of the reactions and responses of others can influence behavior in the direction of conformity. Susceptibility exerted in this manner is not peculiar to any specific type of task, but is found for a wide variety of tasks including violation of prohibitions and acceptance of requests (Barch, Trumbo & Nangle 1957; Blake, Berkowitz, Bellamy, & Mouton 1956; 4 Blake, Helson & Mouton 1956; Freed et al. 1955; Freeman & Fraser I966; Grosser, Polansky & Lippitt 1951; Rosenbaum 1956; Schaoter & Hall 1952; Skanksmitb 196?), the expres sion of attitudes and opinions (Converse & Campbell I968; Bunker 1938; Gordon 1952; He Ison, Blake, Mouton & Olmstead 1956; Horwitz, Pi ana, Goldman & Lee 1955; Kiesler, 2 anna & BeSalvo I966), and reactions to perceptual judgmental and factual materials (Asch 1951; Blake, & Brehm 195^; Clark 1916; Endler I96O; Klein 1970; McConnell & Blake 1953; Misterberg 1950; Olmstead & Blake 1955; Schonbar 19^5; Sherif 1935). Normative vs. Informational In line with the distinction between congruence and movement conformity, the distinction between two types of social influence also must be made. Research on conform ity behavior implies that the influence of social pressure falls into two general categories. One of the most pro lific differentiations between these two motivational bases underlying conformity behavior is that proposed by Beutsch mid Gerard (1955) in distinguishing between "norma tive" and "informational" forms of social influence. Analogous discussions of similar differentiations are Asch’8 (1956) distinction between yielding at the "action level" and at the "judgment level," Thibaut and 5 Strickland’s (1956) consideration of "group" and "task" set and in McBavid’s "message-oriented" and "source- oriented" distinction* Indirectly, but along the same lines, Hiopelle (i960) and Me David (I962, 196*+) have com pared "observational learning" to "blind imitation. " Under the conditions of "normative social influ ence," in an attempt to live up to the expectation of others, an individual’s response to group pressure is assumed to ensure continuation in the group, contribute to its code and norms, and enhance its cohesive qualities (DiVesta 1959). The second pressure influence^ "informational social influence," is the concern of the present investigation. Under conditions of this variable, group membership is relatively unimportant and group influences are only inconspicuously or indirectly created# Therefore, conform ity results, due to the individual’s need to gain correct information about reality and to validate his opinions and make sure they are consistent with the opinions of others. He, therefore, resorts to the judgments of the other group members (to use Thibaut and Strickland’s phrase) "as media tors of fact." (DiVesta 1959,) The following is a typical definition of social conformity: Conformity is defined as some behavioral or attitudinal change that occurs as the result of some real or imagined social influence (see Berg & Bass; Brown 1965; Homans 1961; Kreeh, Crutchfield & Ballachey 1962; Secord & Backman 196^; Walker & Heyns 1962 for similar "textbook" definitions). The operational criterion of conformity underlying most empirical research in this area, including the present study, is . . the public and/or private agreement of an individual with the opinion or judgment of others which he had not held before it was presented to him." (Sampson 1964.) The existence of this tendency is a basis of all human society and accounts for a major part of human conduct. Our aim as psychologists is, in part, the understanding and explanation of behavior. It seems only reasonable, therefore, that research should attempt to elucidate that aspect of behavior which is referred to as conformity. As Hollander (1964) says in reference to the necessity of research in the area of conformity, ". . . among those phenomena lying within the scope of social research, none occupies a more central place than conform ity." Desirability vs. Undesirability of Conformity Judgments concerning the personal or social advan tages or disadvantages of conformity cannot be made easily, since it is apparent that it must have complex origins and ramifications. Independence of expression fulfills a 7 significant social function which is illustrated by genera tions of new thoughts and discoveries. Gardner (1963) points out that repressing independence and individuality is apt to result in social decay because, . • the capacity of society for continuous renewal depends ulti mately upon the individual." (Hollander 1967.) Neverthe less, Heisman (1950) has brou#it it to our attention the view that we are becoming excessively "other-directed." Mbyte (1956) informs us that our Protestant Ethic is being replaced by a Social Ethic; as he phrases it, which dic tates that "being a good team player" is the highest good. Packard (1959) documents the attempts of each social class of our society to imitate the behavior, and even the thoughts, of those individuals of a higher social stratum. This account of social critics illustrating the same general point could easily be lengthened, but perhaps sufficient examples have been mentioned to document the present concern over the professed decline of individual ism* On 1he other hand, if conformity is viewed as the adherence to social expectancies, it becomes evident that for society to function in an organized fashion, conform ity has to exist. Nevertheless, a frequent criticism of our society today is that we live in an "age of conformity.' Mben used in this manner, "conformity" is seen as something 8 which undercuts individuality. However, it is unavailing to condemn any behavior automatically because it demon strates conformity. Society would be in a chaotic state if we lacked accepted rules of behavior. It is often con formity that makes it possible for us to adapt to our environment. (Klein 1970.) Perhaps we are not living in an "age of conformity," Conceivably, we do not cohere to norm and role expectation more closely than individuals of previous times, but we definitely have become more aware and concerned about such matters in recent years. As this concern has developed, a concomitant effort by behavioral and social scientists has grown to understand the principles underlying social con formity. Although men of all ages have directed attention to such issues, it is only recently that any measure of systematic investigation has been accomplished. Perhaps it would be of value to trace some landmarks of significant relevance to the area of social conformity. In this way it will become apparent how some of the procedures of the present investigation were derived, and the reasons for conducting the present investigation will also become more clear. Historical Landmarks in the Area of Social Conformity In 1890 farde’s book. Laws of Imitation, was 9 published. He professed that imitation was the primary factor accounting for group behavior. Tarde believed that the principle of imitation, by itself, was adequate in explaining group behavior. He went so far as to declare, "Society is imitation." LeBon’s book, The Crowd, appeared in 1895. He utilized the principle of suggestion and, therefore, undoubtedly was influenced by Charcot’s work in the area of abnormal psychology. leBon focused his concern on individual changes in behavior when in a crowd situation. He was interested in defining the nature of the "crowd man." Be Bon felt that members of a crowd resembled hypno tized individuals and that crowd behavior necessarily occurred on the level of the least intelligent members. According to LeBon, the gathering together of individuals is all that is required to compose a crowd. At this point the purported "Law of Mental Unity" automatically goes into effect, resulting in the individual’s loss of his individ uality. He becomes a "mechanical man" which immediately transforms suggested ideas into action. (Asch 1952.) There were obvious drawbacks in both Tarde’s and LeBon’s explanation of group behavior. Neither employed experimental techniques or carefully gathered empirical data. They both tried to account for social phenomena with oversimplified explanations and neglected to delineate 1 0 the underlying processes. In 1908 Ross* book, Social Psycholomr. was published. He also tried to account for group behavior in terms of an all-encompassing principle. His principle appeared to be a combination of Tarde *s and LeBon*s principles. Ross, how ever, did not distinguish between imitation or suggestion. He also followed the Tarde-LeBon tradition and failed to utilize experimental approaches or empirical data. His evidence was primarily anecdotal. He also supported LeBon’s "Doctrine of Group Mind" which maintained that the laws of individual psychology are not adequate to explain group behavior. In 1916 Floyd Allport was initiating his research. His concern was to study the effect of the presence of others on individual performance. Allport, however, broke away from the Tarde-LeBon-Ross tradition. He employed an entirely different conceptual design, namely, a type of behaviorism. He also conducted a variety of experimental investigations in group psychology. Allport maintained, in opposition to the "Doctrine of Group Mind," that group concepts are unnecessary for explaining behavior in groups. Allport wrote, "There is no psychology of groups which is not essentially and entirely a psychology of individuals." In Allport’s experiments, his subjects were required to do a number of tasks, including: crossing out vowels in news 11 articles, free associating, multiplying numbers and judging the weights of objects. The tasks were performed under three conditions; while entirely alone, while alone but aware that others were working elsewhere simultaneously, and while together with others at the same place. It was found that the subject’s orientation was different when in the presence of others. In the group situation, there was a greater concern for what others thought. For example, weight judgments were more moderate. Kiesler (I969) has said, in reference to Allport’s work, "The heuristic value of Allport’s work lies in demonstrating that the mere presence of others can be psychologically important." The work of Allport led the way for future experi mental investigation of the area of social conformity. For example, Jenness (1932) conducted experiments on the effect of group discussion on judgments. The subjects in these experiments were asked to judge the number of beans in a bottle. The subjects then discussed the judgments to arrive at a single judgment, after which each made a second set of individual judgments. The results continually indi cated convergence, that is, movement toward a group standard. This research was one of the first direct attempts to investigate convergence of judgment about a group norm, although Allport and others had previously addressed themselves to similar issues. 1 2 In 1935 Muzafer Sherif contributed significantly to the area of social conformity when he published the first of his autokinetic experiments. Sherif (1935) approached the issue of conformity in terms of "social norms." In the autokinetic situation, a point of light is projected in a completely dark room. This point of light appears to move. Hence, the name autokinetic, meaning self-moving. It has been demonstrated that the degree of movement reported by subjects varies among individuals. Initially, each of Sherif’s subjects were tested in a darkened room alone. Each subject, for several trials, was required to look at the point of light and report the direction and degree of movement (in inches). Following this procedure, each sub ject was returned to the room, however, this time in the company of other subjects. They reported their judgments aloud as they watched the point of light and in a short time they converged toward a group standard of apparent movement. Sherif tested his subjects again alone after they had made the judgments in the presence of others. It was found that the subjects retained the group norm rather than their own standard as a basis for judging the extent of apparent movement. Sherif’s work demonstrated the pro found influence other people can have on the perceptions of an individual. Since Sherif*s pioneering work, con formity to group norms and intra-group influence have 13 become focal issues in the area of social psychology. Hollander (1967) has said, in reference to the Sherif study, "This experiment did much to encourage further experimentation on the psychological aspects of group phenomenon. It also showed the harmonious way in which group concepts, like social norms, could be subjected to psychological analysis." (Klein 1970.) Leon Festinger, trained in the tradition of Lewinian group dynamics, has proved to be one of the most prominent researchers in the field of social psychology. Festinger focused on analyzing pressures toward conformity. Fes tinge r (1950) proposed two major sources of pressure toward conformity. He labeled the first source "social reality." By this he meant that a belief is perceived as being "valid" or "proper" to the degree to which it is attached to a group of individuals with similar beliefs. Consequently, members of a group are motivated to "comply" with those beliefs of the group. The second source of pressure toward uniformity occurs because conformity con veniences the attainment of "group goals." Festinger (1953) has also dealt specifically with compliant behavior. He has also forwarded a theoiy of "social comparison" which describes the individual as relying on the group for evalua tion of abilities as well as beliefs. The final historical landmark that shall be l4 discussed is the work of Solomon Asch (1951)* This was an extension of Sherif’s earlier work. In Aseh’s experiments, a group of individuals (typically eight), are seated in a room next to one another. Before them they are shown a standard line and a set of two or three unequal comparison lines. The subjects are required to select from the com parison lines that one which is equal in length to the standard line. The comparison lines are labeled and the subjects are instructed to make their judgments aloud. Of the eight individuals tested simultaneously, only one is an actual subject the rest are confederates of the experi menter. The confederates had met before the actual experi ment and had received instructions that on certain trials they were to respond unanimously with wrong judgments. This procedure confronted the subject with a unanimous contradiction of his own perception on these trials. The results indicated that the subjects made a distinct move ment toward the group’s distortion of accuracy. Overall, approximately 33 percent of the judgments of the subjects were in the pro-majority, or incorrect direction. Hollander (1967), commenting on Asch’s work, states that "Laboratoiy experimentation of this kind supports the hypothesis that our reports of what we perceive are sus ceptible to the influence of others. . . (Klein 1970.) The studies mentioned above have been considered the 15 landmarks in the area of social conformity. There have been vast nmabers of studies not mentioned here that have also contributed greatly to the area. Nevertheless, this review generally brings one to the present state in the area of social conformii^y research. The Present Investigation There are over twenty million people above the age of sixty-five in America. It is interesting, and of con cern, that although they represent a potentially strong political force, they have appeared to fail to demand those things which they affirm they desire and need and those economic and social rights which they deserve. Inferring from the hypotheses of several investigators, this phenomena may be due to a disengagement, a decline or witiidrawal from participation in communii^y and social activities by the older individual (Gumming & Henry 196I; I'oskett 1955; Henry & Gumming 1959; Mayo 1950; Tallent & Lucas 1956; Wagner I96O; Zborowski 1962). It has also been hypothesized that the older individual *s failure to strive for rights and fulfillment of needs is due to a change in social role, a lack of the necessary energy level, a lack of social status or a lack of opportunity (Spangler & Bxomas I962). However, the question arises as to whether these qualities attributed to older individuals can completely account for their failure 16 to strive for rights and fulfillment of needs* Perhaps it is the tendency of older individuals to be conforming that causes their acceptance of our current middle-age, and, even more so, youth-oriented society. (Klein 1970.) It was the type of theorizing stated above, that led the writer to conduct an experiment (Klein 1970) in order to obtain a partial answer to the question of age differ ences in social conformity. This experiment was directed at determining if age, sex and task difficulty acted as determinants of social conformity in a laboratory visual perceptual judgment situation. The subjects were required to judge which of two stimuli (circular discs) was greater in size. Employing a modified Crutchfield (1955) apparatus, young and old subjects were subjected to contrived group pressures toward erroneous perceptual judgments. An analysis of the data indicated that older subjects con formed significantly more often than younger subjects. The following question arose, leading to the present investigation: Was the grater degree of conformity shown by the older subjects, in the above mentioned experiment, a tendency peculiar to that experimental task or is there a general tendency for the older person to be more conforming regardless of the task? The issue of “task specific'* behavior, in reference to the older individual, has been considered in a recent 17 investigation by Birr en (1970). This experiment considered the hypothesis of cautiousness in elderly subjects. The subjects were required to sort decks of cards according to the side of the card on which the smaller of two circles appeared. The cautiousness hypothesis leads to the expec tation of a marked slowing in judgment time by the older individual when sorting decks of maximum ambiguity (that is, when there is no difference between the size of the circles). It was found that even under tasks of maximum ambiguity the older subjects did not increase their judg ment times significantly. Birren, in interpreting the results, alluded to the simplicity of the task and pointed out that, perhaps, on more complex tasks this result would not have been observed. That is, perhaps cautious behavior is “task specific.“ This study demonstrates the need to be concerned with the issue of experimental findings being “task specificFor exan^le, it would be unwise, based on the Klein (1970) study mentioned above, to make the general statement that “older individuals are more conforming than younger individualsTherefore, the present investigation was conducted in order to determine if support could be given to the validity of such a statement. Kiesler (I969) has stated, “We need more studies testing the ^limits' of hypotheses— that is, testing logical extensions of hypothe ses and possible limitations of hypotheses.” Even more 18 specifically, Graham (1962) has stated, “It would be very interesting if one could conduct experiments on conforming in a number of different situations . • •“ In a social conformity situation, which is the setting for this experiment, the subject is confronted with an apparent disagreement between his own judgments or opinions and a unanimous contradictory consensus of the other subjects. What does the subject do under these cir cumstances? Hoes he depend on the evidence of his own senses and opinions and respond independently, or does he yield to the judgment or opinions of the group? The present investigation was concerned with deter mining if age, sex, and task difficulty act as determinants of conforming behavior in a series of laboratory experi ments. These conformity experiments covered a variety of different tasks including: a modified replication of Klein's (1970) study; an auditory perceptual judgment study; an auditory signal detection study; a study on problem solving; and a study on social attitudes. The primary focus of this investigation was to determine if older individuals are more conforming than younger individuals on each of these tasks. A secondary concern of the present investigation was to ascertain differences in conformity due to the nature of the experi mental task. 19 Presented immediately "below is a review of some of the literature which reflects on the issue of generality of conformity behavior* Studies both supporting and refuting the idea of conformity as a general tendency is presented. Following this section, a review of the literature is pre sented discussing each of the independent variables of the investigation as they relate to conformity. A section is also presented on the nature of the experimental task. This literature review illustrates why the particular hypotheses tested in this investigation were formulated. The Generality of Conformity Behavior' The contention in the first portion of this section is that the data indicate that susceptibility to social influence is a general tendency. The second portion of this section presents the opposite view and illustrates findings that do not support a generality of conformity across a variety of tasks. There have been a number of investigators who have demonstrated that conformity in one situation tends to be correlated with conformity in other situations. Harper and Tuddenham (1964) have shown that conformity to acquaint ances is similar to conformity to strangers. Sears (1963) has reported a generality of various forms of dependency in children. Âsch (1956) found that college students tend to obtain the same conformity scores early and late in the 2 0 task. In an experiment by Blake, He Ison and Mouton (1956), three different tasks were employed: reporting the number of metonome clicks heard, expressing attitudes to state ments and solving arithmetic problems. The results indi cated a high degree of individual consistency in conforming within tasks and a considerable amount of consistency between tasks. Nakamura (1958) found the reliability of “objective conformity” (perceptual) items high (.87 for men, .80 for women). Vaughn (1964) found some consistency in conformity or nonconformity for 20 percent of his sub jects studied across four situations. Frye and Bass (1958) and Beloff (1958 ) found a close relationship between face- to-face “acquiescence” and general “conformity” to social norms. Tuddenham (1957) obtained highly consistent individual differences across 7% trials concerning judging visual targets, information, and opinion. MaoBride (1958), using the same measuring procedures as Tuddenham, found a high degree of individual consistency in conforming even after he had raised or lowered the self-confidence of his subjects. Abe Ison and Lesser (1959) found that children who conformed to teachers’ assignments were also likely to conform (as measured projectively) to the judgments of their mothers. Oorrelation between the two scores in some classes was as high as . 90. Ferguson (1944) has reported that there is 21 generality of conformity within the area of attitude judg ments. Responses were obtained for three different atti tude scales (Religionism, Humanitarianism, and Nationalism). It was found that individual shifts from initial judgments toward an assumed majority position showed consistency for all three measures. Ferguson stated, in discussing his results, “It is thus apparent that suggestibility to group opinion is not specific to the attitude variables in ques tion, and may to some extent, therefore, be considered as a general personality trait.” However, this study dealt with attitudes only; it did not offer a basis for assessing the generality of conformity with other types of tasks. Crutchfield (1955), using a battely of 21 items including tasks on perceptual judgments as well as attitudes, reported a significant degree of generality of conformity as determined by split-half correlations. Rosner (1957) raised the question as to whether the individual consistencies in yielding or not yielding were specific to the judgment of lines or whether they extended beyond this type of situation. In this investigation the tasks included three memoiy tasks (nonsense words, nonsense figures, and a simple paragraph), a questionnaire and the line judgments used by âsch (1951)* It was found that response to group pressure was significantly consistent between tasks at a single sitting. The subjects who were 2 2 high eonfomers on one task tended to he high oonformers on the other tasks. Likewise, low conformera on one task were found to he low conformera on the other tasks. Phi coeffi cients ranging from .33 to .54 were obtained between pairs of tasks with respect to high and low yielders. Back and Davis (1965) investigated consistency of individual behavior to group pressure in three different situations— a perceptual judgment task, self-reported acceptance of peer group norms, and self-reported accep tance of authority pressures. The data confirmed the authors’ expectations that conforming to group pressure would show a small but reliable consistency from situation to situation. Milgram (196I) has demonstrated consistent differ ences between two countries in amount of movement conform ity* Norwegian university students were found to be more conforming than French university students across five situations related to the Crutchfield (1955) procedure. Regarding these results, Milgram stated, “No matter how the data are examined they point to greater independence among the French than among the Norwegians. “ He related this difference in independence to variations between the two national cultures. There is an abundance of data indicating that per- suasibility on one issue is positively related to persuasi- bility by other messages on other issues. Since 23 persuasibility studies also demonstrate susceptibility to social influence, an example seems appropriate. It will be sufficient to present one study in order to exemplify the extent of the relationship. Janis and Field (1959) presented a set of ten per suasive messages to about 185 high school students. Each message was supposedly from a different newspaper reporter, and they included a variety of topics (each with a pro and eon form). The subject’s persuasibility for each message was scored ’0’ or ’1’ on the basis of a median split with respect to amount of change. Tetraohoric correlations were computed between the impacts of each pair of messages resulting in 4-5 correlations. Of the 45 correlations, only six proved to be negative. Of the 39 positive correlations, 25 were significant at the .05 level. Therefore, this study, demonstrating a predominance of significant positive correlations, suggests an underlying general factor of susceptibility to social influence. It should be realized, however, that correlation does not indicate identity. Rarely do the correlations in the studies mentioned above account for even as much as 50 percent covariance between tests. The lowness of the typical correlations partially reflect the unreliability of the tests. It also suggests that scores on conformity tests reflect reliable variance other than on a common ^ g- conformity factor. Ferguson’s (1944) study mentioned above does reveal an approximation to a unitary factor solution, but veritably on the average his correlations were quite low. There have been a number of Investigators who have argued for the specif icily of conformity. For example, in discussing personaliiy and conformity, Hollander and Willis (1967) state, ”. . . it is increasingly clear that the search for sovereign attributes of a conforming personality have not been especially fruitful.” They go on to say, “True, for any particular situation individual differences are invariably observed, and these are often substantial, but it is also true that conformity in one situation is not generally a veiy reliable predictor of conformity in other situations.” Linton (1955) employed three different tasks to study the specificity of conformity behavior. The tasks included an autokinetic situation with one confederate, a syllogism test concerned with the effects of the content of the syllogisms in relation to the subject’s attitudes on relevant issues, and an attitude change test concerned with matters of “low personal significance.” The data indicated that the tests were not significantly intercorre lated. Goldberg (1954) utilized a task where his subjects had to judge intelligence from pictures before and after 25 contrived group norms were supplied. The data indicated that susceptibility to influence was highly specific to each picture. Goldberg said in summary, “Ihtercorrelations between conformity under all experimental conditions were low. Conformity was not generalized but appeared to be highly situational. This was considered as evidence against the existence of any general personality character istic of suggestibility or conformity.” The findings presented above do not clearly indicate the generality or specif icily of conformity behavior. It must be remembered, however, that none of these studies nor any others in the area of social conformity, considered the generality of conformity for different age groups • The following section discusses conformity in reference to age. Evidence is presented to indicate why it is a hypothesis of this investigation that older individuals will be more con forming than younger individuals on various experimental tasks. Age and Conformity Since the pioneering work of Sherif in 1935> numer ous studies have been conducted in the area of social con formity. Although these studies are different in many respects, they generally all have one thing in common— they fail to investigate possible age differences or age changes in conformity behavior. 2^ One of the two studies that has concerned itself with age and conformity is a recent investigation by Kahana and Coe (1969)* Individual and situational determinants of conforming behavior were investigated in a home for the aged. The residents were rated by staff members on the degree to which they conformed to the rules and expecta tions of the home. Data obtained through interviews indi cated that both situational and individual factors had a significant role in affecting conformity behavior. Inte gration in the informal organization of the home and adjustment were individual factors that were most signifi cantly related to conformity^ Salieney of the rule for both the home and the individual, among the situational factors, appeared to be important dimensions of conformity. Although the study of Kahana and Coe did not take an experimental approach in investigating age and conformity, it did indicate that older individuals are subject to social norms and standards. However, the question about the disparity of conformity between older and younger individuals, which is the primaly question, still remained unanswered. (Klein 1970.) In reference to age, it has consistently been found that younger individuals are more reactive and susceptible to social influence than are older individuals (AbeIson & Lesser 1959; Berenda 1950; Binker 1938; Luehins & Luchins 27 1955). However, these studies investigated age differences in conformity "behavior "between young and old children; or they compared young children with college age individuals. The question as to whether this finding would be consistent throughout the adult age span led to Hein’s (1970) inves tigation . The study by Hein (1970) is the most recent inves tigation of age differences in social conformity. As dis cussed earlier this investigation was concerned with age, sex and task difficulty as predictors of social conformity Subjects ranging in age from 16 to 21 were compared with an older group (60 - 86 years) on susceptibility to informa tional social influence upon visual perceptual judgments. A series of visual perceptual judgments were presented on slides to the subjects. The task was to judge which of two paired stimuli (circular discs) presented on each slide was greater in size. Employing a modified Crutchfield apparatus, the subjects were subjected to con trived group pressures directed toward erroneous perceptual judgments. The data indicated the following: (1) older subjects conformed significantly more often than younger subjects; (2) there was not a significant difference between males and females in conformity behavior; (3) con formity occurred more frequently as the stimulus ambiguity of the perceptual task increased; and (4) conformity scores 28 increased more in older subjects as stimulus ambiguity increased. These results clearly indicated that older individuals were more conforming. However, the question arises, which is the concern of the present investigation, as to whether this finding is “task specific.” The major hypothesis of the present investigation (which is stated more formally later) is that older individuals will be more conforming than younger persons on a variety of experimental tasks in similar conformity situations. The following discussion of the literature on aging and conformity will attempt to illustrate the impor tance of this hypothesis. Each study or group of studies on the same topic is not strong enough by itself to warrant this hypothesis, since many of the techniques employed in these studies leave much to be desired. However, taken together, these studies seem to warrant hypothesizing that there is a “general tendency” for older individuals to be more conforming than younger individuals. Age, manifest needs and conformity.— In this section a number of studies will be presented which demonstrate a relationship between scoring high or low on various needs and degree of conform! 1y. Age differences on these needs will also be presented. When older subjects score signifi cantly higher or lower on a need that has been shown to be related to conformity in the same direction, support is 29 given to the view that the older subject will be more con forming than the younger subject. Personality inventories, in which all subjects are asked identical questions, and the responses are objec tively scored, have their value in providing an objective measure of a response. The response can be related to other variables apart from its meaning. If people who score high on “aggression” are better lawyers, are less likely to commit suicide, or are alcoholics, it does not matter if we call it “aggression” or just “X“— the fact is that it becomes a predictor of behavior (Koponen 1957). Results consistently show that those individuals who are more susceptible to social influence are more likely to be low on “need achievement” and high on “need affiliation.” Thibaut and Strickland (1956) showed that in an Asch situation, conformity was increased by making subjects more “interaction-oriented” by imposing a “group set” on them rather than a “task set.” A number of other investigations have also found conformity to be greater among interact ion-oriented (need affiliation) subjects as compared with task-oriented (need achievement) subjects. In all of these investigations, accepting the norm, the group decision, or the simulated influence of the group necessitates that the subject reject the truth of what he actually perceives or believes. In these analyses, 30 conformity maintains social approval at the expense of task accomplishment. It also prevents the risk of being ostracized or rejected by the group for failing to be in agreement with them. Likewise, a conformer avoids the risk of showing himself to be in error. Therefore, it is not surprising that individuals most desirous of obtaining or maintaining social approval conform in these situations. Hardy’s (1957) investigation showed that those sub jects with low affiliation needs were also the subjects who conformed least. Raven (1959) found that individuals who do not feel rejected by others and are less concerned about what others think of them conform less than those that do feel such concern. Moeller and Applezweig (1957) have shown, utilizing the Behavior Interpretation Inven- toiy, that those subjects high in the need for social approval (need affiliation) and low in the need for self approval (need achievement) were most likely to conform in an Asch line judging situation— compared to those low in need for social approval and high in need for self approval. Support of Moeller and Applezweig’s finding is offered by Schroder and Hunt (1958) and by MeDavid (1959)# An investigation by Tuddenham (1958) reported correlations between the yielding-independence continuum and various rating variables and self-report tests. The tests included the Edwards Personal Preference Schedule, 31 the Gough California Personality Inventory, the Barron Ego Strength and Independence of Judgment Scales, the Taylor Manifest Anxiety Scale, and the Welsh “A” and ”R” factor!- ally based scales of the MMPI. It was found that those subjects who were high on need affiliation and low on need achievement were those who tended to yield most often. Krebs (1958) used a measure of Murray’s need- achievement and related it to conformity. He found, using judgments from memory as to which of two slides had con tained certain objects, that subjects with high-need achievement were more independent than those scoring low on need achievement. Samelson (1957) and Walker and Heyns (1962) have also found that subjects scoring low on need achievement are more likely to be susceptible to social influence. An investigation by Spangler and Thomas (I962) explored the effects of age, sex and physical disability upon manifest needs as defined by the Edwards Personal Preference Schedule. The results indicated that the need for affiliation score tended to increase with age. Koponen (1957) administered the Edwards Personal Preference Schedule to a nationwide sample of adults. The sample covered urban and rural areas of 1181 counties in 48 states. The EPPS was completed by 4,031 males and 4,932 females ranging in age from under 30 to over 55. Koponen 32 found that older subjects scored higher on need affiliation and lower on need achievement. Kuhlen and Johnson (1952) and McClelland (1953) have demonstrated the greater importance of achievement needs in early adult years. A study by Veroff, Atkinson, Feld and Gurin (I96O), utilizing projective pictures, showed high points in need achievement in young adulthood and middle age followed by a decrease in later years. The relationships demonstrated above— individuals scoring high on need affiliation and low on need achieve ment are more conforming; older as compared to younger individuals score high on need affiliation and low on need achievement— support the hypothesis that older individuals would conform more readily in a group situation. A number of studies have investigated what the rela tionship dominance, deference, submissiveness, and ascen dance has with conformity behavior. For example, Berg and Bass (1961) report ascendant and dominant persons to be more resistant to conformity pressures, whereas submissive persons were reported to be more susceptible to social influence • Results consistently shov/, in a variety of conform ity situations, that those who are more susceptible to con formity pressures are more likely to be submissive (Beloff 1958; Berenda 1950; Bray 1950; Kelson, Blake & Mouton 1 33 Helson Blake, Mouton & Glmstead 1956; Jenness 1932; Kalman 1950). Likewise, it has been consistently reported that those who yield to group pressure or social influence are less dominant (Barron 1953» Bray 1950; Hoffman 1953» Kelman 1950; Tuddenham 1958). There have been a few investigations that have related ascendancy, dominance, deference, and submi ssive ne ss to age. Bendig (i960) examined Guilford-Zimmerman Temperament Schedule scores of men up to age 60. He found that scores on ascendancy decreased with age. The studies by Spangler and Thomas (I962) and Koponen (1957) j mentioned above, found that older individuals score lower on the dominance and higher on the deference scale of the Edwards Personal Preference Schedule than younger individuals. Schaie and Strother (I968) reported results of veiy intelligent older subjects age 20 to 80. In comparison with a matched sample of graduate students in their mid twenties the older subjects were found to score higher on deference. These results compare favorably with the age differences found by Gavron (1965) who compared middle-aged people with young individuals in groups of psychiatric patients. It was found that middle-aged patients scored higher on deference. In a study by Meugarten and Gutman (1958), employing the Thematic Apperception Test, older men 34 were characterized as more submissive than younger men. The relationships shown above— individuals scoring high on dominance and ascendancy are less conforming, and those scoring high on submissiveness and deference are more conforming; older as compared to younger individuals score low on dominance and ascandaney and high on submi ssivene ss and deference— support the hypothesis that older individ uals will be more conforming than younger individuals. Age, perception of one’s parents and conformity.— Another line of reasoning which led to the hypothesis that older subjects would be more conforming than younger sub jects on various tasks, is the results of the questionnaire of Klein’s (1970) study. It was found that the older sub jects as compared, to the younger subjects perceived their parents as more dominating, restricting and rejecting. Berg and Bass (I96I) alude to this issue when they say, “Persons who conform more in a pressure situation can be characterized as perceiving their parents as harsh, puni tive , restricting and rejecting. . . Evidence from anthropological literature also sup ports the above-mentioned relationship. For example, societies in which conforming behavior predominates, propa gate the tendency from generation to generation by follow ing certain child-rearing processes. Among the Bathonga, an extremely conformist culture, it is inconceivable for 35 children to question their elders, their chiefs or their gods (Mead 1939). Champney (1941) investigated the significance of parental attitudes and behavior on the subsequent conform ity found in their children. Ghampney suggested that parents who dominate their children, restrain their chil dren's attempts to solve their own problems, and demand obedience above everything else, are likely to create docile, shy, self-conscious, conforming children. Mowrer (1939) has maintained that children from autocratic homes or schools are likely to become either domineering or autocratic themselves or highly conforming and susceptible to the influence of others. A study by Mussen and Kagan (1938) provides a more specific test of the relationships being considered. They found that in an Asch situation, subjects who were extreme conformists were more likely to perceive their parents on the Thematic Apperception Test as harsh, punitive, restrictive and rejecting. The relationships discussed above— individuals per ceiving their parents as rejecting, restricting, and domi nating are more conforming— older individuals, as compared to younger individuals, perceive their parents as more dominating, restrictive, and rejecting— support the hypoth esis that older subjects will be more conforming than younger subjects on a variety of tasks. 36 Age. eautiouaaQ.ess and conformity.— There seems to he a growing body of literature from various perspectives and with various types of data which demonstrates a relation ship between increasing age past adulthood and cautiousness, A review by Botwinick (1964) Indicated that, although many of the relevant studies have not been conceptualized in this way, there appears to be enough communality among them that this relation is a repeated finding in the study of aging. (Botwiniek I966. ) Some of the studies which lend support to the view of an increase in caution with age include Welford’s (1951) study which related accuracy and speed measurements in a psyehomotor task; Eisdorfer, Axelrod and Wilkie (1963) and Korchin and Basowitz (1957) who studied verbal learning; Basowitz and Korchin (1957) who investigated the ability of perceptual closure; Wallaeh and Kogan (I96I) who were interested in the decision process; Botwiniek, Brinley and Robin (1958) and Silverman (1963) who measured perceptual discrimination; Craik (1962, 1964) who measured performance of aiming ability; and Craik (1965), Milner, Walker, and Beech (1967), and Rees and Botwiniek (1971) who studied applications of signal detection theory in a detection situation. It should be made clear that an increase in cautiousness with age is one of the interpretations these investigators offer in 03j)lanation of their results. 37 In Korchin and Basowitz*s (1957) study, a young and old group of subjects (mean ages 26.8 and 78.1 years), were administered three paired-associate learning tasks. They were familiar word pairs, nonsense equations and false equations. The older and younger subjects were found to differ in their behavior during learning. The younger subjects were more apt to make some response to the stimu lus items. They posed i^po the ses whether correct or incorrect. The older subjects, however, responded cor rectly or not at all. The proportion of “no response” errors remained high through learning. One of the explana tions that Korchin jnd Basowitz posit for this behavior is that the older individual is more cautious. That is, despite encouragement to guess, it is possible that the older individual needs to attain a greater degree of cer tainty l>e fore he is willing to report. Although the sub ject knows that the item is equally incorrect, whether he gives no response or a wrong response, he apparently pre fers the error of omission to that of comBiission. This behavior of inhibiting a response in an uncertain situa- tioh, tends to support the view that older individuals would be more conforming, as compared to younger individ uals, especially under tasks of high difficulty and those that are ambiguous. It seems reasonable to conclude that, in a conformity situation, increasing the stimulus 38 ambiguity would lead to an increase in the subjective uncertainty of the individual’s judgment and his dependence on information from others. Consequently, this varialbe would be positively related to conformity* In a study by Wiener (1958), subjects were required to rate their feel ings of certainty about their judgments in an Asch situa tion. It was found that there was a significant positive relationship between uncertainty and conformity. In a later study, Wiener (1959) confirmed the importance of uncertainty as a variable in conformity research. In a study by Eisdorfer, Axelrod and Wilkie ( 1963), stimulus exposure and learning, in both young and old sub jects, was investigated. lists of eight words equated for association value, size, and structure were employed in a serial rote learning task with a one-second interword interval and a forty-five-second in ter trial interval. Three groups of old subjects (mean age 66.5 years) and three groups of young subjects (mean age 37.3 years) received exposure times of 4, 6, and 8 seconds, making up a 3x2 analysis of variance design (Eisdorfer, Axelrod, and Wilkie (1963). The results supported the postulate of Korchin and Basowitz of an increase in cautiousness with age. Older subjects gave a higher ratio of omission errors relative to total errors than did younger subjects, indi cating that the older individuals were more cautious. This 39 again supports the view that older individuals would con form more readily. That is, in an uncertain situation (a conformity situation), the older individual is more apt to take the “cautious behavior path” and conform to the perceptual judgments of the others in his group. In a study by Basowitz and Korchin (1957), a young (mean age 26.8 years and an old (mean age 78.1 years) group were compared in performance on two tests of perceptual closure. The two tests were the Gestalt Completion and the Thurstone adaptation of Gottschaldt*s Concealed Figures. The Gestalt Completion procedure requires the capacity to organize, and the Thurstone test requires the capacity to resist closure. The older subjects’ performance was dis tinctly inferior on both tasks. “Moreover, an analysis of ’wrong’ responses on Gestalt Completion revealed signs of perseveration, fixation on details and concretization by the aged individuals, as well as a more general tendency toward vagueness, and poor articulation of form” (Basowitz and Korchin 1957). The older subjects also Omitted a sig nificantly greater number of items on Gestalt Completion. One interpretation of this finding was that the older sub jects demonstrated excessive cautiousness. That is, . . a defensive reluctance to venture responses for fear of recognizing their inadequacy.” (Basowitz and Korchin, 1957.) This finding could also be interpreted as evidence for the hypothesis of increased conformity with age* In a conformity situation, the subject cannot manifest his cautiousness by not responding, because this alternative is not open to him. He may, however, as the present study proposes, manifest his cautious behavior in the form of Mann (1959)» iu his review of studies relating personality variables to conformity behavior, illustrates six studies which demonstrated a high relationship between conservatism and conformity. Conservatism was the only personality variable that was consistently related to con- Â study by Silverman (1963) compared young and old subjects with regard to ”confidence level required before responding” with ability factors held constant, fhe young and old subjects were matched on ability to recognize words presented tachistoscopically under a forced-response condi tion. fhe subjects were then compared with regard to the number of responses given on an equivalent task which was administered under a voluntaiy response condition. Older subjects gave fewer responses relative to their perceptual ability. These results provided some support to the view of increased caution in responding with age. Again, this increases cautiousness with age lends support to the hypothesis that older individuals will be more conforming. since compared to acting independently against group con formity pressure, it is the safe and more cautious mode of behavior. An approach to psychophysical problems has been sug gested in the last fifteen to twenty years. This is signal detection theoiy described by Swets, Tanner and Birdsall (1961) and by Green and Swets (I966). Signal detection theoiy provides independent measures of response criterion (readiness or reluctance to respond) and sensoiy efficiency. It is possible, therefore, through utilization of this type analysis, to determine to what extent age decrements in performance are due to increased caution as opposed to a decline in sensoiy efficiency. For a complete description of signal detection theory, the reader is referred to Green and Swets (I966). It is adequate for the present purpose to note that the analysis provides independent measures of cautiousness (p) and sensoiy efficiency (d*). High values of p indicate caution and high values of d* indicate high sensoiy efficiency. Craik (1965) employed signal detection theory in an investigation of whether old subjects adopt a more cautious criterion in a detection situation. Stimuli were presented to the subject via headphones which were comprised of either a three-second burst of white noise or the same white noise plus a faint 1,000 e.p.s. phre tone. In one 42 segment of this experiment, a block of fifty stimuli were presented to the subject to which he had to respond ”yes” or ”no” with regard to the presence of a signal (the 1,000 c.p.s. pure tone). A young (age range 21-35, mean age 28.1 years) group of twenty male subjects were compared with an old (age range 60-80, mean age 64*2 years) group of twenty male subjects. The results of this portion of the experi ment indicated that in the yes-no series the old subjects adopted a more cautious criterion. The mean values for log p were 0.03 for the young group and 1.05 for the old group. This difference was statistically significant (t = 2.31; p < .05). An experiment by Milner, Walker, and Beech (1967), utilizing signal detection theory, also found that older subjects adopt a decision criterion that could be inter preted as being more cautious. They used psychiatric patients ranging in age from seventeen to fifty-four and found a significant correlation between age and p (r = ‘ * ‘ 0.79; P < .001) in an auditory yes-no detection situation. Rees and Botwinick (1971) compared young and old subjects with regard to measures of decision (p) and sen sory processes (d9 bn an auditory signal detection task. There was not a significant difference between the two age groups in their detectability performance. However, older subjects adopted higher criterion values than younger subjects. 43 These studies approached the issue of increased caution with age in an experimental manner. They provide support for the hypothesis of an increase in cautiousness with age. Therefore, these studies also lend support to the position that older individuals will he more conform ing than younger individuals since cautiousness has been shown to be highly related to conformity (Mann 1959» Wiener 1958; Wiener 1959), A study by Wallach and Kogan (1961) was specifically directed at the issue of cautiousness in advanced age. The issue was studied by fixing the limits of cautiousness to the choice of two given alternatives: a rewarding but ri#:y alternative, or a less rewarding but safer one. They measured this by employing a questionnaire which consisted of twelve "life situations." A central character was involved in each of the twelve life situations and he was confronted with two alternatives. The task of the subject was to make each choice for the central character by indi cating the likelihood of success that was judged sufficient to decide upon the more risky course. The results indi cated that the older subjects required more certainty of outcome before they would select the more risky alternative. It was found, therefore, that the older subjects, in con trast to the younger controls, responded in a more cautious or conservative way in making decisions. This study may 44 also toe interpreted as supporting the view that older indi viduals are more conforming. It is a greater risk, in the conformity situation, to disagree with the majority opinion; and it is a much lower risk to conform. Botwinick (1966) has extended Wallach and Kogan^s investigation and has obtained the same results. Botwinick says, "It is clear that it is unequivocal that cautiousness increases in advanced age." All the studies mentioned above, which indicate an increase in cautious behavior by older individuals, tend to support the hypothesis of greater conformity on the part of older individuals. That is, the older individual, being more cautious, would conform more often to group influence than younger individuals, since it is a more cautious, conservative, and less risk-involving mode of behavior. This conclusion can also be drawn on the basis of the investigations mentioned by Mann (1959) and hy Wiener (1958, 1959), Age, anxiety, self-concent, ego strength and con formity.— A number of studies have demonstrated a positive relationship between anxiety and conformity behavior. In a study by Walters and Karol (I960), subjects were isolated and were then put in a social influence situation. They found that those subjects who experienced greater anxiety while isolated subsequently showed greater susceptibilily to social influence. In a study by Walters, Marshall and Shooter (i960), subjects who rated high on anxiety con formed more to the experimenter’s suggestions in a Sherif autokinetic situation. In an experiment by Parley (I966), female subjects were made anxious in anticipation of an electric shock. These subjects were found to conform more in an Asch situation than those who were not made to feel anxious. A study by Janis (1955) showed that students scoring high on test and social anxiety were more persuasi- ble and susceptible to social influence than those subjects who scored low on the two types of anxiety. Studies by Berkowitz and Cottingham (I96O), Insko, Arkoff, and Insko (1965), Leventhal and Miles (1964), and Leventhal, Singer and Jones (1965) have found a positive relationship between anxiety arousal and opinion change. There is some evidence that relates age with anxieiy. It seems reasonable that due to social and physical losses, together with increasing responsibilities and commitments, an increase in anxiety with age is likely to result. Veri tably , a number of writers, for example, Kaufman (1940) and Atkin (1940), who have attempted theoretical explana tions of the aging process, have considered anxiety generated by social and physical losses as the fundamental independent variable. Studies by Powell and Perraro (I96O) and Olsen and Elder (1958) suggest an increase in anxiety with age. In both of these investigations, reaction time to stimulus words was employed as a measure of threat or stress. Reaction times to words such as "unhappy," "rest less," "afraid," "worry," and "anxious" were considered generalized "anxiety stimuli" as compared to words such as "priest" or "church" which would be representative of the religious aspect of life. Both studies found reaction time to the anxiety stimulus words to increase with age. In reference to the relationship between anxiety and age, Kuhlen (1964) has stated, "Although the data are by no means as extensive as one might wish, either with respect to the range of symptoms sampled or the range of ages, the evidence does suggest that increasing age brings increasing susceptibility to stress and threat." fhe relationships considered above— individuals scoring higher on anxieiy are more susceptible to conform ity pressures; the increase in anxiety with age— support is given to the hypothesis that older individuals will be more conforming than younger individuals. Numerous investigations have shown a relationship between self-concept and conformity behavior. Typically, self-concept means attitudes regarding oneself. That is, one’s opinion of his own personal worth, and the degree to which a person is complacent and self-confident. In a review of a number of studies of conformity 47 and personality across a variety of situations, Linton and Graham (1959) conclude that patterns of personality do make a person more or less susceptible to social influence. Along with the Moeller and Applezweig (1957) study, they particularly stress the central role of the self-concept, in terms of self-esteem, as a general factor which is inversely related to conformity. This position is supported, by the work of Bray (1950), Kelman (1950), Cohen (1959), Janis and Field (1959) and many others. Thorndike (1938) investigated the effect of group discussion upon the correctness of group decisions with the factor of majority influence considered. He found that the responses of the majority exerted less influence on those subjects who had confidence in themselves. In Hochbaum’s (1954) study, it was also shown that the self-confident subject was less susceptible to the conformity pressures of his group. Asch (1952), Kelley and Lamb (1957), and Wiener (1956) have all shown that less self-confident subjects are more likely to conform to a simulated majority opinion* Mttes and Kelley (1956) have found that when esteem differences are simulated, subjects tend to estimate their own ability and self-esteem according to the esteem they are led to believe they have. This, subsequently, influ ences whether or not they will conform to the norms of the group. Similar studies have resulted in the same 48 conclusions (Mausner 1954; Kelley & Shapiro 1954). The importance of self-confidence may explain why PiVesta (1959) found that conformity decreased as subjects acquired experience with their tasks* League and Jackson (1964), employing a click counting task, report that sub jects scoring low on self-esteem on a personality test con formed more closely to a contrived group judgment than did subjects who scored high. There are a number of investigations that have reported a relationship between age and self-concept. Anticipation of the change in self-concept with age would lead one to predict a curvilinear change. That is, self- concept would become more favorable during the time of gains and increased status, and less favorable during the period when losses are being encountered. A study by Lehner and G^hderson (1953), employing a draw-a-person test, found a curvilinear relationship of self-concept to age. The results of their study showed that men tended to draw larger figures the older they got up to about 30 years of age, thereafter, they drew smaller pictures. Women were found to draw larger pictures up to age forty, after which they began to draw smaller ones. It can be inferred that these trends reflect trends in self- evaluation, since it is often considered that in such picture drawing the person’s self-image is projected. 49 In a systematic effort to investigate self-concept, Mason (1954) administered various measures of self-concept to several groups of subjects from different backgrounds. Mason found that indigent old people had more negative self-concepts than did a group of independent middle-class elderly, and both, in turn, had a more negative self- concept than did a more youthful low-economic group. An identical age trend was reported by Seward (1945), who administered intelligence tests to younger and older college professors and recorded the comments his subjects made during the course of the experiments. It was found that older group members made twice as many se If-belittling remarks than did the younger group members. A study by Lehner and Silver (1948) and one by Giedt and 2ehner (195D employed a draw-a-person test in an investigation of age changes in self-concept. They required their subjects to draw a person and then to label the age of the person drawn. It was found that subjects under 25 tended to label their drawings with ages older than their own, while those over 25 labeled the drawings with younger ages. The interpretation offered for these results suggested that approximately age 25-30 is the idealized age in the American culture. Herman’s (1949) findings also suggest that increasing age may be a threat to one’s self-concept, especially for women. Herman found 50 that older women more often failed to record their date of birth in biographical sources. In reference to the evi dence on age changes in self-concept, Kuhlen (1959) has stated, "... the data seem to suggest that, consciously or unconsciously, the individual, as he gets older, has a less positive attitude toward himself." The relationships discussed above— individuals with lower self-concepts are more likely to conform; older individuals as compared to younger individuals score lower on self-concept measures— support the hypothesis that older subjects will be more conforming than younger subjects. Previous empirical and theoretical studies, although not perfectly, concur, that the overoonformist has less "ego strength." For example, Hoffman (1953), using judg ments of line lengths, found conformers to differ from non- conformers by scoring lower on ego strength measures. The personality measures obtained by Hoffman were from a sentence-completlon test, two questionnaires, and the Thematic Apperception Test. Crutchfield (1955) reports similar findings. The older individual has been shown to have a lower "ego strength" than younger age groups (Meugarten & Gutmann 1958; Rosen I960; Shanon & Sharon 1965). Rosen (I96O) divided ego strength into "ego involve ment" and "quality of ego involvement." It was found that 51 the extent of involvement of the ego in outer world events decreases with age. She measured ego involvement by two scores— the number of introduced figures in TAT stories, and the presence in a stoiy of any element of content involving "conflict" or "controversy." Rosen also found that the quality of ego involvement declines with age; the shift was from assertive to passive affect. Rosen also measured quali*^ of ego involvement in two ways— a rating of the emotional intensity of the TAT stories, and a count of whether the activity described in the stories was assertive or passive. Similar results were found by Shanon and Sharon (1965) who tested Israeli men aged 39-61. They also found the number of optimistic themes, positive end ings to stories and achievement themes fell with age. The relationships mentioned above— individuals scoring low on ego strength are more likely to conform; older individuals as compared to younger individuals score lower on measures of ego strength— support the hypothesis that older individuals will be more conforming than younger individuals. Sex and Conformity Khower (1935) found women to be more influenceable than men. A number of investigators before and since have confirmed the findings that females are more conforming than males (Burt 1920; Crutchfield 1955; Jenness 1932; 52 Kirkpatrick 1936; fuddeiüiam, Mac Bride & Zahn 1958; and others)* Beloff (1958), Tuddenham (1957) and Applezweig and Moeller (1958) found women to he much more conforming than men when confronted with pressure to conform by their groups* This finding was substantiated by Coleman, Blake, and Mouton (1958) who obtained correlations between diffi culty of general information items and the tendency to con form of .58 for men and .89 for women. Tuddenham found similar differences when comparing college men and women. However, other studies have failed to find a sex difference in conformity behavior (Phelps 1966; Klein 1970). Studies on acquiescence have shown little relation to sex (Christie & Lindauer 1963)* In fact, men tend to obtain slightly higher acquiescence scores than women (Mahler 1962; Hilgard, Lauer, & Melei 1965). Kahana and Coe’s (1969) study on conformity found that in contrast to the major portion of the literature, males were more con forming than females. Of the 24 women that participated in the study, only 10 rated high on conformity. On the other hand, of the nine men who participated, all scored high on Endler’r S (1965) study is another example of a con formity study that failed to find a sex difference in degree of conformity. In Endler’s study, however, all the 53 confederates were male, while the subjects were both male and female. In most of the above-mentioned studies where a sex difference was discovered, the sex of the subjects was the same as the confederates. Endler posits that perhaps females conform less to male confederates than they do to female confederates. He further postulates that this may be because they identify to a larger degree with other females, or, because they are more concerned with obtaining social approval from females as a group than from males as a group. Therefore, Endler maintains that in his investi gation the presence of male confederates may have reduced the conforming behavior of the females to the conformity level of the males. There continues to be a discrepancy among experi mental findings in the degree of conformity between males and females. Therefore, the present investigation will be interested in determining if females are more conforming than males. It has been well established, indicated by the studies mentioned above, that when the subjects in the conformity situation are of the same sex, there are sex differences in conformity behavior (females being more con forming). It will be a hypothesis of this investigation, in contrast to Endler’s (1965) data, that females will be more conforming, on each of the experimental tasks investi gated, even when the number of males and females in the 54 conformity situation are equal. This hypothesis is partially based on a study by Allen and Crutchfield (1963). They investigated whether experimentally reinforced conformity would generalize to subjective stimuli. The experimenters told the subjects that the false group consensus was correct on objective stimuli. An increase in conformity to group pressure was found on objective stimuli receiving such authoritative confirmation of the group’s responses. The results also indicated that this experimentally reinforced conformity generalized to other objective and subjective material on which reinforcement was not given. Pertaining to the current issue, it was found, taking both reinforcement and nonreinforcement conditions together, females were clearly more conforming than males on vocabulary items and the trend was similar for perceptual items. The results of this investigation demonstrated that females were not only more conforming than males, but also that this result tended to generalize over all the tasks included in the study, both objective and subjective. Further support for this hypothesis is based on the investigations of Koponen (1957) and Spangler and Thomas (1962), In both of these studies, females were found to score significantly higher than males on the affiliation scale of the Edwards Personal Preference Schedule. As 55 mentioned earlier, individuals scoring high on affiliation needs have been found to be more susceptible to conformity pressures (Applezweig & Moeller 1957; Raven 1959; Thibaut & Strickland 1956; Tuddenham 1958). The Koponen and Spangler & Thomas studies found females to score significantly lower than males on the achievement scale of the EPPS. As dis cussed previously, those individuals scoring low on need achievement have been found to be more susceptible to con formity pressures (Krebs 1958; SameIson 1957; Thibaut & Strickland 1956; Tuddenham 1958; Walker & Heyns 1962). The Koponen and Spangler and Thomas studies also found females to score significantly lower than males on the dominance scale of the EPPS. Again, as described earlier, individuals scoring low on dominance have been shown to be more susceptible to social influence (Barron 1953; Bray 1950; Hoffman 1953; Kelman 1950). Another basis for the hypothesis that females will be more conforming than males in each of the sub-experjments was a discussion by Bass (1961). He maintains that sex differences in conformity behavior may be involved with motivational orientation. He argues that men are more "task-oriented,” while women are more "social-interaetion oriented.” Therefore, he maintains that because women are more concerned than men with obtaining social approval from other individuals, women should tend to be more conforming. 56 Bass* discussion was based on studies carried out at \ Louisiana State University and on the study by Tuddenham and Mac Bride (1958). In the Louisiana State studies, it was found that college men were much more task oriented, more concerned with getting the job done, getting the correct answer and doing the best job possible. The women subjects, however, were more interaction oriented, were concerned about receiving social approval from others, and were concerned about being accepted members of their groups, It should also be mentioned, that the Klein (1970) study, mentioned above, did not find a statistically sig nificant sex difference. However, the data did show an obvious trend indicating that females are more conforming than males. Task Difficulty and Conformity The present investigation also investigated the effect of the ambiguity or difficulty of the stimulus. It has been shown that the more ambiguous the stimulus and the more difficult the task, the greater is the proportion of subjects conforming to the objectively incorrect judgment of others (Asch 1951; Asch 1956; Berenda 1950; Carpenter, Carpenter & Weiner 1956; Chipman 1966; Coleman, Blake & Mouton 1958; Endler 196O; Fisher, Williams & Lubin 1957; Hunter I968; Kelley & Lamb 1957; Klein 1970; London & Lim 1964; Miller & Tiffany 1963; Weiner 1956; and others). 57 In reference to task difficulty in perceptual con formity situations, it has been illustrated that when the perceptual situation is clearer and more distinct, and when the perceptual disposition is stronger, conformity is less likely to occur (Coffin 1941). It has been demonstrated that the larger the contrast on a discrimination judgment, the less conformily will occur (Asch 1948; Asch 1956; Crutchfield 1955; Klein 1970; Luchins 1944; Thrasher 1954). The more difficult the observational conditions (e.g., the speeded presentation of clicks in a metronome click count ing situation), the more conformity will occur (Blake, Kelson & Mouton 1957). In Sharif’s (1935) experiments employing the auto kinetic phenomenon, the stimulus was highly ambiguous in reference to the judgment of the distance the stationary spot of light appeared to move. Considering the absence of physical cues to distance, and the situation which required absolute judgments, the tendency for the judgments of the subjects to converge does not seem surprising. The data led Sherif (1948) to state, ". , . the greater the vague ness of the stimulus field, the greater the influence of ... suggestion, prestige, social pressure and propaganda.** At about the same time, îkirphy (1937) had found a marked tendency for subjects judging which of two cards had more dots on it to modify their judgments to conform to the 58 majority. It was shown that the influence of the majority opinion was appreciable when ambiguity was introduced by having approximately an equal number of dots on each card. Luchins (1955) utilized a situation similar to Asch’s in an investigation of conformity in a perceptual situation. He required his subjects to judge which was the shorter of two lines differing in length by from one-sixteenth to one inch. It was found that generally the influence of the confederates was greatest when the stimulus was ambiguous. That is, when the difference between the two lines was least. A study by Luchins (19^5) is another example illustrating an increase in conforming behavior under tasks of high difficulty. Children aged eleven and thirteen were required to interpret ambiguous drawings. A confederate’s answer preceded the subject’s response. The children tended to offer the same interpretation as the confederate when the drawing left enough scope for ambiguity. However, when the drawing presented a clearly structured object, the subjects tended to remain independent of the con federate’s interpretations. In a study by Blake, Helson, and Mouton (1956), subjects were asked to solve arithmetic problems without paper and pencil in a "simulated group" situation. The results indicated that the subjects were more influenced by the decision of "others" as the problems became more 59 difficult. In a study by Kanareff and Lanzetta (1958), subjects were required to judge which of two recorded tones was highest in pitch. Ambiguity was maximal in that the second tone was equal in pitch to the first on all test trials. The investigators found an especially high degree of imitation of partners’ choices in this situation. The studies mentioned above appear to indicate that as the task becomes more difficult, or as the stimulus becomes more ambiguous, there is an increased tendency for the subject to depend on or look to others as sources of information for the selection of the correct answer. A direct relationship appears to be self-evident when the following question is asked: Are conformity pressures more effective when subjects are faced with ambiguous stimuli? However, all studies have not yielded this relationship. The experiments by Schonbar (19^5), utilizing an autokinetic situation like that of Sherif (1955), found that there may still be a marked social effect or conver gence of judgments when subjects are required to judge the movement of a point of light which actually does move. This is a much less ambiguous situation than the one reported by Sherif where the point of light is always stationary. In a study by Olmstead and Blake (1955), group pressure was exerted on subjects to conform to the number of metronome clicks heard. The results indicated that 60 group pressure operates even when the discrepancy is larger than one or two metronome clicks. This illustrates that group pressures to conform are present and do have an effect on an individual's perceptual judgment in situations of low task difficulty. In a study by Blake, Helson & Mouton (1956), the difficulty of the items in a metronome counting task was varied to ascertain the effect of difficulty of task on susceptibility to group opinion. It was assumed that there would be a greater yielding to group pressure under con ditions of higher rates of clicks. It was also assumed that it would be more difficult for subjects to judge the correct number of clicks as the rate of clicks increased. However, the data of their study indicated that as the rates of clicks increased in difficulty (14-0, 1?0, 200 clicks per minute), there was not a comparable rise in conformity. The authors point out, however, that if the metronome clicks had been calibrated to provide a greater graduation in difficulty than was achieved, the expected relationship of increased difficulty with increased con formity would have probably been attained. This study, nevertheless, must be considered as contradictory to the findings of Asch (1956), Chipman (1966), Hunter (1968), Endler (i960) and to the findings of many other studies (mentioned above) which have found this relationship. 61 A variation of the original Asch (1951) experiment was conducted by Asch (1951) in an exploratory attempt to ascertain the effect of increasing the contradiction between the majority and the critical subject. The purpose of this was to determine if increasing the magnitude of discrepancy would aid the critical subjects in asserting their independence and in disclaiming the majority more decisively. In this experiment, there was a total of ten length-of-line judgments, five of which were critical. The judgments of the confederates departed greatly from the actual state of affairs by amounts ranging from one to seven inches. Under these conditions, 28 percent of the critical responses were errors identical to those of the con federates. In a control group, the errors were 2 percent of the total number of critical judgments. The mean number of errors was 1 . 4 - in the experimental group and 0 . 1 in the control group. In referring to this finding, Asch said, "Increasing the objective discrepancies here did not abolish the majority effect nor appreciably decrease it." This illustrated that, in contradiction to most reported studies, decreasing ambiguity does not decrease conformity. While susceptibility and conformity may vary directly with task difficulty, results of experiments reported do not consistently demonstrate the validity of this relationship. The present investigation, therefore, was partially aimed at substantiating the following 62 relationship within each of the five sub-experiments : Conformity will be greater at higher levels of task difficulty. Nature of the Experimental Task A secondary concern of the present investigation was to ascertain differences in conformity due to the nature of the experimental task. Tasks in previous con formity experimentation have ranged from questions that have literal answers, to social attitudes that cannot be presumed to have "correct" responses (Blake, Helson, & Mouton 1958; Crutchfield 1955; Ferguson 194-4; Festinger & Thibaut 1951)» The tasks that consists of material based on fact and obvious perceptual judgments have been shown to be most resistant to change from conformity pressures. However, items such as attitudes toward social problems, which obtain their validity from a social frame of reference, appear to be more subject to social influence. In reference to the issue of the nature of the tasks, Blake and Mouton (1961) have stated: Conformity behavior increases when it is necessary for an individual to rely more heavily on the responses of others in making his own adjustment. Attitudes are more easily shifted than are reactions to factual or logical items, probably because atti tudes are more social in character. In investigations where the ease of discrimi nation had been varied by changing the conditions of presentation, it was found that conformity increased as 63 subjects were prevented from utilizing an objective frame of reference in giving their judgments (Deutsch & Gerard 1 9 5 5 ; Luchins 1 9 4 4 ; Raven & Rietsema 1 9 5 7 ) - In the present experiment, the nature of the tasks range from a clear-cut perceptual judgment task (Experi ment I) to a task involving social attitudes (Experiment V); that is, from a task with an objective frame of reference to a task with a social frame of reference. The tasks in the present investigation increase in ambiguity from Experiment I to Experiment V. In Experi ment I, the stimulus is always present while the subject is making his response. In Experiment II and III, the stimuli are not physically present while the subject is making his response, but rather they must be judged from memory. Deutsch and Gerard (1955) found more conformity on line judgments when responses were made several seconds after the stimuli had been removed. In Experiment IV, the correct answer (objective frame of reference) is never presented to the subject. In Experiment V, there is no correct answer since the task deals with attitudes. Issues of social opinion are much more ambiguous, subject to argument and disagreement. Matters of fact are more easily tested and less likely to be ambiguous. Therefore, it is hypothesized (stated more formally in the next section) that conformity will increase from Experi ment I to Experiment V because correct answers to the 64 - stimulus presentations will become less easy to obtain with assuredness. Hypotheses The above literature survey led to the formulation of four hypotheses that this study tested. Before listing them, it may be wise to define the term "informational social influence" as an influence to accept information obtained from another or others. That is, "social influence is regarded as informational when the source of information is regarded as a cue to the correct answer ..." (Graham 1962). "Social conformity" is defined here as ". . . the public and/or private agreement of an individual with an opinion or judgment he had not held before it was presented to him." (Sampson 1964-. ) The following hypotheses were tested using social conformity as the dependent variable: Hypothesis I: Informational social influence is greater upon the judgments or opinions of older individuals than upon the judgments or opinions of younger individuals. Further this hypothesis will hold in each of the sub experiments on all five tasks employed. Hypothesis II. Informational social influence is greater upon the judgments and opinions of females than upon the judgments or opinions of males. Further, this hypothesis will hold in each of the sub-experiments on all five tasks employed._______ 65 Hypothesis III. Informational social influence is greater upon the judgments or opinions of the subjects as the stimulus of the task becomes more ambiguous (i.e., as the task becomes more difficult). Further, this hypothesis will hold in each of the five sub-experiments on all five tasks employed. Hypothesis IV: Informational social influence is greater upon the judgments or opinions of the subjects as the nature of the experimental tasks proceed from an objective to a social frame of reference (i.e., there will be an increase in conformity from Experiment 1 to Experi ment V). Compliance vs. Private Acceptance Arriving at a distinction between "compliance" and "private acceptance" has been a concern of social psy chologists for a number of years. Although there have been few experiments that have dealt with this distinction, a number of social psychologists have at least discussed the issue (Festinger 1953, 1957; French & Raven 1959; Hovland, Janis & Kelley 1953; Johoda 1959; Kelley & Volkart 1952; Kelman 1961; Lewin 1951; Newcomb, Turner & Converse 1965; Secord & Backman 1964). The precise terms used in the distinction have varied from investigator to investigator. Dichotomies used in the distinction include : public versus private atti tudes, overt versus covert behavior, compliance versus 66 internalization and own versus induced forces. When conducting an empirical investigation on con formity, failure to consider the distinction between compliance and private acceptance could result in treating responses which are extremely different psychologically as identical. When an individual is faced with a unanimous disagreement from a group, he may comply or not comply. Whatever his public response, the person may privately agree or disagree with the group. Therefore, there are four possible outcomes of a conformity experiment : (1) compliance and private acceptance, (2) compliance without private acceptance, (3) no compliance and no private acceptance, and (4) private acceptance without compliance. Two apparently identical responses of public conformity may, therefore, actually reflect two totally disimilar psychological states (compliance with or without private acceptance). The present investigation adopted the distinction made by French and Raven (1959). They have noted, although stated as a dichotomy, that the distinction between com pliance and private acceptance may represent an underlying psychological continuum. The continuum could be expressed as the degree to which the subject’s behavior is dependent upon the presence of some external agent. The "conforming" behavior is considered "compliance" if it is evidenced only in the presence of the person or persons who induced 67 the original conformity behavior. The "conforming" behavior is considered as "private acceptance" if this behavior subsequently does not depend on the presence of the inducing agent. Theoretically, in between these two extremes can be varying degrees to which the behavior is evidenced without the presence of the other. Presumably, this would represent varying degrees of private acceptance. Lewin (1951) was pointing in a similar direction in his distinction between "own" and "induced" forces. For example, Lewin states, "Forces may correspond to a person's own needs .... Many psychological forces ... do not, however, correspond to his own wishes but to the wishes of another person .... These forces can be called induced forces." The similarity between Lewin's distinction and the distinction between compliance and private acceptance is relatively clear. The existence of an "induced" force acting on a person and having an effect on his behavior corresponds to compliance without, necessarily, private acceptance. If the "induced" force becomes an "own" force, then you would have the situation of private acceptance accompanying the compliance. French (1944) further elaborates on the dis tinction between "own" and "induced" forces. He discusses the rejection or acceptance of "induced" forces in the following manner: 68 An induced force which is accepted to a high degree produces in the person additional own forces in the same direction, so that the behavior instigated becomes relatively independent of the inducing agent and will occur even if his power field is removed. But an induced force which is rejected produces in the person opposing forces with the result that induced behavior will cease as soon as the inducing field is withdrawn. (French 1944). Here, again, is presented a distinction between compliant behavior with and without private acceptance. The question arises as to how one tests for private acceptance. This becomes a meaningful question since in any conformity study it is possible to distinguish between four types of situations which may appear similar on the surface (these are described above). An example of one of these four outcomes will clarify the problem of failing to distinguish between compliance and private acceptance. Coch and French (1948) provide an example of a person who complies but privately rejects the group norm. They discuss the case of a woman clothes-presser they observed in their investigation of the resistance to behavioral changes in an industrial setting. The person was placed in a work group that had its own production norms and insisted on conformity to those norms by group members- When the presser joined the group, the rest of the group was producing at a rate of approximately 50 units per hour. After the new girl had been in the group a few weeks and became accustomed to the work, her production began to exceed that of the rest of the girls. The group 69 quickly began, in the words of Coch and French, to "scape goat" the girl. The scapegoating included several forms of harassment. The enforcement of the group was effective. The girl's productivity promptly dropped and subsequently stayed at the same level as the rest of the group. At this point there is no clear evidence whether this was merely compliance without private acceptance or whether the girl now privately accepted the group norm as the appropriate level of productivity. When the group was later dispersed in the factory, and were no longer present, the girl's productivity doubled within four days and stabilized at this higher figure. This example provides a clear indication of compliance without private acceptance. This study represents a good paradigm for the joint study of compliance and private acceptance. Most studies in the social psychological literature have employed either measures of compliance or measures of private acceptance, but seldom both simultaneously. This historical fact is unfortunate because it would be of interest to examine how the two types of conformity inter act. In reference to this issue, Kiesler ( 1 9 6 9 ) has stated : The notion of private acceptance is obviously an important aspect of the study of group pressure and "conformity." However, ... it has received relatively little study. One obvious reason for this is that it is more difficult to study private acceptance experimentally than it is either to study compliance or to ignore the distinction 70 altogether. Needed even more, however, are experi ments in which both variables are studied at once. There have been a few investigations that have experimentally considered the distinction between compliance and private acceptance. For example, in Sherif's (1935) original investigation, one of the considerations was to ascertain whether the effect of the social influence was impermanent or continuous. Sherif tested the subjects' autokinetic judgments 24 hours after the initial session. It was found that the social influence continued to have an effect on these subjects when tested alone after a day had elapsed. In a study by Bovard (1948), subjects were tested alone 28 days after the initial conformity session. The subjects' judgments, of the autokinetic stimuli, were found to be close to their judgments obtained in the presence of a confederate. Similar results have been reported by Rohrer et al. (1954), who used a year as the interval of time between observations. In a study by Hardy (1957), the topic of divorce was used in an attempt to study compliance and private acceptance. The subjects were first required to give public responses. The subjects then went into séparaLe rooms and filled out a questionnaire to give their private responses. Nearly 50 percent of the subjects complied by changing their opinion markedly toward the group, as compared with their previous private responses. 71 Private acceptance was also found. Approximately 4-5 per cent of the subjects privately shifted toward the group on at least half of the twelve statements. There have been a few other studies that have investigated the effect of social influence on opinions which have found compliance to be accompanied by private acceptance (Gerard 1954; Raven 1959). The present investigation considered both com pliance and private acceptance under the rubric of conform ity. Although the major concern of the investigation was with that portion of conformity designated as compliance, an experimental attempt was made to distinguish between these two forms of social behavior. The question raised above as to how to test for both compliance and private acceptance in the same experiment must be considered. If the effects of social influence result in a subject publicly complying, one can not determine solely on this basis whether or not the compliance which is observed is accompanied by private acceptance. If it is accompanied by private acceptance, the behavior being exhibited should persist even when the group exerting the influence is removed. On the other hand, if compliance is not accompanied by private acceptance, then the individ ual should revert to his former mode of behavior when the group exerting the pressure is removed. Therefore, the operational way in which the present investigation 72 identified private acceptance was in terms of the after effects remaining from social influence when the conformity pressures had been alleviated. The actual research design is presented in the General Method section which follows. Restatement of the purpose of the present investi gation.— The purpose of the present investigation was to focus on determining whether the greater degree of con formity expressed by older subjects in a visual perceptual judgment situation (Klein 1 9 7 0 ) is unique to that situation, or whether it is a general tendency for the older person to be more conforming regardless of the experimental task. It was also the purpose of this study to investigate sex differences in conformity behavior and variation in con formity behavior due to ambiguity of the stimulus on a variety of experimental tasks. As a secondary concern, this investigation was interested in determining if con formity behavior varies due to the nature of the experi mental task. CHAPTER II GENERAL METHOD Subjects One hundred and twenty subjects were recruited for participation in the general experiment. The subjects were all paid volunteers. Sixty of the subjects were young, ranging in age from 17 to 24 years with a mean age of 18.6 years. Sixty of the subjects were old, ranging in age from 60 to 81 years with a mean age of 67.2 years. Half of the subjects in each age group were male the other half female. There were twenty-four subjects (half young and half old, half male and half female) in each of the five sub-experi ments. The young subjects were students attending the University of Southern California. The older subjects were recruited from the Institute of Lifetime Learning in Long Beach, California. This institute offers adults over the age of 55 continued learning through lectures, classes and seminars. The number of subjects utilized in this experiment was determined by the considerations of the size of the differences expected, the power of the designs, the control measures employed, and the practical aspects of obtaining subjects. Considering these points 24 subjects per experiment appeared to be a sufficient number to determine ______________________ 75_______________________ 74 if the main effects were significant, to test the inter actions, and to make certain post hoc comparisons. Experiments designed to evaluate the role of homogeneity among subjects regarding their degree of skill, and equality of ability and knowledge have generally failed to illustrate a consistent relationship between this factor and susceptibility to conformity pressure (Eestinger et al. 1 9 5 2 ; Eestinger & Thibaut 1 9 5 1 ; Gerard 1 9 5 4 ; Hovland, Janis & Kelley 1 9 5 3 ; Janis & Hovland 1 9 5 9 ) # Therefore, in the present study, no strict emphasis was placed on equating subjects on these factors. However, as much care as possible was taken to select subjects that had attained at least a minimum educational level. Conformity Apparatus There were three types of conformity apparatus employed in the present investigation. The apparatus used by Klein ( 1 9 7 0 ) was utilized in sub-experiments I, III, and V. In Experiment II, a modification of the apparatus used by Mathie ( 1 9 5 9 ) was utilized. In Experiment IV, a technique used by Deutsch and Gerard ( 1 9 5 5 ) was modified to measure conformity. Klein's ( 1 9 7 0 ) annaratus.— The devising of laboratory techniques to measure conformity was pioneered b y Sherif ( 1 9 3 5 ) and by Asch ( 1 9 5 1 ) . A group of experi ments was reported by Asch ( 1 9 5 1 ) , which have already been 75 described, in which subjects were given the task of com paring the lengths of successive sets of lines. The experimenter instructed all the members, except one, to respond erroneously. The single uninstructed member was then faced by a situation in which all the other group members were perceived to be in error. The question was whether or not the uninformed subject would conform to the incorrect judgment of the confederates. A practical limitation of Asch's procedure was the necessity for the experimenter, and also his confederates, to spend an hour or longer testing each subject. The limitation of Asch's technique and apparatus led to designing the apparatus used by Klein ( 1 9 7 0 ) , which was based on the apparatus employed by Crutchfield ( 1 9 5 5 ) . This modified Crutchfield apparatus makes it possible to test four subjects at a time in separate stalls. It pro vides a two-choice answer reporting system, utilizing a switch and light communication system that enables the experimenter to introduce false information into the system. In front of each subject is a panel which has eight six-watt bulbs placed side by side in pairs down the face of the panel. Each pair of bulbs represents each of the four subjects (as far as the subjects are concerned). At the base of each of the subject's panels are two mercury switches. Mercury switches were used since they 76 silent; this prevents the subjects from knowing when the other subjects respond. The two switches on the subject's panel light only the two bulbs indicating his response. They also light two bulbs on the experimenter's panel, in order for the experimenter to record the subject's response The experimenter's panel also has eight bulbs, two repre senting each of the four subjects that are tested at the same time, which enables the experimenter to record the subjects' responses. At the base of the experimenter's panel are six mercury switches. These control the lighting of the remaining six bulbs on each of the subject's panels. Each of the four adjacent panels have side wings which form an open cubicle. In this way, although each individ ual is sitting side by side to his fellow subjects, he is unable to see their response panels. In the present testing situation no confederates are required. All four subjects are tested simultaneously in a thoroughly standardized situation. Also, the experi menter is able to exercise highly flexible control of the simulated group judgments. Modified Mathie (1959) apparatus.— This apparatus seats four people next to one another separated by wooden panels forming a row of open cubicles. This apparatus does not require any confederates, and tests four subjects simultaneously. In front of each subject, on a "flip-type" 77 rack is a stack of index cards each having one number printed on it from one to fifty. The subject responds on each trial by turning over the card which indicates his response. On a table in front of the experimenter, facing the subjects, are three "flip-type" racks each with a stack of index cards numbered from one to fifty. By turning over particular cards, the experimenter is able to intro duce false information into the system. This apparatus allows multiple-alternative tasks to be employed. The actual procedure for utilizing this apparatus is described when discussing Experiment II. Klein ( 1 9 7 1 ) apparatus.— This apparatus is based on a technique used by Deutsch and Gerard (1955). This apparatus seats four people next to one another, each in an open cubicle. No confederates are required, and all four subjects are tested at the same time. The apparatus provides an unlimited answer-reporting system. Each sub ject is given a "magic slate" on which he is told to record his responses to the stimuli presented. In front of each subject, is a rack covered with foam rubber on which to place his "magic slate" after recording his response. The experimenter has three "magic slates" on which he records the supposed responses of the subjects. He actually records false information deceiving the sub ject. In front of the experimenter are three racks on which he places the three "magic slates" after recording 78 the contrived responses of the subjects. This apparatus makes it possible to use tasks that have multiple and un defined response outcomes. The procedure for using this apparatus is presented when discussing Experiment IV. General Procedure This investigation was comprised of five separate experiments. Each followed the same research design; however, they varied slightly in statistical design and analysis. Experiments I through V were all interested in age, sex and task difficulty as predictors of social conformity. In the succeeding chapters, each of the sub experiments are described in detail which includes a brief introduction, a complete methodology (subjects, apparatus, procedure, directions, data analysis) and results. Below is a list of the five conformity sub-experi ments. Experiment I: A visual perceptual judgment task— the subject’s task was to judge which of two circular discs was largest in size. (A modified replication of Klein's (1970) study.) Experiment II: An auditory perceptual judgment task— the subject's task was to judge the number of metronome clicks heard. Experiment III: An auditory signal detection task— the subject's task was to report whether he heard a pure tone plus noise, or noise alone, in his right ear while 79 sometimes receiving additional information in his left ear. Experiment IV: A problem solving task— the sub ject's task was to solve one-operation arithmetic problems. Experiment V: A social attitude task— the subject's task was to rate statements on nationalism as to whether he agreed or disagreed with them. Research design.— The five experiments followed an identical research design. The reason for arriving at the particular design which was employed was primarily to obtain a measure of both compliance and private acceptance as two distinct measures of social conformity. The following describes the research design for each of the sub-experiments : First, individual performance was evaluated under "alone" conditions. Next, the subject was tested in a "conformity" situation where conformity pressures were exerted to shift responses away from those under the alone condition. The subject's behavior was then again measured under the alone condition after the conform ity pressures were relieved. Finally, the subject's behavior was again measured under the alone condition one week later. Compliance, which will be referred to as con formity, was evidenced if the subject's performance in the conformity situation was different from his performance the first time he was tested alone. Private acceptance was evidenced when performance under the post-conformity alone 80 condition was different from the pre-conformity alone condition in the direction of the divergent reports (con- Itrived) given in the conformity situation. Further evidence' . for private acceptance was evident when the performance I under the second post-conformity alone condition (one week later) was different from the pre-conformity alone con dition. This last test determined whether the subjects retained the group norms even after the social influence had been alleviated over a period of time. Diagram- matically, this research design takes the following form: I A C A---------------- A ■ initial one week I session later I A tested under "alone" condition J C tested under "conformity" condition I Alone situation.— Subjects were scheduled to come I ! to the experiment in groups of four. The subjects were j then taken to the room where one of the three types of I I apparatus was set up. They were asked to seat themselves I I in one of the four stalls. Each subject responded inde- Ipendently without any knowledge of the responses of the I other subjects. The instructions were given by a tape recording (the subjects were also given a printed copy of the directions to read), describing the apparatus and the task they were to perform. When the experimenter was assured that the subjects understood the requirements of 81 the task, the series of trials began. This same procedure was followed after the conformity situation and again one week after the initial session. Conformity situation.— The subjects were already seated in one of the four stalls of one of the three types of apparatus. The instructions were given by a tape recording (the subjects were also given a printed copy of the directions to read), informing the subjects that this portion of the experiment was interested in their per formance in a group situation. The instructions described the apparatus and the fact that it was designed so that each subject was not only able to report his own response, but that he would also be informed of the responses of the other group members. The subject then selected a sealed envelope at random, in which he found a card indicating the order in which he was to respond (either 1st, 2d, $d, or 4th). Each envelope contained a card with "4th" printed on it. This led each subject to believe that the other three members of the foursome had been assigned to the 1st, 2d, and 3d response-order positions. The task was then described. A portion of the trials were critical trials, occurring randomly, on which the contrived report of the responses of the three peers displayed a unanimous selection of an incorrect response. The remaining trials were innocuous where peer judgments were accurate. The innocuous trials were included in order to maintain the 82 subject's acceptance of the genuineness of the apparatus and the situation. When the experimenter was assured that all the subjects understood the requirements of the task, the series of trials began. Actually, the subject had a wrong understanding of the situation because he had been deceived. In the case of those subjects tested with the Klein (1970) apparatus, the apparatus was not actually wired the way the subject had been informed. There was actually no connection between the panels of the four subjects. Rather, they were wired indentically to a control panel where the experimenter sat. It v/as the experimenter who sent all the information which appeared on the panels of the sub jects. The wiring was in parallel so that whatever signals were sent by the experimenter, simultaneously and identi cally appeared on all four subject's panels. In the case of the modified Mathie apparatus, and the apparatus based on the modified Deutsch and Gerard technique, the experi menter controlled the information the subjects received by turning over particular cards or by writing false infor mation on the experimenter's "magic slate" respectively. Therefore, on the critical trials, the subjects received the unanimous incorrect responses simultaneously. The quantitative scores of the subjects are not able to account entirely for the reasons of the various conforming tendencies that were observed. Therefore, an 83 attempt to understand the behavior of the subjects stemmed from observations of the subjects and from a post-experiment questionnaire. (An example of the questionnaire used is given in the Appendix, pp. 294-296.) The questionnaire aimed at ascertaining the following: what the subjects actually experienced during the experiment; whether or not the subjects were deceived; whether they considered the other subject's responses in arriving at their own responses; etc.; personality characteristics of the sub jects (self-concept, happiness, shyness, cautiousness); the subjects' child-rearing experiences. General Data Analysis The actual scoring procedures and statistical designs varied slightly among the five experiments; these are described in the following chapters. Basically, they consisted of various analysis of variance designs and post hoc comparisons. Although the statistical analyses differed slightly, each of the five experiments conducted the same basic tests. The following describe these tests: Scoring: 1st "A" score = performance score when first tested alone. 2d "A" score = performance score when tested alone immediately following the conformity situation. 3d "A" score = performance score when tested alone one week after the initial testing. "C" score = performance score under the _________________ conformity situation._________________ 84 Test 1: "C" score minus 1st "A" score.— Scores in the first alone situation on the critical trials were subtracted from scores in the conformity situation on the critical trials. These difference scores were used in the analysis. This test determined if there were signi ficant differences between the age groups, the sexes, and the levels of task difficulty in the degree of compliance (conformity). Test 2: 2d "A" score minus 1st "A" score.— Scores in the first alone situation were subtracted from scores in the second alone condition (post-conformity). These difference scores were used in the analysis. This test determined if there were significant differences between the age groups, the sexes, and the levels of task diffi culty in degree of private acceptance. Test 3: 3d "A" score minus 1st "A" score.— Scores in the first alone condition were subtracted from scores in the third alone condition (one week later). These difference scores were used in the analysis. This test determined if there were significant differences between the age groups, the sexes, and the levels of task difficulty in degree of private acceptance. It also determined the persistence of the group norms after the social influence had been alleviated over a period of time. The results of tests 1, 2, and 3, iu each sub experiment, were separately quantitatively compared 85 (presented in the chapter on General Results). This entailed three 2x2x5x5 analysis of variance designs, one for each of the three tests (1, 2, and 5)- Score entries in each cell were based upon the results of each of the five experiments. The main purpose of these combined analyses was to determine if there was a significant difference between old and young subjects on conformity (compliance and private acceptance) over all the tasks combined; to see if there was a significant difference between male and female subjects on conformity (compliance and private acceptance) over all the tasks combined; to see if there was a significant difference beween the levels of task difficulty on conformity (compliance and private acceptance) over all the tasks combined; and to determine if there was a significant difference between the types of tasks on degree of conformity (compliance and private acceptance). Post hoc comparisons were also utilized in analyzing the data on the degree of conformity between the various tasks. The question as to whether old or young subjects are more conforming on each of these tasks, was answered by the individual analyses that were conducted in each sub-experiment (described in the following chapters). CHAPTER III EXPERIMENT I It has been consistently demonstrated that knowledge of the reactions and responses of others can influence behavior in the direction of conformity. Specifically in the area of visual perceptual judgments, this result has been found by a number of investigators (Asch 1951 ; Endler I960; Kanareff & Lanzetta I960; Klein 1970; Luchins 1955; Schonbar 194-5; Sherif 1955; etc.). In a visual perceptual judgment conformity situ ation, which was the setting for this experiment, the subject is faced with an apparent disagreement between his own clear perception and a unanimous contradictory con sensus of the other subjects. The question arises as to what the subject will do when confronted with these cir cumstances. Will he rely on the evidence of his own senses and respond independently, or will he yield to the judgment of the group? This experiment focused on determining if task difficulty, age, sex and an interaction of these variables have an effect on conformity behavior. More specifically, this experiment attempted to answer the question: Do age, sex, and task difficulty act as predictors of social con formity in a laboratory visual perceptual judgment task? 86 87 This experiment was essentially a modified repli cation of the experiment by Klein (1970). The primary difference was the statistical design and analysis employed, and, consequently, the number of subjects utilized. This experiment tested the following hypotheses using social conformity as the dependent variable : Hypothesis I: Informational social influence is greater upon the visual perceptual judgments of older individuals than upon the visual perceptual judgments of younger individuals. Hypothesis II: Informational social influence is greater upon the visual perceptual judgments of females than upon the visual perceptual judgment of males. Hypothesis III. Informational social influence is greater upon the visual perceptual judgments of the subjects as the stimulus of the task becomes more ambiguous (i.e., as the task becomes more difficult). PEOCEDUEËS Twenty-four subjects were used in this experiment. Half of the subjects were young ranging in age from 17 to 23. The other half of the subjects were old, ranging in age from 64 to 76. There was an equal number of males and females in both age groups. (A more complete description of the subjects is given in the General Method section, p. 73.) 88 The Klein (1970) conformity apparatus was utilized for this experiment. (This apparatus has been described in full detail in the General Method section, pp. 74-76.) The visual stimuli were presented on 55 nim slides by a Kodak #860 carousel projector. Method Each subject was first tested alone, then in the conformity situation, then again alone. After a period of one week, he was again tested in the alone situation. (This procedure and the reasons for employing it have been discussed in the General Method section, pp. 78-82.) Visual task.— The subject was informed that a series of slides would be projected on a screen in front of him. He was told that each slide would show a pair of white discs against a black background. He was instructed that his task was to judge which of the two discs was the largest, and to indicate his choice by turning on his left switch if he felt the disc on the left was largest or the right switch if he felt the disc on the right was largest. When performing in the conformity situation he was reminded to respond in the response position he chose at random (all Ss responded in the 4th position due to the deception employed). A series of 56 trials (slide presentations) fol lowed. Twenty-four were critical trials, occurring 89 randomly, on which the judgment of three peers displayed a unanimous selection of the wrong member pair (conformity situation). The remaining 12 trials were innocuous where peer judgments were accurate. The innocuous trials were included in order to maintain the subject's acceptance of the genuineness of the apparatus and the situation. The critical trials were of low task difficulty, medium task difficulty and high task difficulty. There were eight critical trials representing each of the three levels of task difficulty, and there were four innocuous trials representing each of the levels of task difficulty. In the low-task difficulty situation the radii of the two discs were in a ratio of 16 to 10; in the medium-task difficulty situation the radii of the two discs were in a ratio of 16 to 14; and in the high-task difficulty situ ation the radii of the two discs were in a ratio of 16 to 15* The conditions of the task of this experiment is presented in Table 89 of the Appendix, p. 298. Directions for the alone situation.— This experi ment is concerned with perceptual judgments. A series of slides will be projected on the screen in front of you. On each slide will be two white discs against a black background. Your task will be to judge which of the two discs, the one on the left or the one on the right, is the largest. If you feel the disc on the left is largest, turn on the left switch; if you feel the disc on the right is the largest, turn on the right switch. After you make your judgments, leave your switch on until 1 tell you to turn it off. There will be no time limit for your judgments, however, they should not take long. When each slide appears, make your judgment and wait for me to tell you to turn off ______your switch. Are there any questions?____________ 90 Directions for the conformity situation.— This portion of the experiment is concerned with per ceptual judgment in groups. Again, a series of slides will he projected on the screen in front of you. On each slide, as before, there will be two white discs against a black background. Again, your task will be to judge which disc is larger, the one on the left, or the one on the right. In this portion of the experiment, you will not only make your own judgments, but you will also be informed of the judgment of the other subjects. You will each be responding in a predetermined order, which will be decided at random. In each of these envelopes, one of which each of you will pick at random, is a card with a response order printed on it (either 1st, 2d, 3d, or 4th). Do not show it to anyone after I see it; just keep it face up in front of you. Ihis card indicates the response position in which you are to make your judgment. Again, if you feel the disc on the left is larger, turn on the left switch; if you feel the disc on the right is larger, turn on the right switch. On the panel in front of you, each of the pairs of bulbs will represent the responses of the sub jects who have selected that response position. Your switches will turn on those two bulbs that correspond to the response position you have selected. You will make your judgment response after you see either the left or right bulb light up in the response position that precedes the one you have selected. Again, after you make your judgment, leave your switch on until I tell you to turn it off. When the next slide appears, wait for your turn to respond and then make your judgment. If, for example, you select the 1st response position, then you will make your response as soon as the slide appears. If you have selected the 3d position you will wait for the first two subjects to judge which of the figures is largest, which will be indicated by either the left or right bulb lighting up in their position. There will be no time limit for your judgments, however, they should not take long. Are there any questions? 91 Data Analysis The "alone score" was the number of times the sub ject responded with an incorrect judgment on the critical trials in the alone situation. The "conformity score" was the number of times the subject agreed with the contrived group consensus on the critical trials in the conformity situation. Tests 1, 2, and 3, described in the General Method, were conducted. Each of these tests were separately accomplished by statistically analyzing the data by a 3x2x2 mixed analysis of variance design. These were two- between, one-within groups designs. There was repeated measurements over levels of task difficulty. That is, each subject was tested under all three levels of task diffi culty. This resulted in testing three main effects and four interaction effects for the test of compliance (test l) and for the tests of private acceptance (test 2 and 3). Having found a significant main effect for task difficulty in tests 1 and 2, Scheffe*s (1959) method of multiple comparisons was employed to find the source of the effect in each test. The results of the comparisons made are presented in the General Results section, p. 239. 92 RESULTS The data were statistically analyzed by the utili zation of three 2x2x5 mixed analysis of variance designs. Complete analysis of variance summary tables are given in Table 1 (compliance— C), Table 2 (private acceptance in the initial session— 1st P.A.) and Table 3 (private acceptance after one week— 2d P.A.). The data relevant to the hypothesis that older individuals are more conforming than younger individuals are given in Tables 4, 5 and 6 and are diagrammatically presented in Figures 1, 2, and 3- Table 4 gives the mean conformity (G) scores of the two age groups, and indicates that the main effect of age was significant (P = 5*225; p < .05). Figure 1 graphically illustrates the difference in mean conformity (0) scores of the two age groups. It can be seen from the figure that mean conformity scores increased with age. The data pro vide support for the hypothesis, that informational social influence is greater upon the visual perceptual judgments of older individuals than upon the visual perceptual judgments of younger individuals (as measured by com pliance). Table 5 gives the mean conformity (1st P.A.) scores of the two age groups. The main effect of age was not significant at the .05 level of significance. Figure 2 95 TABLE ' 1 COMPLETE ANALYSIS OF VARIANCE SUMMARY TABLE FOR COMPLIANCE EXPERIMENT I Source of Variation df 88Q MSQ F Age 1 4 .5 0 0 4 .5 0 0 5.226* Sex 1 .500 .500 .581 Age X Sex 1 .222 .222 .258 Error 20 1 7 .2 2 2 .861 Task Difficulty 2 16.028 8.014 20.317*** Age X T. D. 2 4.083 2.041 5. 176** Sex X T. D. 2 .250 .125 .5 1 7 Age X Sex x T. D. 2 .5 2 8 .264 . 669 Error 40 15 .7 7 7 .5 9 4 *p < .05 **p < .025 ***p < .001 94 TABLE 2 COMPLETE ANALYSIS OF VARIANCE SUMMARY TABLE FOR THE FIRST TEST OF PRIVATE ACCEPTANCE EXPERIMENT I Source of Variation df SSQ MSQ F Age 1 .222 .222 1.081 Sex 1 .056 .055 0 .2 7 0 Age X Sex 1 .056 .055 0.270 Error 20 4.111 .205 Task Difficulty 2 6.861 5.450 15.424* Age X T. D. 2 .861 .450 1.685 Sex X T. D. 2 .028 .015 0.054 Age X Sex x T. D. 2 .028 .015 0.054 Error 40 10.222 .255 p < .001 95 TABLE 3 COMPLETE ANALYSIS OF VARIANCE SUMMARY TABLE FOR THE SECOND TEST OF PRIVATE ACCEPTANCE EXPERIMENT I Source of Variation df _ . _ S SQ MSQ F Age 1 .125 .125 2.568 Sex 1 .125 .125 2.568 Age X Sex 1 .014 .015 0.265 Error 20 1.056 .055 Task Difficulty 2 .561 .181 2.600 Age X T. D. 2 .085 .041 0.600 Sex X T. D. 2 .085 .041 0.600 Age X Sex x T. D. 2 .028 .014 0.200 Error 60 2.778 .069 Mean Conformity Score T Yotmg Old Fig. 1. Mean conformity scores (compliance) of young and old subjects (combining the three levels of task difficulty) for Experiment I. Eange of scores also indicated. TABLE 4 EXPEEIMEET I MEAH C01IF0EMITÏ SCOEES (COMPLIANCE) OP OLD AND YOUNG SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOE MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 1). Age N Mean Score F (from Analysis of Variance) Young 12 1.085 5.226* Old 12 2.585 *p < .05 97 Mean Conformity Score 22 20 18 16 14 12 11 10 Young Old 2. Mean conformity scores (1st private of young and old subjects (combining the three levels of task difficulty) for Experiment I. Range of scores also indicated. Ei acceptance TABLE 5 EXPERIMENT I MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF OLD AND YOUNG SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 2). Age N Mean Score F (from Analysis of Variance) Young Old 12 12 .666 .999 1.081 98 Mean Conformity Score 22 20 18 16 14 12 11 10 Young Old Fig. 5. Mean conformity scores (2nd private acceptance) of young and old subjects (combining the three levels of task difficulty) for Experiment I. Eange of scores also indicated. (TABLE 6 EXPERIMENT I MEAN CONFORMITY SCORES (2nd PRIVATE ACCEPTANCE) OF OLD AND YOUNG SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 3). Age Young Old N 12 IP Mean Score .083 .335 F (from Analysis of Variance 2.368 99 graphically presents the mean conformity (1st P.A.) scores of the two age groups. It is apparent that conformity scores were only slightly higher for the older subjects. The mean conformity (2d P.A.) scores of the two age groups are presented in Table 6. The main effect of age was not found to be significant at the .0 3 level of significance. Figure 3 presents the difference in the mean conformity (2d P.A.) scores of the two age groups. It is evident that conformity scores were not significantly different; the mean scores were only slightly higher for the older subjects. The hypothesis that females would conform more than males was not confirmed by the data pertaining to com pliance, initial private acceptance, or private acceptance after one week. The data relevant of this hypothesis are given in Tables 7, 8, and 9 and are presented graphically in Figures 4, 3, and 6. Tables ?, 8, and 9 present the mean conformity (C, 1st P.A., and 2d P.A. respectively) of the male and female subjects. It is apparent that the main effect of sex was not significant in any of the three analysis of variance analyses. Figures 4, 3, and 6 illustrate the difference in the mean conformity (C, 1st P.A., and 2d P.A. respective ly) scores of the male and female subjects. It is clear from the figures, that the female subjects were only slightly more conforming than the male subjects. 100 Mean Conformity Score 24 23 22 21 20 19 18 17 1 6 13 1 4 15 12 11 10 9 8 7 6 5 4 5 2 1 t Male Female Fig. 4. Mean conformity scores (compliance) of male and female subjects (combining the three levels of task difficulty) for Experiment I. Range of scores also indicated. TABLE 7 EXPERIMENT I MEAN CONFORMITY SCORES (COMPLIANCE) OF MALE AND FEMALE SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF VARIANCE— TABLE l). Sex N Mean Score F (from Analysis of Variance) Male 12 1.585 Female 12 2.083 .380 101 Mean Conformity Score 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Male Female Fig. 3- Mean conformity scores (1st private acceptance) of male and female subjects (combining the three levels of task difficulty) for Experiment I. Range of scores also indicated. TABLE 8 EXFERIMERT I MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF MALE AND FEMALE SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF SEX (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 2). Sex N Mean Score F (from Analysis of Variance Male 12 .750 .270 Female 12 .916 102 Mean Conformity Score 24 25 22 21 20 19 18 17 15 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 Female Male Fig. 6. Mean conformity scores (2d private acceptance) of male and female subjects (combining the three levels of task difficulty) for Experiment I. Range of scores also indicated. TABLE 9 EXPERIMENT I MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OF MALE AND FEMALE SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF SEX (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 5). Sex N Mean Score Male Female 12 12 .085 .355 F (from Analysis of Variance) 2.568 105 The hypothesis, that'informational social influence is greater upon the visual perceptual judgments of the subjects as the task becomes more difficult, was supported by the data pertaining to compliance and the first test of private acceptance. The data germane to this hypothesis are given in Tables 10, 11, and 12 and are illustrated in Figures 7, 8, and 9- Mean conformity (C) scores for the three levels of task difficulty, averaged over both age and sex, are presented in Table 10. It is evident that the main effect of task difficulty is highly significant (F = 20.517i p < .001). Figure 7 illustrates graphically the difference in the mean conformity (C) scores of the subjects tested under the three levels of task difficulty. It is clear from the figure, the higher the level of task difficulty, indicating greater stimulus ambiguity, the higher the con formity (C) scores. Having obtained a significant F value for the main effect of task difficulty (C), the data were analyzed to find the source of the effect. This was accomplished by employing Scheffe*s (1959) method of multiple comparisons. The results of the comparisons made are presented in Table 85 in the General Results section, p. 259. Table 11 gives the mean conformity (1st P.A.) scores for the three levels of task difficulty averaged over both age and sex. It is apparent that the main effect 104 Me^ Conformity Score 8 7 6 5 4 5 2 1 High Task Low Task Medium Task Difficulty Difficulty Difficulty Fig. 7- Mean conformity scores (compliance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment I. Range of scores also indicated. TABLE 10 EXPERIMENT I MEAN CONFORMITY SCORES (COMPLIANCE) OE SUBJECTS EOR THE THREE LEVELS OP TASK DIPPICULTY (COMBINING AGE AND SEX). P VALUE POR MAIN EPPECT OP TASK DIPPICULTY (TAKEN PROM ANALYSIS OP VARIANCE--TABLE 1). Level of Difficulty N Mean Score P (from Analysis of Variance) LTD 24 .125 MTD 24 .458 20.517* HTD 24 1.250 *p < .001 105 8 7 6 Mean Conformity Score 1 Low Task Difficulty Medium Task Difficulty High Task Difficulty Pig. 8. Mean conformity scores (1st private acceptance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment I. Range of scores also indicated. TABLE 11 EXPERIMENT I MEAN CONPORMITY SCORES (1st PRIVATE ACCEPTANCE) OP SUBJECTS POR THE THREE LEVELS OP TASK DIPPICULTY (COMBINING AGE AND SEX). P VALUE POR MAIN EPPECT OP TASK DIPPICULTY (TAKEN PROM ANALYSIS OP VARIANCE— TABLE 2). Level of Difficulty N Mean Score P (from Analysis of Variance) LTD MTD HTD *p < .001 24 24 24 .000 .125 .708 15.424* 106 8 7 6 Mean Conformity Score 5 Medium Task High Task Low Task Difficulty Difficulty Difficulty Fig. 9. Mean conformity scores (2d private ac ceptance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment I. Range of scores also indicated. TABLE 12 EXPERIMENT I MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OF SUBJECTS FOR THE THREE LEVELS OF TASK DIFFICULTY (COMBINING AGE AND SEX). F VALUE FOR MAIN EFFECT OF TASK DIFFICULTY (TAKEN FROM ANALYSIS OF VARIANCE- TABLE 5). Level of Difficulty N Mean Score F (from Analysis of Variance) LTD MTD HTD 24 24 24 .000 .041 .166 2 .6 0 0 107 of task difficulty was significant (F = 15*424, p < .001). The mean conformity (1st P.A.) scores of the subjects tested under the three levels of task difficulty are presented graphically in Figure 8. It is clear from the figure, the higher the level of task difficulty the higher the conformity (1st P.A.) scores. In order to find the source of the significant main effect for task difficulty (1st P.A.), Scheffe*s (1959) method of multiple comparisons was employed. The results of the comparisons made are presented in Table 85 in the General Results section, p. 259* Table 12 gives the mean conformity (2d P.A.) scores for the three levels of task difficulty averaged over both age and sex. It is apparent that the main effect of task difficulty for private acceptance after one week was not significant at the .05 level of significance (F = 2.600, p < .10). The conformity (2d P.A.) scores of the subjects tested under the three levels of task difficulty are graphically presented in Figure 9* It is evident that the conformity scores increased with levels of task diffi culty, however, the difference in the mean scores was not significant. The analysis of the data pertaining to compliance, presented in Table 1, also indicated a significant age by task difficulty interaction effect (F = 5*176, p < .025)* That is, with increases in task difficulty conformity (C) 108 scores increased for both young and old subjects. However, the increase at each level was greater for the older subjects. CHAPTER IV EXPERIMENT II Conformity investigations in the area of perceptual judgments have typically focused on the visual sense modaility (Asch 1951; Kanareff & Lanzetta I960; Kassarjian & Kassarjian 1962; Klein 1970; Luchins 1955; Schonbar 1945; Sherif 1955; Vidulich & Kaiman 1961; and many others). However, there are a few investigations that have con sidered the auditory sense modality in perceptual conformity studies. In the majority of these investigations, the subject's task was to count the number of clicks heard from a metronome (Blake & Brehm 1954; Blake, Helson & Mouton 1956; MeDavid 1959; Olmstead & Blake 1955)* The present conformity experiment also used the metronome click counting task. The apparatus and procedure was different in the present auditory judgment experiment from those in the past. However, the use of this type of auditory judgment task was based on those investigations mentioned above. The present experiment was directed at determining if a situational condition, age, sex and an interaction of these variables have an effect on conformity behavior. More specifically, this experiment attempted to answer the question: Does age, sex, and task difficulty act as 109 110 predictors of social conformity in a laboratory auditory perceptual judgment task? This experiment tested the following hypotheses using social conformity as the dependent variable: Hypothesis I: Informational social influence is greater upon the auditory perceptual judgments of older individuals than upon the auditory perceptual judgment of younger individuals. Hypothesis II: Informational social influence is greater upon the auditory perceptual judgments of females than upon the auditory perceptual judgments of males. Hypothesis III: Informational social influence is greater upon the auditory perceptual judgments of the subjects as the stimulus of the task becomes more ambiguous (i.e., as the task becomes more difficult). PROCEDURES There were twenty-four subjects used in this experiment. Half of the subjects were old, ranging in age from 61 to 75* The other half of the subjects were young, ranging in age from 18 to 25. There were an equal number of males and females in both age groups. (A more complete description of the subjects is given in the General Method section, p. 75- The Modified Mathie (1959) conformity apparatus was employed in this experiment. (This apparatus has been Ill described in greater detail in the General Method section, pp. 76-77*) The stimuli were presented on magnetic tape via Elega #DR-111 C stereo headphones. A Concord Mark III tape recorder was used. Method In the initial session, the subjects were first tested alone, then in a conformity situation, and then again alone. They were retested in the alone situation one week after their initial testing. (This research design and reasons for utilizing it have been discussed in the General Method section, pp. 79-81.) Auditory task.— The subject was instructed that he would be hearing a series of metronome clicks. He was told that his task would be to judge how many clicks he had heard. He was instructed to indicate his judgment by turning over that card which had printed on it the number of clicks he had heard. When performing in the conformity situation, he was reminded to respond to the response position he had chosen at random (all Ss responded in the 4th position due to the deception employed). A series of 56 trials (metronome click presen tations) followed. Twenty-four trials were critical trials, occurring randomly, on which the contrived judg ments of the three peers displayed a unanimous selection of an incorrect number of clicks. There were 12 innocuous 112 trials where peer judgments were accurate. The innocuous trials were included in order to maintain the subjects acceptance of the genuineness of the apparatus and the situation. On the critical trials the incorrect judgments were discrepant from the actual number of clicks by 1, 2, 3, or 4 clicks. The variation of the reported discrepancy was introduced to determine the effect of differences in the magnitude of the reported discrepancy on conformity be havior. It should be noted that this was a tangential issue and not a variable of primary concern. However, the statistical design does include background discrepancy as a variable. The critical trials were of low task difficulty, medium task difficulty and high task difficulty. There were eight critical trials representing each of the three levels of task difficulty, and four innocuous trials representing each of the levels of task difficulty. Under low task difficulty, the rate of presentation of the clicks was 138 per minute. Under medium task difficulty, the clicks were presented at a rate of 184 per minute. Under high task difficulty, the clicks were presented at a rate of 224 per minute. The click discrepancies were balanced over the rates of presentation to prevent con founding, The number of actual clicks per trial were also balanced over the rates of presentation to prevent 115 confounding. The conditions of the task for this experi ment is presented in Table 90 in Appendix, p. 500. Direction for the alone situation.— This experi- ment is concerned with perceptual judgments. You will be hearing a series of metronome clicks. Your task will be to judge how many clicks you hear in each series. After you make your judgment, turn over that card on the rack in front of you which indicates the number of clicks you have heard. The cards are numbered from one to fifty. After you ' make your judgment leave your card turned over, facing me, until I indicate to return it to its normal position. Each series of clicks will begin five seconds after a tone is sounded. There will be no time limit for your judgments, however, they should not take long. Are there any questions? Directions for the conformity situation.— This portion of the experiment is concerned with perceptual judgment in groups. Again, you will be hearing a series of metronome clicks. Again, your task will be to judge how many clicks you heard in each series. Each series of clicks will begin five seconds after a tone is sounded. In this portion of the experiment, you will not only make your own judgments, but you will also be informed of the judgments of the other subjects. You will each be responding in a pre determined order which will be decided at random. In each of these envelopes, one of which each of you will pick at random, is a card with a response order printed on it (either 1st, 2d, 5d, or 4th). Do not show it to anyone after I see it; just keep it face up in front of you. This card indicates the response position in which you are to make your judgment. On the table in front of me you can see three racks of cards numbered from one to three, each containing cards numbered from one to fifty. Each of these racks will represent the responses of the subjects who have selected that response position. You will make your judgment after you see the card I turn over for the subject who selected the response position that preceeds your response position. Since no one follows the person who has selected the fourth position, his response need not be shown in order to save time. Again, after you make your judgment, leave your card turned over until I indicate to turn it back. When the 114 next series of clicks has been completed, wait for your turn to respond and then make your judgments. If, for example, you have selected the first response position, then you would make your response as soon as the series of clicks was completed. If you have selected the 3d position, you will wait for the first two subjects to make their judgments, which will be indicated by the cards I turn over on the first two racks in front of me, then you will make your judgment. There will be no time limit for your judgments, how ever, they should not take long. Are there any questions? Data Analysis The "alone score" was the frequency the subject responded with an incorrect answer on the critical trials (in the direction of what the contrived report was in the conformity situation) when tested alone. The "conformity score" was the frequency the subject responded with an answer identical with, or in the direction of, the con trived group consensus on the critical trials in the conformity situation. Tests 1, 2, and 5, described in the General Method section, were conducted. Each of these tests were separately accomplished by statistically analyzing the data by 2x2x3x4 mixed analysis of variance designs. These were two-between, two-within groups designs. There was repeated measurements over levels of task difficulty and over the number of discrepant clicks reported by the contrived group. This resulted in testing four main effects and 11 interaction effects for the test of 115 compliance (test l) and for the two tests of private ac ceptance (tests 2 and 3). Having found a significant main effect for task difficulty in tests 1, 2, and 3, Scheffe's (1959) method of multiple comparisons was employed to determine the source of the effect in each case. The results of the comparisons made are presented in the General Results section, p. 239. RESULTS The data were statistically analyzed hy using three 2x2x3x^ mixed analysis of variance designs. Complete analysis of variance summary tables are given in Table 13 (compliance— C), Table 14 (initial test of private ac ceptance— 1st P.A.) and Table 15 (private acceptance after one week-2d P.A.). The data pertaining to the hypothesis, that older individuals are more susceptible to informational social influence in making their auditory judgments, are given in Tables 16, 1?, and 18 and are diagrammatically presented in Figures 10, 11, and 12. The mean conformity (C) scores of the two age groups are presented in Table 16. The main effect of age was not significant (F = 4.093, p < .10). Figure 10 graphically illustrates the difference in mean conformity (C) scores of the two age groups. It is clear from the figure, that COMPLETE TABLE 15 ANALYSIS OF VARIANCE SUMMARY TABLE FOR COMPLIANCE EXPERIMENT II 116 Source of Variation df 88Q MSQ F Age 1 1.253 1.253 4.093* Sex 1 .281 .281 0.918 Age X Sex 1 .005 .005 0.011 Error 20 6.124 .506 Task Difficulty 2 7.425 3.711 17.996** Age X T. D. 2 1.256 .628 3.047* Sex X T. D. 2 .062 .051 0.151 Age X Sex x T. D. 2 .006 .005 0.016 Error 40 8.249 .206 Discrepant Report 5 20.545 6.781 25.121** Age X D. R. 5 1.954 .651 2.221* Sex X D. R. 5 .927 .309 1.053 Age X Sex x D. R. 3 .260 .086 0.296 Error 60 17.596 .293 T. D. X D. R. 6 7.187 1.197 7.258** Age X T. D. X D. R. 6 2.076 .546 2.091* Sex X T. D. X D. R. 6 .104 .017 0.104 Age X Sex x T. D. X D. R. 6 .457 .072 0.440 Error 120 19.858 .165 *p < .10 **p < .001 ■ COMPLETE FOR THE 117 TABLE 11 ANALYSIS OF VARIANCE SUMMARY TABLE FIRST TEST OF PRIVATE ACCEPTANCE EXPERIMENT II Source of Variation df SSQ MSQ. F Age 1 .281 .281 1.680 Sex 1 .051 .051 0.186 Age X Sex 1 .005 .005 0.020 Error 20 3.347 .167 Task Difficulty 2 5.540 1.670 15.287** Age X T. D. 2 .457 .218 1.740 Sex X T. D. 2 .020 .010 0.082 Age X Sex x T. D. 2 .006 .005 0.027 Error 40 5.027 .125 Discrepant Report 3 3.677 1.225 9-540** Age X D. R. 3 .204 .068 0.531 Sex X D. R. 3 .121 .040 0.515 Age X Sex x D. R. 3 .058 .012 0.099 Error 60 7.708 .128 T. D. X D. R. 6 2.020 ^ .336 2.566* Age X T. D. X D. R 6 .568 .061 0.467 Sex X T. D. X D. R 6 .284 .047 0.561 Age X Sex x T. D. X D. R. 6 .076 .012 0.097 Error 120 15.748 .151 *p < .025 **p < .001 COMPLETE FOR THE 118 TABLE 15 ANALYSIS OF VARIANCE SUMMARY TABLE SECOND TEST OF PRIVATE ACCEPTANCE EXPERIMENT II Source of Variation df 8SQ MSQ F Age 1 .015 .015 0.217 Sex 1 .015 .015 0.217 Age X Sex 1 .015 .015 0.217 Error 20 1.277 .065 Task Difficulty 2 .256 .128 3.305** Age X T. D. 2 .006 .005 0.089 Sex X T. D. 2 .006 .005 0.089 Age X Sex x T. D. 2 .006 .005 0.089 Error 40 1.555 .058 Discrepant Report 3 .180 .060 1.857 Age X D. R. 3 .015 .004 0.142 Sex X D. R. 3 .180 .060 1.857 Age X Sex x D. R. 3 .015 .004 0.142 Error 60 1.944 .052 T. D. X D. R. 6 .151 .021 0.742 Age X T. D. X D. E 6 .048 .008 0.273 Sex X T. D. X D. E 6 .581 .065 2.148* Age X Sex x T. D. X D. R. 6 .048 .008 0.273 Error 120 3.555 .029 *p < .10 **p < .05 119 Mean Conformity Score 24- 25 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 Old Young Fig. 10. Mean conformity scores (compliance) of young and old subjects (combining the three levels of task difficulty) for Experiment II. Range of scores also indicated. TABLE 16 EXPERIMENT II MEAN CONFORMITY SCORES (COMPLIANCE) OF OLD AND YOUNG SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 15). Age N Mean Score F (from Analysis of Variance) Young 12 2.000 4.095* Old 12 5.585 *p < .10 120 Mean Conformity Score 24 23 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 Young Old Fig. 11. Mean conformity scores (1st private acceptance; of young and old subjects (combining the three levels of task difficulty) for Experiment II. Range of scores also indicated. TABLE 17 EXPERIMENT II MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF OLD AND YOUNG SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 14). Age N Mean Score F (from Analysis of Variance) Young 12 1.083 1.680 Old 12 1.833 121 Mean Conformity Score 24 23 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 Old Young Fig. 12. Mean conformity scores (2d private acceptance; of young and old subjects (combining the three levels of task difficulty) for Experiment II. Range of scores also indicated. TABLE 18 EXPERIMENT II MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OF OLD AND YOUNG SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 15). Age . N Mean Score F (from Analysis of Variance) Young 12 .555 .217 Old 12 .500 122 older subjects were more conforming than the younger subjects. However, this difference was not significant at the .05 level of significance. Tables 1? and 18 give the mean conformity (1st P.A. and 2d P.A. respectively) scores of the two age groups. The main effect of age, in both analyses, was not signi ficant at the .05 level of significance. Figures 11 and 12 graphically present the mean conformity (1st P.A. and 2d P.A. respectively) scores of the two age groups. It is evident that the mean conformity scores (1st P.A. and 2d P.A.) were only slightly higher for the older subjects. The hypothesis that females are more conforming than males was not confirmed by the data. The data rele vant to this hypothesis are given in Tables 19, 20, and 21 and are presented graphically in Figures 15, 14, and 15. Tables 19, 20, and 21 give the mean conformity (C, 1st P.A. and 2d P.A. respectively) scores of the male and female subjects. The main effect of sex was not signifi cant in any of the three analysis of variance analyses. Figures 15, 14, and 15 diagrammatically illustrate the mean conformity (C, 1st P.A. and 2d P.A. respectively) scores of the male and female subjects. It is clear from each of the figures, that it was the male subjects who were slightly more conforming. The hypothesis, that the more difficult the task the more likely the subject is to be susceptible to Mean Conformity Score 2 4 23 22 2 1 2 0 19 1 8 17 1 6 15 1 4 13 1 2 1 1 1 0 9 8 7 6 5 4 5 2 1 — Male Female Fig. 15. Mean conformity scores (compliance) of male and female subjects (combining the three levels of task difficulty) for Experiment II. Range of scores also indicated. TABLE 19 EXPERIMENT II MEAN CONFORMITY SCORES (COMPLIANCE) OF MALE AND FEMALE SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF SEX (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 1$). Sex N Mean Score F (from Analysis of Variance) Male Female 12 12 3.166 2.416 .918 124 Mean Conformity Score 24 25 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 Male Female Fig. 14. Mean conformity scores (1st private acceptance; of male and female subjects (combining the three levels of task difficulty) for Experiment II. Range of scores also indicated. TABLE 20 EXPERIMENT II MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF MALE AND FEMALE SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF SEX (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 14). Sex N Mean Score F (from Analysis of Variance) Male Female 12 12 1.585 1.553 .186 125 Mean Conformity Score 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Male Female Fig. 1 3. Mean conformity scores (2d private acceptance; of male and female subjects (combining the three levels of task difficulty) for Experiment II. Range of scores also indicated. TABLE 21 EXPERIMENT II MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OF MALE AND FEMALE SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF SEX (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 13). Sex N Mean Score F (from Analysis of Variance) Male Female 12 12 .300 553 217 126 informational social influence, was supported by the data pertaining to compliance and to both tests for private acceptance. The data germane to this hypothesis are given in Tables 22, 23, and 24 and are illustrated in Figures 16, 17, and 18. Mean conformity (C) scores for the three levels of task difficulty, averaged over both age and sex, are pre sented in Table 22. It is evident that the main effect of task difficulty is highly significant (F = 17-996, p < .001). Figure 16 illustrates the difference in the mean conformity (C) scores of the subjects tested under the three levels of task difficulty. It is clear from the figure, the higher the level of task difficulty, indicating greater stimulus ambiguity, the higher the conformity (G) scores. Having obtained a significant F value for the main effect of task difficulty (C), the data were analyzed by Scheffe’s (1959) method of multiple comparisons to find the source of the effect. The results of the comparisons made are presented in Table 83 in the General Results section, p. 239- Table 23 gives the mean conformity (1st P.A.) scores for the three levels of task difficulty averaged over both age and sex. It is apparent that the main effect of task difficulty is significant (F = 13-287, p < .001). The conformity scores (1st P.A.) of the subjects tested 127 8 7 6 Mean Conformity Score Low Task Medium Task High Task Difficulty Difficulty Difficulty Fig. 16. Mean conformity scores (compliance of subjects for the three levels of task difficulty (combining age and sex) for Experiment II. Range of scores also indicated. TABLE 22 EXPERIMENT II MEAN CONFORMITY SCORES (COMPLIANCE) OF SUBJECTS FOR THE THREE LEVELS OF TASK DIFFICULTY (COMBINING AGE AND SEX). F VALUE FOR MAIN EFFECT OF TASK DIFFICULTY (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 15). Level of Difficulty N Mean Score F (from Analysis of Variance) LTD 24 .230 MTD 24 .750 17.996* HTD 24 1.791 *p < . 001 12 8 8 7 Mean Conformity Score Low Task Medium Task High Task Difficulty Difficulty Difficulty Fig. 17. Mean conformity scores (1st private acceptance; of subjects for the three levels of task difficulty (combining age and sex) for Experiment II. Range of scores also indicated. TABLE 23 EXPERIMENT II MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF SUBJECT FOR THE THREE LEVELS OF TASK DIFFICULTY (COMBINING AGE AND SEX). F VALUE FOR MAIN EFFECT OF TASK DIFFICULTY (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 14). Level of Difficulty N Mean Score F (from Analysis of Variance) LTD 24 .085 MTD 24 .291 15.287* HTD 24 1.085 *p < .001 129 8 7 6 Mean Conformity Score Medium Task High Task Low Task Difficulty Difficulty Difficulty Fig. 18. Mean conformity scores (2d private acceptance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment II. Range of scores also indicated. TABLE 24 EXPERIMENT II MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OF SUBJECTS FOR THE THREE LEVELS OF TASK DIFFICULTY (COMBINING AGE AND SEX). F VALUE FOR MAIN EFFECT OF TASK DIFFICULTY (TAKEN FROM ANALYSIS OF VARIANCE- TABLE 15). Level of Difficulty N Mean Score F (from Analysis of Variance) LTD 24 .000 MTD 24 .125 3.305* HTD 24 .291 *p < .05 150 under the three levels of task difficulty are presented graphically in Figure 17* It is apparent from the figure, the higher the level of task difficulty, the higher the conformity scores. In order to determine the source of the signifi cant main effect of task difficulty (1st P.A.), Scheffe's (1959) method of multiple comparisons was employed. The results of the comparisons made are presented in Table 85 in the General Results section, p. 259• Table 24 presents the mean conformity scores (2d P.A.) for the three levels of task difficulty averaged over both age and sex. The main effect of task difficulty for private acceptance after one week was significant (F = 5.505, p < .05). The difference in the mean con formity (2d P.A.) scores of the subjects tested under the three levels of task difficulty is illustrated diagram matical ly in Figure 18. It is apparent that as the difficulty of the task increases, conformity scores increase significantly. Scheffe's (1959) method of multiple com parisons was employed to find the source of the significant main effect for task difficulty (2d P.A.). The results of the comparisons made are presented in Table 85 in the General Results section, p. 259. The main effect of "discrepancy of contrived back ground reports" was found to be significant in the test for 151 compliance (F = 25.121, p < .001— Table 15), and for the first test of private acceptance (F = 9.5^0, p < .001— Table 14). That is, the closer the contrived report to the actual number of clicks sounded, the greater the conformity to that contrived report. The analysis of the data pertaining to compliance. Table 15, and to the first test of private acceptance. Table 14, also indicated a significant task difficulty by discrepancy of contrived background report interaction effect (F = 7.258, p < .001; F = 2.58, p < .025 respec tively) . That is, with increases in task difficulty conformity (C, and 1st P.A.) scores increased for all four discrepant background reports, however, the increase in conformity at each level of task difficulty was greater for the background reports closest to the actual number of clicks sounded. CHAPTER V EXPERIMENT III There have been a number of studies in the past that have investigated auditory detection. However, there appears to be a dearth of conformity experiments that have employed an auditory detection task, especially in the way it will be approached in this experiment. A study by Kanareff and Lanzetta (1958) is, perhaps, the only con formity study that has employed an auditory task that approaches the one that was utilized in this experiment. The task they used was an auditory pitch discrimination task. They found a high degree of imitation of partner choices using pairs of recorded tones and requiring the subject to judge whether the second tone was higher or lower in pitch. The present experiment focused on investigating conformity using an auditory detection task. The idea for the basic task was taken from Broadbent (1985). However, the task of the present experiment was a variation of the task used in the Broadbent experiment. The present investigation was directed at determining if age, sex, and task difficulty act as pre dictors of social conformity in a laboratory auditory detection task. 152 155 Broadbent's (1965) investigation demonstrated that performance on an auditory detection task decreased as attention was diverted away from the stimulus source. It was shown that this decrement was reflected in d'— that is, a reduction of the effective signal strength of the stimu lus. The situation of the experiment called for the subject to attend to a signal in noise on one ear and to a series of digits on the other ear. The results indicated that the subjects* ability to detect the signal dropped from the control to the divided attention condition. This drop was not greater for the older subjects of his experiment. Therefore, older subjects are not at a dis advantage on this type of divided attention task. This experiment tested the following hypotheses using social conformity as the dependent variable : Hypothesis I: Informational social influence is greater upon the auditory detection performance of older individuals than upon the auditory detection performance of younger individuals. Hypothesis II: Informational social influence is greater upon the auditory detection performance of females than upon the auditory detection performance of males. Hypothesis III: Informational social influence is greater upon the auditory detections of the subjects as the stimulus of the task becomes more ambiguous (i.e., as the task becomes more difficult). 154 PROCEDURES There were twenty-four subjects used in this experiment. Twelve of the subjects were old, ranging in age from 50 to 74. Twelve of the subjects were young, ranging in age from 17 to 22. There were an equal number of males and females in both age groups. (A more complete description of the subjects is given in the General Method section, p. 75.). The Klein (1970) apparatus was utilized in this experiment for measuring conformity. (This apparatus has been described in detail in the General Method section, pp. 74-76.) A tape recording was made in order to present the stimulus material to the subjects. The tape recording on the right channel consisted of an audiometric reproduction of the pure tone at the test frequency and band limited random white noise. The tape recording on the left channel consisted of a series of digits. The following equipment was employed to reproduce the stimuli on the right channel of the tape: a Tektronix type 162 waveform generator; four Tektronix type 161 pulse generators; a Grason Stadler model 901-B noise generator; two Grason Stadler model 829-E electronic switches; a General Radio model 1515 oscilator; a Tektronix type 502-A oscilloscope; a mixer; an Ampex model AG 440 tape recorder. A diagram of the set-up used 155 in making the tape is presented in Figure 67 of the Appendix, p. 299* The subjects each had a pair of Elega model DR-1110 stereo headphones through which they received the stimuli. Method The subjects on the first testing were first tested alone, then in the conformity situation and then again alone. They were tested alone an additional time one week after the initial testing. (This research design and the reasons for employing it have been discussed in the General Method section, pp. 78-82.) Auditory detection task.— On each trial, the subject received a 4-second burst of limited band random white noise in his right ear. On half of the trials, a 1000 Hz tone was also presented. On the signal plus noise trials, the tone began one second after the noise onset and ended one second before the noise offset. The signal to noise ratio on the trials where the tone was present was +5dB. That is, the tone was 5dB more intense. The subjects were told that the likelihood of a trial being noise alone or signal plus noise is equal, but that these would occur randomly. The subject was instructed that he would be pre sented a series of trials in which either noise alone or a pure tone signal plus noise would occur in his right ear. 156 He was told his task would he to determine whether or not a signal was present. He was instructed that if he felt that there was a signal present he was to turn on his right switch and if he felt there was not a signal present he was to turn on his left switch. The subject was also instructed that on certain trials he would be hearing digits spoken in his left ear. He was told to ignore the digits on some trials and to pay attention to them on others. When responding in the conformity situation the subject was reminded to respond in the response position he had chosen at random (all subjects responded in the 4th position due to the deception employed.) A series of 56 trials followed. Twenty-four of the trials were critical trials, occurring randomly, on which the contrived consensus of the three peers was inaccurate. The other 12 trials were innocuous where peer responses were unanimously correct. The innocuous trials were included in order to maintain the subject's acceptance of the genuineness of the situation and apparatus. There was an equal number of critical trials with noise alone and with signal plus noise. There was also an equal number of innocuous trials with noise alone and with signal plus noise. There were eight critical trials at each of three levels of task difficulty, and four innocuous trials at each of the three levels of task difficulty. Under high 157 task difficulty, during the detection interval, the subject heard a series of digits in his left ear and he was told to pay attention to these digits. Under medium task difficulty, during the detection interval, the subject also heard the series of digits in his left ear but was told to ignore them. Under low task difficulty, during the detection interval, nothing was presented in the sub ject's left ear. The information in the left ear increases the difficulty of detecting the signal in the right ear, especially when the subject is instructed to pay at tention to that information. The conditions of the task for this experiment is presented in Table 91 of the Appendix, p. 501. Directions for the alone situation.— This experi- ment is concerned with auditory detections. Your task will be to respond on a number of trials as to whether or not you hear a signal in addition to noise on your right ear, or just noise alone. If you feel a signal was present on a trial turn on your right switch. If you feel only noise was present, turn on your left switch. Leave your switches on until I indicate to turn them off. In your left ear, you will sometimes hear a series of digits. If I indicate to ignore them or to pay attention to them follow my in structions. After you make your judgments as to the presence of a signal, write down the digits you heard. There is an equal number of trials with signals present and signals absent, however, they will be randomly distributed among the trials. There will be no time limit for your judgments, however, they should not take long. Are there any questions? Directions for the conformity situation.— This por tion of the experiment is concerned with auditory 158 detection in groups. Again, your task will be to respond on a number of trials as to whether or not you hear a signal, in addition to noise, or just noise alone in your right ear. In your left ear, as before, there will sometimes be a series of digits. If I indicate to ignore them, or pay attention to them, follow my directions. In this portion of the experiment you will not only make your own responses, but you will also be informed of the responses of the other subjects. You will each be responding in a predetermined order which will be decided at random. In each of these envelopes, one of which each of you will pick at random, is a card with a response order printed on it (either 1st, 2d, 5d, or 4th). Do not show it to anyone after I see it; just keep it face up in front of you. This card indicates the response position in which you are to make your responses. Again, if you feel that there was a signal present on a trial, turn on your right switch. If you feel there was only noise present on a trial, turn on your left switch. On the panel in front of you, each of the pairs of bulbs will represent the responses of the subjects who have selected that response position. Your two switches will turn on those two bulbs that correspond to the response position you have selected. You will make your response after you see either the left or right bulb light up in the response position that preceeds the one you have selected. If, for example, you select the first response position, you would make your response as soon as the trial interval was over. If you select the third position, you will wait for the first two subjects to respond as to the presence of a signal, as indicated by either the left or right bulbs lighting up in both the first and second positions on your panel. Again, there will be an equal number of trials with and without a signal present, and they will be randomly distributed in the series. After you make your responses leave your switch on until I tell you to turn it off. On the trials where digits are presented, write the digits you heard down after making your judgment as to the presence of a signal. There will be no time limit for your responses, however, they should not take long. Are there any questions? 159 Data Analysis The "alone score" was the number of times the subject made an incorrect response on the critical trials in the alone situation. The "conformity score" was the number of times the subject agreed with the contrived group consensus on the critical trials of the conformity situation. Tests 1, 2, and 5, described in the General Method section, were conducted. Each of these tests was accomplished separately by statistically analyzing the data by 2x2x5 mixed analysis of variance designs. These were two-between, one-within groups designs. This resulted in testing three main effects and four interaction effects for the test of compliance (test 1) and for the tests of private acceptance (test 2 and 5). There was repeated measurements over the levels of task difficulty. That is, each subject was tested under all three levels of task difficulty. Having found a significant main effect for task difficulty in tests 1, 2, and 5, Scheffe's (1959) method of multiple comparisons was employed to find the source of the effect in each case. The results of the comparisons made is presented in the General Results section, p. 259. 14-0 EESULTS Three 2x2x3 mixed analysis of variance designs were used to analyze the data statistically. Complete analysis of variance summary tables are given in Table 23 (compliance— C), Table 26 (initial test of private ac ceptance— 1st P.A.) and Table 27 (private acceptance after one week— 2d P.A.). The data pertaining to the hypothesis that older individuals are more conforming than younger individuals are given in Tables 28, 29, and 30 and are graphically illustrated in Figures 19, 20, and 21. Table 28 presents the mean conformity scores (0) of the two age groups, and indicates that the main effect of age was significant (F = 3-599, P < -03). The difference in mean conformity (C) of the two age groups is graphically illustrated in Figure 19- It can be seen from the figure, that mean conformity scores increased with age. The data provide support for the hypothesis that informational social influence is greater upon the auditory detections of older individuals than upon the auditory detections of younger individuals (as measured by compliance). Table 29 gives the mean conformity (1st P.A.) scores of the two age groups. The main effect of age was not significant at the .03 level of significance (F = 4.036, p < .10). Although the main effect was not 141 TABLE 23 COMPLETE ANALYSIS OP VARIANCE SUMMAEI TABLE FOR COMPLIANCE EXPERIMENT III Source of Variation df SSQ MSQ F Age 1 10.888 10.888 5.599 Sex 1 .222 .222 0.110 Age X Sex 1 .033 .035 0.027 Error 20 40.355 2.016 Task Difficulty 2 56.999 18.499 21.346 Age X T. D, 2 4.111 2.033 2.371 Sex X T. D. 2 .111 .033 0.064 Age X Sex x T. D. 2 .111 .033 0.064 Error 40 34.666 .866 < .03 = * p < .001 142 TABLE 26 COMPLETE ANALYSIS OP VARIANCE SUMMARY TABLE FOR THE FIRST TEST OF PRIVATE ACCEPTANCE EXPERIMENT III Source of Variation_____________df______ SSQ______ MSQ________F_ Age 1 5.013 3.012 4.036 Sex 1 .123 .123 0.101 Age X Sex 1 .123 .123 0.101 Error 20 24.722 1.236 Task Difficulty 2 22.027 11.013 13.042 Age X T. D. 2 5.694 1.847 2.187 Sex X T. D. 2 .230 .123 0.148 Age X Sex x T. D. 2 .230 .124 0.148 Error 40 55.777 .844 *p < ,10 **p < .001 145 TABLE 27 COMPLETE ANALYSIS OP VARIANCE SUMMARY TABLE FOR THE SECOND TEST OF PRIVATE ACCEPTANCE EXPERIMENT III Source of Variation df __ __ MSQ. F Age 1 .123 .123 0.217 Sex 1 .123 .123 0.217 Age X Sex 1 .123 .123 0.217 Error 20 11.499 .575 Task Difficulty- 2 1.555 . 666 4.000* Age X T. D. 2 .000 .000 0.000 Sex X T, D. 2 .000 .000 0.000 Age X Sex x T. D. 2 .000 .000 0.000 Error 40 5.666 .166 144 Mean Conformity Score 24 23 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 Old Young Fig. 19. Mean conformity scores (compliance) of young and old subjects (combining the three levels of task difficulty) for Experiment III. Range of scores also indicated. TABLE 28 EXPERIMENT III MEAN CONFORMITY SCORES (COMPLIANCE) OF OLD AND YOUNG SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 25). Apie N Moan Score F (from Analysis of Variance) Young 12 2.583 5.599* Old 12 4.9I6 *p < .05 145 Mean Conformity Score 24 25 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 Old Young 20. Mean conformity scores (1st private of young and old subjects (combining the Fi acceptance three levels of task difficulty) for’Experiment III. Range of scores also indicated. TABLE 29 EXPERIMENT III MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF OLD AND YOUNG SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 26). Age N Mean Score F (from Analysis of Variance Young 12 1.416 4.056* Old 12 3.000 *p < .10 146 Mean Conformity Score 24 25 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 2 Old Young Fig. 21. Mean conformity scores (2d private acceptance/ of young and old subjects (combining the three levels of task difficulty) for Experiment III. Range of scores also indicated. TABLE 50 EXPERIMENT III MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OF OLD AND YOUNG SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 27). Age N Mean Score F (from Analysis of Variance) Young 12 .500 .217 Old 12 .750 147 significant at the adopted significance level, it can he seen in Figure 20 that the mean conformity score was substantially higher for the older group. Table 50 presents the mean conformity (2d P.A.) scores of the two age groups and indicates that the main effect of age was not significant at the .05 level of significance. Figure 21 presents diagrammatically the mean conformity (2d P.A.) scores of the two age groups. It is apparent from the figure that conformity scores were only slightly higher for the older subjects. The hypothesis that females are more conforming than males was not confirmed by the data pertaining to compliance, the first test of private acceptance, or private acceptance after one week. The data relevant to this hypothesis are given in Tables 51, 52, and 55 and are presented graphically in Figures 22, 25, and 24. The mean conformity (C, 1st P.A., and 2d P.A.) scores of the male and female subjects are given in Tables 51, 52, and 55 respectively. It can be seen from each of the tables, that the main effect of sex was not significant in each of the three analysis of variance analyses. Figures 22, 25, and 24 present graphically the mean conformity scores (C, 1st P.A., and 2d P.A. respective ly) of the male and female subjects. It is evident from Figure 25, that in the first test of private acceptance the female subjects were only slightly more conforming than 148 Mean Conformity Score 24 23 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 Female Male Fig. 22. Mean conformity scores (compliance) of male and female subjects (combining the three levels of task difficulty) for Experiment III. Range of scores also indicated. TABLE 31 EXPERIMENT III MEAN CONFORMITY SCORES (COMPLIANCE) OF MALE AND FEMALE SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF SEX (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 23). Sex N Mean Score F (from Analysis of Variance) Male Female 12 12 5.916 5.583 .110 149 Mean Conformity Score 24 23 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 ± ± Male Female Fig. 25. Mean conformity scores (1st private acceptance; of male and female subjects (combining the three levels of task difficulty) for Experiment III. Range of scores also indicated. TABLE 52 EXPERIMENT III MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF MALE AND FEMALE SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF SEX (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 26). Sex N Mean Score F (from Analysis of Variance) Male 12 2.083 .101 Female 12 2.355 150 Mean Conformity Score 22 21 20 18 16 14 12 10 Male Female Fig. 24. Mean conformity scores (2d private acceptance; of male and female subjects (combining the three levels of task difficulty) for Experiment III. Range of scores also indicated. TABLE 53 EXPERIMENT III MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OF MALE AND FEMALE SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF SEX (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 27). Sex N Mean Score F (from Analysis of Variance) Male 12 .750 .217 Female 12 .500 151 the male subjects. Figures 22 and 24 indicate that in the case of compliance and private acceptance after one week, it was the male subjects who were slightly more conforming. The hypothesis, that informational social influence is greater upon the auditory detections of the subjects as the task becomes more difficult, was supported by the data pertaining to compliance and to both tests of private ac ceptance. The data relevant to this hypothesis are given in Tables 54, 55» and 56 and are illustrated in Figures 25, 25, and 27- Mean conformity (C) scores for the three levels of task difficulty, averaged over both age and sex, and are presented in Table 54. It is clear that the main effect of task difficulty is highly significant (F = 21.546, p < .001). The difference in the mean conformity (G) scores of the subjects tested under the three levels of difficulty is illustrated in Figure 25* It is apparent from the figure that the higher the level of task diffi culty, the higher the conformity (0) scores. Having obtained a significant F value for the main effect of task difficulty (C), the data were analyzed by Scheffe*s (1959) method of multiple comparisons to find the source of the effect. The results of the comparisons made are presented in Table 85 in the General Results section, p. 259. 152 8 7 6 Mean Conformity Score 5 ± Low Task Difficulty Medium Task Difficulty High Task Difficulty Fig. 25- Mean conformity scores (compliance) of subjects for the three levels of task difficulty (com bining age and sex) for Experiment III. Range of scores also indicated. TABLE 54- EXPERIMENT III MEAN CONFORMITY SCORES (COMPLIANCE) OF SUBJECTS FOR THE THREE LEVELS OF TASK DIFFICULTY (COMBINING AGE AND SEX). F VALUE FOR MAIN EFFECT OF TASK DIFFICULTY (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 25). Level of Difficulty N Mean Score (from Analysis of Variance) LTD MTD HTD 'p < .001 24- 24- 24- .4-16 1.166 2.166 21.54-6* 15$ Mean Conformity Score 8 7 6 5 4 3 2 1 Medium Task High Task Low Task Difficulty Difficulty Difficulty Fig. 26. Mean conformity scores (1st private acceptance; of subjects for the three levels of task difficulty (combining age and sex) for Experiment III. Eange of scores also indicated. TABLE 55 EXPERIMENT III MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF SUBJECTS FOR THE THREE LEVELS OF TASK DIFFICULTY (COMBINING AGE AND SEX). F VALUE FOR MAIN EFFECT OF TASK DIFFICULTY (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 26). Level of Difficulty N Mean Score F (from Analysis of Variance) LTD 24 .208 MTD 24 .500 15.042* HTD 24 1.500 *p < .001 154 8 7 6 Mean Conformity Score 4 i Low Task Medium Task High Task Difficulty Difficulty Difficulty Fig. 27. Mean conformity scores (2d private acceptance; of subjects for the three levels of task difficulty (combining age and sex) for Experiment III. Range of scores also indicated. TABLE 36 EXPERIMENT III MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OF SUBJECTS FOR THE THREE LEVELS OF TASK DIFFICULTY (COMBINING AGE AND SEX). F VALUE FOR MAIN ’ EFFECT OF TASK DIFFICULTY (TAKEN FROM ANALYSIS OF VARIANCE- TABLE 2 7). Level of N Mean Score F (from Analysis LTD 24 .041 MTD 24 .208 4.000* HTD 24 .375 < .0 5 155 Table 55 presents the mean conformity (1st P.A.) scores for the three levels of task difficulty averaged over both age and sex. It is apparent that the main effect of task difficulty is significant (P = 13-042, p < .001). The mean conformity scores (1st P.A.) of the subjects tested under the three levels of task difficulty are presented graphically in Figure 26. It is clear from the figure, the higher the level of task difficulty the higher the conformity (1st P.A.) scores. In order to find the source of the significant main effect obtained for task difficulty (1st P.A.), Scheffe * s (1959) method of multiple comparisons was used. The results of the comparisons made are presented in Table 83 in the General Results section, p. 239. Table 36 presents the mean conformity (2d P.A.) for the three levels of task difficulty averaged over both age and sex. The main effect of task difficulty for private acceptance after one week was significant (F = 4.000, p < . 0 5). The difference in the mean conformity scores of the subjects tested under the three levels of task difficulty is illustrated in Figure 27- It is apparent that as the difficulty of the task increases, there is a significant increase in mean conformity (2d P.A.) scores. Scheffe*s (1959) method of multiple com parisons was used to find the source of the significant main effect for task difficulty (2d P.A.). The results of 156 the comparisons made are presented in Table 83 in the General Results section, p. 239• CHAPTER VI EXPERIMENT XV Experimentation in the area of social conformity has generally considered tasks dealing with perceptual judgments (Asch 195'^; Kanareff & Lanzetta 1960; Klein 1970; Schonbar 1945; Sherif 1935; etc.), or attitudes and opinions (Converse & Campbell 1968; Gordon 1952; Kiesler, Zana, & ReSalvo 1966; etc.). In both of these types of tasks, the subject is making a judgment or rating on stimuli presented to him. These tasks do not require the subject to perform any mental manipulations except for the judgment of stimuli. There have been a few investigations that have required the subjects to ’ ’solve problems” (Blake, Helson & Mouton 1956; Coffin 1941). In the study by Blake, Helson, and Mouton (1956) a ’ ’simulated group” technique was used to investigate conformity in solving arithmetic problems without paper and pencil. They found that the more difficult the problems were, the more the subjects were influenced by the solutions of others. The present investigation was based on the study by Blake, Helson and Mouton (1956) in that it also utilized arithmetic problem solving as the task. 157 158 The present investigation was directed at deter mining if age, sex, and task difficulty act as determinants of social conformity in a laboratory arithmetic problem solving task. This experiment tested the following hypotheses using social conformity as the dependent variable : Hypothesis I. Informational social influence is greater upon the arithmetic problem solutions of older individuals than upon the arithmetic problem solutions of younger individuals. Hypothesis II: Informational social influence is greater upon the arithmetic problem solutions of females than upon the arithmetic problem solutions of males. Hypothesis III: Informational social influence is greater upon the arithmetic problem solutions of the sub jects as the task becomes more difficult. PROCEDURES There were twenty-four subjects used in this experiment. Ti^relve of the subjects were young, ranging in age from 18 to 24. Twelve of the subjects were old, ranging in age from 62 to 78. There were an equal number of males and females in both age groups. (A more complete description of the subjects is given in the General Method section, p. 75.) 159 Two additional groups of twenty-four individuals, selected from the same population as the general subject sample, were used to assess the difficulty of arithmetic problems. One of these groups was comprised of older individuals ranging in age from 61 to 75- The other group was young, ranging in age from 17 to 25. Both of these standardizing groups had an equal number of males and females. The conformity apparatus that was used in this experiment was based on the principle of the Klein (1970) and Mathie (1959) apparatus. (This apparatus has been described in greater detail in the General Method section, p. 77.) Method In the initial session, the subjects were first tested alone, then in the conformity situation, and then again alone. They were retested in the alone situation one week after their first testing. (This research design and the reasons for employing it have been discussed in the General Method section, pp. 79-81.) Problem-solving task.— The subject was instructed that he would be given a series of arithmetic problems to solve without the use of paper and pencil. He was told to indicate his solution by writing it on his "magic slate.” He was told to then place the slate in the rack in front 160 of him with the solution facing the experimenter. When performing in the conformity situation, he was reminded to respond in the response position he had chosen at random. (All Ss responded in the 4th position due to the deception employed.) A series of 56 trials of arithmetic problems in volving one operation (addition, subtraction, multipli cation or division) followed. Each problem was presented for five seconds. Twenty-four trials were critical trials, occurring randomly, on which the solutions of the three peers were unanimously incorrect. There were 12 innocuous trials where the contrived solutions were accurate. The innocuous trials were included in order to maintain the subject's acceptance of the genuineness of the apparatus and the situation. The critical trials were of low task difficulty, medium task difficulty, and high task difficulty. There were eight critical trials at each level of difficulty (2 addition, 2 subtraction, 2 multiplication, 2 division). There were four innocuous trials at each level of task difficulty. "Difficulty” of the arithmetic problems was assessed by the performance of the two standardizing groups of 24 subjects before the actual experiment began. Prob lems answered correctly by more than 75 percent of the subjects of a standardizing group were categorized as low task difficulty problems for that age group. Problems 161 answered correctly by more than 75 percent of the subjects of a standardizing group were categorized as low task difficulty problems for that age group. Problems answered correctly by 25 to 75 percent of the subjects of a standardizing group were classified as medium task diffi culty problems for that age group. Problems answered correctly by less than 25 percent of the subjects of a standardizing group were categorized as high task difficulty problems for that age group. By using both age groups in standardizing the difficulty of the problems used, the task did not put either the young or the old subjects at a disadvantage. On the critical trials, the incorrect problem solutions were discrepant from the correct solution by a small or large amount. Those incorrect problem solutions classified as mildly discrepant were those that were in the error range of the standardizing groups. Those classified as highly discrepant were those outside the error range of the standardizing group. The variation of the discrepant background report of the contrived group was introduced to determine the degree to which these differences affect conformity behavior. It should be made clear that this issue is an aside. It is not one of the primary concerns of the investigation. However, the statistical design and analysis does include discrepant background report of the contrived group as a variable. 162 (The conditions of the task of this experiment for both young and old subjects, is presented in Tables 92 and 93 in the Appendix, pp. 302-303.) Directions for the alone situation.— Tliis experi ment is concerned with arithmetic problem solving. You will see a series of one operation arithmetic problems (either addition, subtraction, multi plication, or division). Your task will be to solve the problems without paper and pencil. Each problem will appear for five seconds. After you arrive at a solution, write your answer on the "magic slate" in front of you and then place the slate on the rack in front of you with your answer facing me. Leave the slate on the rack until I tell you to take it off. Are there any questions? Directions for the conformity situation.— This portion of the experiment is concerned with arithmetic problem solving in groups. Again you will be seeing a series of one-operation arithmetic problems. Again, your task will be to solve the problems without paper and pencil. Each problem will appear for five seconds. In this portion of the experiment, you will not only make a response indicating your solution, but you will also be informed of the solutions arrived at by the other subjects. You will each be responding in a predetermined order which will be decided at random. In each of these envelopes, one of which each of you will pick at random, is a card with a response order printed on it (either 1st, 2d, 5d, or 4-th). Do not show it to anyone after I see it; just keep it face up in front of you. This card indicates the response position in which you are to display your solution answers. On the table in front of me you can see three empty racks numbered from 1 to 3# The "magic slate" I place on each of these racks will repre sent the responses of the subjects who have selected that response position. You will make your response after you see the "magic slate" placed on the rack in front of me for the subject who has selected the response position that precedes the one you have selected. Since no one follows the person who has selected the fourth position his response need not be shown in order to save time. Again, after ________you display vour answer, leave your slate on its 165 rack until I tell you to take it off. When the next problem appears, wait for your turn to respond and then display your solution answers. You are allowed to change your answer as long as you do it before you display it. Remember, however, you cannot make calculations on the slate. If for example, you have selected the first response position, then you would display your solution answer immediately after the problem is presented. If you have selected the third position, you will wait for the first two sub jects to display their solutions, which will be indicated by the numbers on the slates in front of me in racks one and two, then you will display your solution answers. Are there any questions? Data Analysis The "alone score" was the frequency the subject responded with an incorrect answer on the critical trials (in the direction of what the contrived report of the group was in the conformity situation) when tested alone. The "conformity score" was the frequency the subject responded with an answer identical with or in the direction of the contrived group consensus on the critical trials in the conformity situation. Tests 1, 2, and 5, described in the General Method section, were conducted. Each of these tests was separately accomplished by analyzing statistically the data utilizing a 2x2x5x2 mixed analysis of variance design. There were two-between, two-within groups designs. There was repeated measurements over levels of task difficulty and discrepancy of contrived background reports. This resulted in testing four main effects and 11 interaction effects for the test of compliance (test 1) and for the____ 164- tests of private acceptance (tests 2 and 3)- Having found a significant main effect for task difficulty in tests 1, 2, and 5, Scheffe's (1959) method of multiple comparisons was employed to determine the source of the effect in each case. The results of the comparisons made are presented in the General Results section, p. 259- RESULTS The data were analyzed statistically by the utilization of three 2x2x5x2 mixed analysis of variance designs. Complete analysis of variance summary tables are presented in Table 5 7 (compliance— C), Table 58 (first test of private acceptance— 1st P.A.) and Table 5 9 (private acceptance after one week— 2d P.A.). The data relevant to the hypothesis that older individuals are more conforming than younger individuals are given in Tables 4-0, 4-1, and 4-2 and are diagrammatically presented in Figures 28, 29, and 50. The mean conformity (C) scores for the two age groups, and the F value for the main effect of age, are presented in Table 4-0. The main effect of age was found to be significant (F = 4-.515, p < .05)- Figure 28 illustrates graphically the difference in mean conformity (C) scores of the two age groups. It is clear from the figure that mean conformity scores increased with age. 165 TABLE 57 COMELETE AITALYSIS OE VARIANCE SUmARY TABLE FOR COMPLIANCE EXPERIMENT IV Source of Variation df s s q MSQ F Age 1 12.250 12.250 4. 515* * Sex 1 5 .4 4 4 5.555 2 . 0 0 6 Age X Sex 1 .000 .000 0.000 Error 20 54.277 2.715 Task Difficulty 2 55.555 1 6 . 7 7 7 5 2 . 5 8 6 * * * * Age X T. D. 2 4 . 1 6 6 2 . 0 8 5 4 . 0 2 1 * * Sex X T. D. 2 1.055 .527 1 . 0 1 8 Age X Sex x T. D. 2 . 1 6 6 . 0 8 5 0 . 1 6 0 Error 40 20.722 . 5 1 8 Discrepant Report 1 1 4 . 6 9 4 14.694 19.097* * * * Age X D. R. 1 1 . 5 6 1 1 . 5 6 1 1 . 7 6 9 Sex X D. R. 1 .111 .111 0.144 Age X Sex x D. R. 1 .444 . 4 4 4 0.577 Error 2 0 15.588 . 7 6 9 T. D. X D. R. 2 2.055 1.027 4. 774* * * Age X T. D. X D. R. 2 1 . 5 8 8 . 6 9 4 5. 225* Sex X T. D. X D. R. 2 .055 .027 0 . 1 2 8 Age X Sex x T. D. X D. R. 2 . 8 8 9 .444 2 . 0 6 4 Error 40 8 . 6 1 0 .215 *p < .10 **p < .05 ***p < .025 ****p < .001 166 TABLE 38 COMPLETE ANALYSIS OP VARIANCE SUMMARY TABLE FOR THE FIRST TEST OF PRIVATE ACCEPTANCE EXPERIMENT IV Source of Variation df SSQ MSQ F Age 1 9.506 9.506 6.475*** Sex 1 3.062 3.062 2.086 Age X Sex 1 .340 .340 0.231 Error 20 29.561 1.468 Task Difficulty 2 17.166 8.583 31.055***** Age X T. D. 2 5.722 1.861 6.733***** Sex X T. D, 2 .500 .250 0.904 Age X Sex x T. D. 2 .222 .111 0.402 Error 40 11.055 .276 Discrepant Report 1 10.562 10.562 23.400***** Age X D. R. 1 2.006 2.006 4.446** Sex X D. R. 1 1.175 1.175 2.600 Age X Sex x D. R. 1 .062 .062 0.138 Error 20 9.027 .451 T. D. X D. R. 2 3.166 1.585 7.550**** Age X T. D. X D. R. 2 1.055 .527 2.516* Sex X T. D. X D. R. 2 .055 .027 0.132 Age X Sex x T. D. X D. R. 2 .000 .000 0.000 Error 40 8.388 .209 *p < .10 < .05 ***p < .025 ****p < .005 **** < .001 167 TABLE 59 COMPLETE POP THE ANALYSIS OP VARIANCE SUMMARY TABLE SECOND TEST OP PRIVATE ACCEPTANCE EXPERIMENT IV Source of Variation df 88Q MSQ P Age 1 4.540 4.540 4.240* Sex 1 .562 .562 0.549 Age X Sex 1 .062 .062 0.061 Error 20 20.472 1.025 Task Difficulty 2 6.791 5.595 14.641**** Age X T. D. 2 2.515 1.256 5.419*** Sex X T. D. 2 .291 .145 0.628 Age X Sex x T. D. 2 .125 .062 0.269 Error 40 9.277 .251 Discrepant Report 1 4.540 4.540 12.550*** Age X D. R. 1 .840 .840 2.429 Sex X D. R. 1 .062 .062 0.180 Age X Sex x D. R. 1 .006 .006 0.020 Error 20 6.916 .545 T. D. X D. R. 2 .847 .425 5.765** Age X T. D. X D. R. 2 .515 .256 2.284 Sex X T. D. X D. R. 2 .125 .062 0.555 Age X Sex x T. D. X D. R. 2 .547 .175 1.545 Error 40 4.499 .112 *p < .10 **p < .05 ***p < .005 * * * * p < . 0 0 1 1 6 8 Mean Conformity Score 2 4 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Young Old Fig. 28. Mean conformity scores (compliance) of young and old subjects (combining the three levels of task difficulty) for Experiment IV. Range of scores also indicated. TABLE 40 EXPERIMENT IV MEAN CONFORMITY SCORES (COMPLIANCE) OF OLD AND YOUNG SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 37). Age N Mean Score F (from Analysis of Variance) Young 12 3.666 4.313* Old 12 7.166 < .03 169 Mean Conformity Score 24 25 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 ± Old Young Fig. 29. Mean conformity scores (1st private acceptance) of young and old subjects (combining the three levels of task difficulty) for Experiment IV. Range of scores also indicated. TABLE 41 EXPERIMENT IV MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF OLD AND YOUNG SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 58). Age...... N Mean Score F (from Analysis of Variance Young 12 2.085 6.475* Old 12 5.166 *p < .025 170 Mean Conformity Score 24 25 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Old Young Fig. 50. Mean conformity scores (2d private acceptance; of young and old subjects (combining the three levels of task difficulty) for Experiment IV. Range of scores also indicated. TABLE 42 E X F E m m m iv MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OE OLD AND YOUNG SUBJECTS (COMBINING THE THREE LEVELS OP TASK DIEPICULTY). E VALUE EOR MAIN EEEECT OE AGE (TAKEN EROM ANALYSIS OE VARIANCE— TABLE 39). Age N Mean Score E (from Analysis of Variance) Young 12 1.085 4.240* Old 12 5.166 *p < .10 171 The data offer support to the hypothesis that informational social influence is greater upon the arithmetic problem- solving performance of older individuals than it is upon the arithmetic problem-solving performance of younger individuals (as measured by compliance). Table 4-1 gives the mean conformity (1st P.A. ) scores for the young and old subjects. The main effect of age was significant (P = 6.4-75, P < .023). Figure 29 presents graphically the difference in the mean conformity (is P.A.) scores for both age groups. It is evident that mean conformity scores increased with age. This data lends further support to the hypothesis that the older individual is more conforming (as measured by initial private acceptance). Table 4-2 presents the mean conformity (2d P.A.) scores of the old and young subjects and indicates that the main effect of age was not significant at the .05 level of significance (P = 4.240, p < .10). Figure 50 illustrates the difference in the mean conformity (2d P.A.) scores for the two age groups. It is clear that although this difference did not reach significance, the conformity scores were generally substantially higher for the older subjects. The hypothesis that females are more conforming than males did not receive support from the data per taining to compliance, the first test of private 172 acceptance or private acceptance after one week. The data relevant to this hypothesis are given in Tables 4-5, 44-, and 4-5 and are graphically presented in Figures 51, 52, and 55. The mean conformity (C, 1st P.A., and 2d P.A.) scores of the male and female subjects are presented in Tables 45, 44, and 45 respectively. It is apparent that the main effect of sex was not significant in any of the three analysis of variance analyses. Figures 51, 52, and 55 illustrate the difference in the mean conformity scores (C 1st P.A., 2d P.A. respectively) of the male and female subjects. It is evident from the figures that the female subjects were only slightly more conforming than the male subjects. The hypothesis, that informational social influence is greater upon the arithmetic problem-solving performance of the subjects as the task becomes more difficult, was supported by the data pertaining to compliance and to both tests of private acceptance. The data relevant to this hypothesis are given in Tables 46, 47, and 48 and are diagrammatically presented in Figures 54, 55, and 56. Mean conformity (C) scores for the three levels of task difficulty, averaged over both age and sex, are pre sented in Table 46. It is apparent that the main effect of task difficulty was highly significant (F = 52.586, p < .001). The difference in the mean confirmity (C) 17$ Mean Conformity Score 24 2$ 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 Male Female Fig. 51* Mean conformity scores (compliance) of male and female subjects (combining the three levels of task difficulty) for Experiment IV. Range of scores also indicated. TABLE 45 EXPERIMENT IV MEAN CONFORMITY SCORES (COMPLIANCE) OF MALE AND FEMALE SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF SEX (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 57). Sex N Mean Score F (from Analysis of Variance) Male 12 4.249 2.006 Female 12 6.585 174 Mean Conformity Score 24 25 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 t Male Female Fig. 52. Mean conformity scores (1st private acceptance; of male and female subjects (combining the three levels of task difficulty) for Experiment IV. Range of scores also indicated. TABLE 44 EXPERIMENT IV MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF MALE AND FEMALE SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF SEX (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 58). Sex N Mean Score F (from Analysis of Variance) Male 12 2.749 2.086 Female 12 4.500 175 Mean Conformity Score 24 25 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 ± ± Male Female Fig. 55. Mean conformity scores (2d private acceptance) of male and female subjects (combining the three levels of task difficulty) for Experiment IV. Range of scores also indicated. TABLE 45 EXPERIMENT IV MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OF MALE AND FEMALE SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF SEX (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 59). Sex N Mean Score F (from Analysis of Variance) Male 12 1.750 .549 Female 12 2.500 176 8 7 6 Mean Conformity Score 4 ± Low Task Medium Task High Task Difficulty Difficulty Difficulty Fig. 54. Mean conformity scores (compliance) of subjects for the three levels of task difficulty (com bining age and sex) for Experiment IV. Range of scores also indicated. TABLE 46 EXPERIMENT IV MEAN CONFORMITY SCORES (COMPLIANCE) OF SUBJECTS FOR THE THREE LEVELS OF TASK DIFFICULTY (COMBINING AGE AND SEX). F VALUE FOR MAIN EFFECT OF TASK DIFFICULTY (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 57). Level of Difficulty N Mean Score F (from Analysis of Variance) LTD 24 .750 MTD 24 1.585 52.586* HTD 24 5.085 *p < . 001 177 8 6 Mean Conformity Score ± Low Task Medium Task High Task Difficulty Difficulty Difficulty Pie. 55* Mean conformity scores (1st private acceptance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment IV. Range of scores also indicated. TABLE 47 EXPERIMENT IV MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF SUBJECTS FOR THE THREE LEVELS OF TASK DIFFICULTY (COMBINING AGE AND SEX). F VALUE FOR MAIN EFFECT OF TASK DIFFICULTY (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 58). Level of Difficulty N Mean Score F (from Analysis of Variance) LTD 24 .458 MTD 24 1.041 51.055* HTD 24 2.125 *p < .001 178 Mean Conformity Score 8 7 6 5 4 5 ± 1 Low Task Medium Task High Task Difficulty Difficulty Difficulty Fig. 56. Mean conformity scores (2d private acceptance; of subjects for the three levels of task difficulty (combining age and sex) for Experiment IV. Range of scores also indicated. TABLE 48 EXPERIMENT IV MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OF SUBJECT FOR THE THREE LEVELS OF TASK DIFFICULTY (COMBINING AGE AND SEX). F VALUE FOR MAIN EFFECT OF TASK DIFFICULTY (TAKEN FROM ANALYSIS OF VARIANCE- TABLE 59). Level of Difficulty N Mean Score F (from Analysis of Variance) LTD 24 .250 MTD 24 .585 14.641 * HTD 24 1.291 *p < .001 179 scores of the subjects tested under the three levels of task difficulty is illustrated in Figure 54. It is clear that the higher the level of task difficulty, the higher the conformity scores. Having obtained a significant F value for the main effect of task difficulty (C), the data were analyzed by Scheffe's (1959) method of multiple comparisons to find the source of the effect. The results of the comparisons made are presented in Table 85 in the General Results section, p. 259- Table 47 presents the mean conformity (1st P.A.) scores for the three levels of task difficulty averaged over both sex and age. It is apparent that the main effect of task difficulty was significant (F = 51.055» P < 001). The mean conformity (1st P.A.) score of the subjects tested under the three levels of task difficulty are presented graphically in Figure 55* It is clear from the figure, the higher the level of task difficulty the higher the conformity (1st P.A.) scores. In order to determine the source of the significant main effect obtained for task difficulty (1st P.A.), Scheffe's (1959) method of multiple comparisons was employed. The results of the comparisons made are presented in Table 85 in the General Results section, p. 259. 180 Table 48 presents the mean conformity (2d P.A.) scores for the three levels of task difficulty averaged over both age and sex. The main effect of task diffi culty for private acceptance after one week was signifi cant (P = 14.641, p < .001). The difference in the mean conformity (2d P.A.) scores of the subjects tested under the three levels of task difficulty is illustrated in Figure 56. It is clear that as the difficulty of the task increased, there was a significant increase in mean conformity scores. Scheffe's (1959) method of multiple comparisons was used to determine the source of the sig nificant main effect for task difficulty (2d P.A.). The results of the comparisons made are presented in Table 85 in the General Results section, p. 259. The main effect of "discrepancy of contrived back ground report" was found to be significant in the test for compliance (F= 19.097» P < .001— Table 57)» the first test of private acceptance (F = 25.400, p < .001— Table 58), and in the test of private acceptance after one week (F = 12.550, p < .005— Table 59). That is, the closer the contrived report to the actual problem answer, the greater the conformity to that contrived report. A significant task difficulty by discrepancy of contrived background report interaction effect was also found for compliance (F = 4.774, p < ,025— Table 57)» the first test of private acceptance (F = 7-555» P < .005— 181 Table 38), and in the test for private acceptance after one week (F = 5*766, p < .05— Table 59)- That is, with increases in task difficulty conformity scores (0, 1st P.A. and 2d P.A.) increased for both discrepant background reports. However, the increase in conformity at each level of task difficulty was greater for the background reports closest to the actual problem answers. A significant age by task difficulty interaction effect was also obtained for compliance (F = 4.021, p < .001 — Table 58) and for the test of private acceptance after one week (F = 5-419, p < .005— Table 59). That is, with increases in task difficulty conformity scores (0, 1st P.A., and 2d P.A.) increased for both young and old sub jects; however, the increase at each level was greater for the older subjects. The age by discrepancy of contrived background report interaction effect was significant in the first test of private acceptance (F = 4.446, p < .05— Table 58). That is, when the contrived report approached the actual problem answer, conformity increased for both age groups; however, the increase was greater for the older subjects. CHAPTER VII EXPERIMENT V Investigations have repeal/odly studied suscepti bility to social influence utilizing perceptual tasks (Asch 1951 ; Kanareff & Lanzetta 1 9 6 0 ; Klein 1 9 7 0 ; Olmstead & Blake 1 9 5 5 ; Sherif 1 9 5 5 ; and many others). There have also been a number of investigations on susceptibility to social influence in reference to various attitudes and opinions (Allen & Crutchfield 1 9 6 5 ; Blake, Helson & Mouton 1956; DiVesta 1 9 5 8 ; Ferguson 1 9 4 4 ; Helson, Blake, Mouton & Olmstead 1 9 5 6 ; Horwitz, Piana, Goldman & Lee 1 9 5 5 ; and others). Although the conformity studies with perceptual tasks are, perhaps, more experimental in regard to control, they are somewhat less comparable to "real life" situations than are attitude task conformity studies. It would appear beneficial for conformity studies to consider both types of tasks in order to obtain a broader perspective of the area. The study of attitudes has occupied a central place in social psychology during the past 50 years. Cardno (1955) has stated, "Attitude entails an existing pre disposition to respond to social objects which, in inter action with situational and other dispositional variables, guides and directs the overt behavior of the individuals." 1 8 2 185 If the attitude of an individual toward an object is known, it can be used in conjunction with situational and other dispositional variables to explain, understand and predict reactions of that individual to that object. In addition to knowing the attitude, knowing the principles governing the change of the attitude will provide mean ingful information. It is the change in attitudes, or more specifically, the susceptibility of an individual to social influence which leads to a change in his attitudes, that was of concern in this experiment. The present experiment was directed at determining if age, sex, and task difficulty act as predictors of social conformity in a laboratory social attitude task. The specific attitudes this investigation focused on were those toward internationalism-nationalism. This experiment tested the following hypotheses using social conformity as the dependent variable : Hypothesis I: Informational social influence is greater upon the attitude ratings of older individuals than upon the attitude ratings of younger individuals. Hypothesis II: Informational social influence is greater upon the attitude ratings of females than upon the attitude ratings of males. hypothesis III: Informational social influence is greater upon the attitude ratings of the subjects as the task becomes more difficult (i.e., as the task becomes 184 mo re amb iguou s ) • PROCEDURES Twenty-four subjects were used in this experiment. Half of the subjects were young, ranging in age from 17 to 22. The other half were old, ranging in age from 55 to 81. (A more complete description of the subjects is given in the General Method section, p. 75-) Two additional groups of 24 individuals, selected from the same population as the general subject sample, were used to determine the percentage of young and old individuals who agree or disagree with the various attitude statements used in this experiment. One of these groups was comprised of younger individuals, ranging in age from 18 to 22. The other group consisted of older individuals, ranging in age from 62 to 79- Both of these standardizing groups had an equal number of males and females. The Klein (1970) conformity apparatus was utilized for this experiment. The apparatus is described in full detail in the General Method section, pp. 74-76. Attitude scale.— The attitude scale that was used in this experiment was the Worldmindedness Scale developed by Sampson & Smith (1957). The scale is designed to measure nationalistic-internationalistic attitudes. Sampson and Smith referred to it as a "social attitudes questionnaire." A complete description of the attitude____ 185 scale is given in the Appendix, p. 297. It should be made clear that the primary purpose of this investigation was not to determine age differences on these attitudes. Rather, the purpose was to determine age differences as to the susceptibility of the ratings of these attitudes to change due to social influence. Method In the initial session the subjects were given the attitude scale alone, then in the conformity situation and then again alone. They were given the scale again one week later in the alone situation. (This research design and the reasons for utilizing it have been discussed in the General Method section, pp. 79-80.) Social attitude task.— The subject was told he would be rating a series of statements by either agreeing or disagreeing with them. He was instructed that a series of slides would be projected on the screen in front of him and that each slide would present a statement. He was told to rate the statement by turning on his left switch if he agreed with the statement and his right switch if he disagreed with it. When performing in the conformity situation he was reminded to respond in the response position he had chosen at random (all ^s responded in the 4th position due to the deception employed). 186 A series of 52 trials (slide presentations) followed. Twenty-four were critical trials, occurring randomly, on which the contrived group consensus displayed a unanimous rating of the statement opposite to the majority of a standardizing group for that statement. The remaining eight trials were innocuous where peer ratings unanimously displayed the majority response given by a standardizing group for that statement. The innocuous trials were included in order to maintain the subject's acceptance of the genuineness of the apparatus and the situation. The critical trials were of low task difficulty, medium task difficulty, and high task difficulty. Of the twenty-four critical trials, there were eight representing each of the three levels of task difficulty. "Difficulty" of the attitude statements was assessed by the performance of the two standardizing groups of subjects before the actual experiment began. If 90 percent (or more) of the subjects of a standardizing group agreed or disagreed with an attitude statement, then it was classified as a low-task difficulty statement. If between 75 percent and 90 percent of the subjects of a standardizing group agreed or dis agreed with an attitude statement, then it was classified as a medium-task difficulty statement. If between 55 per cent and 75 percent of the subjects of a standardizing group agreed or disagreed with an attitude statement, then 187 it was classified as a high-task difficulty statement. By using separate age groups in standardizing the difficulty of the attitude statements, the task did not put either the young or old subjects at a disadvantage. The con ditions of the task for this experiment, for both young and old subjects, is presented in Tables 94 and 95 of the Appendix, pp. 507-508. An example of the attitude scale is presented in the Appendix, pp. 504-506. Directions for the alone situation.— This experi ment is concerned with social attitudes. A series of slides will be projected on the screen in front of you. On each slide will be a state ment. Your task will be to rate each statement on the basis of your agreement or disagreement with it. If you agree with the statement, turn on your left switch. If you disagree with the statement, turn on your right switch. After you make your ratings, leave your switch on until I tell you to turn it off. When each slide appears make your rating and wait for me to tell you to turn off your switch. There will be no time limit for your ratings, however, they should not take long. Are there any questions? Directions for the conformity situation.— This portion of the experiment is concerned with social attitudes in groups. Again, a series of slides will be projected on the screen in front of you. On each slide, as before, a statement will be shown. Again your task will be to rate each statement on the degree to which you agree or disagree with it. Again, the ratings cor respond to the following switches: left switch— agree, right switch— disagree. In this portion of the experiment, you will not only make your own ratings, but you will also be informed of the ratings of the other subjects. You will each be responding in a predetermined order which will be decided at random. In each of these envelopes, one of which each of you will pick at random, is a card with a 188 response order printed on it (either 1st, 2d, 5d, or 4-th). Do not show it to anyone after I see it; just keep it face up in front of you. This card indicates the order in which you are to make your ratings. On the panel in front of you, each of the pairs of bulbs will represent the responses of the subject who has selected that response position. Your switches will turn on those two bulbs that correspond to the response position you have selected. You will make your rating after you see one of the two bulbs light up in the response position that proceeds the one you have selected. If, for example., you have selected the first response position, then you will make your ratings as soon as the slides appear. If you have selected the 5d position, you will wait for the first two subjects to rate the statement, which will be indicated by either the left or right bulbs turning on in the first and second positions. Again, after you make your ratings, leave your switch on until I tell you to turn if off. When the next slide appears, wait for your turn to respond and then make your rating. There will be no time limit for your ratings, however, they should not take long. Are there any questions? Data Analysis The "alone score" was the frequency the subject responded with a non-majority rating on the critical trials in the alone situation. The "conformity score" was the frequency the subject agreed with the contrived group consensus (non-majority ratings) on the critical trials in the conformity situation. Tests 1, 2, and 5, described in the General Method section, were conducted. Each of these tests were ac complished by analyzing the data statistically through the use of a 2x2x5 mixed analysis of variance design. These were two-between one-within groups designs. This resulted 189 in testing three main effects and four interaction effects for the test of compliance (test l) and for the tests of private acceptance (tests 2 and 3). There was repeated measurements over the levels of task difficulty. That is, the subjects were tested under all three levels of task difficulty. Having found a significant main effect for task difficulty in tests 1, 2, and 3, Scheffe's (1959) method of multiple comparisons was employed to determine the source of the effect in each case. The results of the comparisons made are presented in Table 83 in the General Results section, p. 239- RESULTS Three 2x2x3 mixed analysis of variance designs were used to analyze statistically the data. Complete analysis of variance summary tables are given in Table 49 (compliance— G), Table 50 (initial test of private ac ceptance— 1st P.A.) and Table 51 (private acceptance after one week— 2d P.A. ). The data relevant to the hypothesis that older individuals are more conforming than younger individuals are given in Tables 52, 53, and 54 and are graphically presented in Figures 37, 38, and 39. Table 52 gives the mean conformity (C) scores of the two age groups, and indicates that the main effect of 190 TABLE 49 COMPLETE AEALXSIS OF VAEIAITCE SUmAHX TABLE FOE COKFLIANCE EXPEEIMEHT V Source of Variation df 8SQ MSQ F Age 1 21.125 21.125 15.000* Sex 1 4.015 4.015 2.470 Age X Sex 1 .680 .680 0.418 Error 20 32.499 1.625 Task Difficulty 2 76.777 58.588 54.841** Age X T. D. 2 14.533 7.166 10.258** Sex X T. D. 2 1.444 .722 1.051 Age X Sex x T. D. 2 .110 .055 0.079 Error 40 27.999 .6999 *p < .005 **p < .001 191 TABLE 50 COMPLETE ANALYSIS OF VARIANCE SUMMARY TABLE FOR THE FIRST TEST OF PRIVATE ACCEPTANCE EXPERIMENT V Source of Variation df 8SQ. MSQ F Age 1 21.125 21.125 15.518** Sex 1 4.015 4.015 2.550 Age X Sex 1 1.125 1.125 0.709 Error 20 51.722 1.586 Task Difficulty 2 44.194 22.097 19.887*** Age X T. D. 2 6.585 5.291 2.962* Sex X T. D. 2 2.527 1.265 1.157 Age X Sex x T. D. 2 .250 .125 0.112 Error 40 44.444 1.111 *p < .10 **p < .005 ***p < .001 192 TABLE 51 COMPLETE ANALYSIS OF VARIANCE SUMMARY TABLE FOR THE SECOND TEST OF PRIVATE ACCEPTANCE EXPERIMENT V Source of Variation df Bsq __ MSQ F Age 1 9.588 9.588 6.787 Sex 1 5.555 5.555 2.570 Age X Sex 1 .500 .500 0.561 Error 20 27.666 1.585 Task Difficulty 2 25.861 12.950 9.946 Age X T. D. 2 4.527 2.265 1.741 Sex X T. D. 2 2.194 1.097 0.844 Age X Sex x T. D. 2 .085 .041 0.052 Error 40 51.999 1.299 *p < .01 **p < .001 193 Mean Conformity Score 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 I Young Old Fig. 37" Mean conformity scores (compliance) of young and old subjects (combining the three levels of task difficulty) for Experiment V. Range of scores also indicated. TABLE 52 EXPERIMENT V MEAN CONFORMITY SCORES (COMPLIANCE) OF OLD AND YOUNG SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 49). Age . . N Mean Score F (from Analysis of Variance) Young 12 4.416 13.000* Old 12 7.666 *p < .005 194 Mean Conformity Score 24 23 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 Old Young Fig. 58. Mean conformity scores (1st private acceptance; of young and old subjects (combining the three levels of task difficulty) for Experiment V. Range of scores also indicated. TABLE 55 EXPERIMENT V MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF OLD AND YOUNG SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 50). Age N Mean Score F (from Analysis of Variance) Young 12 2.855 15.518* Old 12 6.085 *p < .005 195 Mean Conformity Score 24 25 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 Old Young Fig. 59. Mean conformity scores (2d private acceptance; of young and old subjects (combining the three levels of task difficulty) for Experiment V. Range of scores also indicated. TABLE 54 EXPERIMENT V MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OF OLD AND YOUNG SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 51). Age N Mean Score F (from Analysis of Variance) Young 12 2.085 6.787* Old 12 4.250 *p < .01 196 age was significant (F = 15.000, p < .005). The difference in mean conformity (G) scores of the two age groups is illustrated in Figure 57. It is clear that the mean conformity scores increased with age. Table 55 presents the mean conformity (1st P.A.) scores of the two age groups and indicates a significant main effect for age (F = 15.518, p < .005). Figure 58 graphically illustrates the dif ference in the mean conformity (1st P.A.) scores of the two age groups. It can be seen from the figure, that mean conformity scores increased with age. Table 54 gives the mean conformity (2d P.A.) scores of the two age groups. Again, the main effect of age was found to be significant (F = 6.787, p < .01). The difference in the mean con formity scores is diagrammatically presented in Figure 59. It is apparent from the figure, that the mean conformity scores increased significantly with age. The data from all three analyses (compliance, 1st P.A., 2d P.A.) provided support for the hypothesis that informational social influence is greater upon the social attitude ratings of older individuals than upon the social attitude ratings of younger individuals. The hypothesis, that females are more conforming than males, was not confirmed by the data pertaining to compliance, initial private acceptance or private ac ceptance after one week. The data pertaining to this hypothesis are given in Tables 55, 56, and 57 and are 197 Mean Conformity Score 24 2$ 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 3 2 1 Male Female Fig. 40. Mean conformity scores (compliance) of male and female subjects (combining the three levels of task difficulty) for Experiment V. Range of scores also indicated. TABLE 59 EXPERIMENT V MEAN CONFORMITY SCORES (COMPLIANCE) OF MALE AND FEMALE SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF SEX (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 49). Sex N Mean Score F (from Analysis of Variance) Male Female 12 12 5.335 6.750 2.470 198 Mean Conformity Score 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Male Female Male Fig. 41. Mean conformity scores (1st private acceptance) of male and female subjects (combining the three levels of task difficulty) for Experiment V. Range of scores also indicated. TABLE 36 EXPERIMENT V MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF MALE AND FEMALE SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF SEX.(taken from analysis OF VARIANCE— TABLE 30). Sex N Mean Score F (from Analysis of Variance) Male 12 3.750 2.350 Female 12 3.166 199 Mean Conformity Score 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Male Female Fig, 42. Mean conformity scores (2d private acceptance} of male and female subjects (combining the three levels of task difficulty) for Experiment V. Eange of scores also indicated. TABLE 57 EXPERIMEITT V MEAU CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OF MALE AND FEMALE SUBJECTS (COMBINING THE THREE LEVELS OF TASK DIFFICULTY). F VALUE FOR MAIN EFFECT OF SEX (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 51). Sex N Mean Score F (from Analysis of Variance) Male 12 2.499 Female 12 3.835 2.570 200 presented graphically in Figures 40, 41, and 42. Tables 55, 56, and 57 give the mean conformity (C, 1st P.A., and 2d P.A. respectively) of the male and female subjects. It is apparent that the main effect of sex was not significant in any of the three analysis of variance analyses. Figures 40, 41, and 42 illustrate the difference in the mean conformity (C, 1st P.A., 2d P.A. respectively) of the male and female subjects. It is evident from the figures, that the female subjects were only slightly more conforming than the male subjects. The hypothesis, that informational social influence is greater upon the social attitude ratings of the sub jects as the task becomes more difficult, was supported by the data pertaining to compliance and to both tests of private acceptance. The data relevant to this hypothesis are given in Tables 58, 59, and 60 and are illustrated diagrammatically in Figures 45, 44, and 45. Mean conformity (C) scores for the three levels of task difficulty, averaged over both age and sex, are pre sented in Table 58. It is apparent that the main effect of task difficulty is highly significant (F = 5'^* 841, p < .001). The difference in the mean conformity (C) scores of the subjects tested under the three levels of task difficulty is illustrated in Figure 45. It is evident that there was a significant increase in conformity scores as task difficulty increased. 201 8 7 6 Mean Conformity Score 4 Low Task Medium Task High Task Difficulty Difficulty Difficulty Fig. 45. Mean conformity scores (compliance) of subjects for the three levels of task difficulty (combining age and sex) for Experiment V. Range of scores also indicated. TABLE 58 EXPERIMENT V MEAN CONFORMITY SCORES (COMPLIANCE) OF SUBJECTS FOR THE THREE LEVELS OF TASK DIFFICULTY (COMBINING AGE AND SEX). F VALUE FOR MAIN EFFECT OF TASK DIFFICULTY (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 49). . Level of Difficulty N Mean Score F (from Analysis of Variance) LTD 24 .875 MTD 24 1.791 54.841* HTD 24 5.575 *p < .001 ■2Ü2 8 7 Mean Conformity Score 5 4 t Low Task Medium Task High Task Difficulty Difficulty Difficulty Fig. 44. Mean conformity scores (1st private acceptance; of subjects for the three levels of task difficulty (combining age and sex) for Experiment V. Range of scores also indicated. TABLE 59 EXPERIMENT V MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE OF SUBJECTS FOR THE THREE LEVELS OF TASK DIFFICULTY (COMBINING AGE AND SEX). F VALUE FOR MAIN EFFECT OF TASK DIFFICULTY (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 50). Level of Difficulty N Mean Score F (from Analysis of Variance) LTD 24 • 666 MTD 24 1.250 19.887* HTD 24 2.541 *p < .001 205 Mean Conformity Score 8 7 6 5 ± Low Task Medium Task High Task Difficulty Difficulty Difficulty Fig. 45* Mean conformity scores (2d private acceptance; of subjects for the three levels of task difficulty (combining age and sex) for Experiment V. Range of scores also indicated. TABLE 60 EXPERIMENT V MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OF SUBJECT FOR THE THREE LEVELS OF TASK DIFFICULTY (COMBINING AGE AND SEX). F VALUE FOR MAIN EFFECT OF TASK DIFFICULTY (TAKEN FROM ANALYSIS OF VARIANCE- TABLE 51). Level of Difficulty N Mean Score F (from Analysis of Variance) LTD 24 .458 MTD 24 .8555 9.946* HTD 24 1.875 *p < .001 204 In order to find the source of the significant main effect obtained for task difficulty (C), Scheffe's (1959) method of multiple comparisons was employed. The results of 'the comparisons made are presented in Table 85 in the General Results section, p. 259- Table 59 presents the main conformity (1st P.A.) scores for the three levels of task difficulty averaged over both age and sex. It is apparent that the main effect of task difficulty was significant (P = 19-887, p < .001). The difference in the mean conformity (1st P.A.) scores of the subjects tested under the three levels of task diffi culty are presented graphically in Figure 44. It is evident from the figure, the higher the level of task difficulty the higher the conformity scores. Having obtained a significant F value for the main effect of task difficulty (1st P.A.), the data were analyzed by Scheffe's (1959) method of multiple comparisons to determine the source of the effect. The results of the comparisons made are presented in Table 85 in the General Results section, p. 259- Table 60 gives the mean conformity (2d P.A.) scores of the subjects for the three levels of task difficulty. The main effect of task difficulty for private acceptance after one week was significant (F = 9-946, p < .001). The difference in the mean conformity (2d P.A.) scores of the subjects tested under the three levels of task difficulty 205 are presented diagrammatically in Figure 45- It is apparent from the figure, the higher the level of task difficulty the higher the conformity scores. Scheffe's (1959) method of multiple comparisons was used to find the source of the significant main effect of task difficulty (2d P.A.). The results of the comparisons made are given in Table 83 in General Results section, p. 239. The analysis of the data pertaining to compliance, presented in Table 49, also indicated a significant age by task difficulty interation effect (F = 10.258, p < .001). That is, with increases in task difficulty conformity (C) scores increased for both young and old subjects ; however, the increase at each level of difficulty was greater for the older subjects. CHAPTER VIII GENERAL RESULTS Three 2x2x5x5 mixed analysis of variance designs were used to analyze statistically the combined data of the five experiments. These were three between-groups, one within-groups designs. The three between-groups variables were age, sex, and experimental task. The within-groups variable was task difficulty. In each of the analyses there were four main effects and 11 inter actions tested. Complete analysis of variance summary tables are given in Table 61 (compliance— C), Table 62 (initial test of private acceptance— 1st P.A.), and Table 65 (private acceptance after one week— 2d P.A.).^ The general hypothesis of this investigation, in reference to age, stated: Informational social influence is greater upon the judgments or opinions of older subjects than upon the judgments or opinions of younger subjects; further, this hypothesis will hold in all five experiments. The data relevant to the first portion of this hypothesis It was recognized that heterogeneity of variance may have been present. However, this probably would not have affected the significance of the main effects that were found. 206 207 TABLE 61 COMPLETE ANALYSIS OF VARIANCE SUMMARY TABLE FOR COMPLIANCE COMBINED ANALYSIS Source of Variation df 8SQ MSQ F Age 1 59.211 59.211 26.725** Sex 1 5.211 5.211 1.449 Experimental Task 4 99.094 24.775 11.181** Age X Sex 1 .044 .044 0.020 Age X E. T. 4 6.816 1.704 0.769 Sex X E. T. 4 14.761 5.690 1.665 Age X Sex x E. T. 4 1.261 .515 0.142 Error 100 221.551 2.215 Task Difficulty 2 210.688 105.544 159.451** Age X T. D. 2 52.822 16.411 21.724** Sex X T. D. 2 .688 .544 0.455 E. T. X T. D. 8 15.921 1.990 2.654* Age X Sex x T. D. 2 .288 .144 0.191 Age X E. T. X T. D. 8 5.066 .585 0.507 Sex X E. T. X T. D. 8 4.588 .575 0.759 Age X Sex x E. T. X T. D. 8 1.488 .180 0.246 Error 200 151.084 .755 *p < .01 **p < .001 20 8 TABLE 62 COMPLETE ANALYSIS OF VARIANCE SUMMARY TABLE FOR THE FIRST TEST OF PRIVATE ACCEPTANCE COMBINED ANALYSIS Source of Variation df SSQ MSQ. F Age 1 3 5 .0 0 2 55.002 24.785*** Sex 1 4.669 4.669 5.506* Experimental Task 4 71.685 17.920 15.457*** Age X Sex 1 .625 .625 0.469 Age X E. T. 4 16.761 5.440 2.585** Sex X E. T. 4 5.705 1.426 1.071 Age X Sex x E. T. 4 1.561 .540 0.255 Error 100 155.164 1.551 Task Difficulty 2 111.216 55.608 84.054*** Age X T. D. 2 17.772 8 .8 8 6 1 5. 45 1*** Sex X T. D. 2 .858 .419 0.654 E. T. X T. D. 8 10.785 1.547 2.057** Age X Sex x T. D. 2 .549 .175 0.264 Age X E. T. X T. D. 8 5.005 .575 0.567 Sex X E. T. X T. D. 8 2.994 .574 0.565 Age X Sex x E. T. X T. D. 8 .705 .088 0.155 Error 200 152.515 .661 *p < .10 **p < .05 ***p < .001 COMPLETE FOR THE 209 TABLE 63 ANALYSIS OF VARIANCE SUMMARY TABLE SECOND TEST OF PRIVATE ACCEPTANCE COMBINED ANALYSIS Source of Variation df SSQ MSQ F Age 1 9.344 9.344 10.809*** Sex 1 1.599 1.599 1.850 Experimental Task 4 52.016 15.004 15.045*** Age X Sex 1 .177 .177 0.205 Age X E. T. 4 8.961 2.240 2.591* Sex X E. T. 4 3.485 .870 1.007 Age X Sex x E. T. 4 .627 .156 0.181 Error 100 86.445 .864 Task Difficulty 2 27.172 15.586 51.393*** Age X T. D. 2 4.258 2.119 4.897** Sex X T. D. 2 1.216 .608 1.405 E. T. X T. D. 8 15.216 1.902 4.595*** Age X Sex x T. D. 2 .105 .052 0.121 Age X E. T. X T. D 8 5.372 .671 1.551 Sex X E. T. X T. D 8 1.785 .222 0.515 Age X Sex x E. T. X T. D. 8 .338 .042 0.097 Error 200 86.553 .452 *p < .05 **p < .005 ***p < .001 210 are given in Tables 64, 65, and 66 and are illustrated in Figures 46, 47, and 49. Tables 67, 68, and 69 and Figures 49, 50, and 51 present data pertaining to the first and second portion of this hypothesis. Table 64 gives the mean conformity (C) scores of the two age groups, averaged over the five experimental tasks, and indicates that the main effect of age was sig nificant (F = 26.725, P < .001). The difference in mean conformity (G) scores of the two age groups is illustrated in Figure 46. It is clear that the mean conformity scores, averaged over the five experimental tasks, increased sig nificantly with age. Table 65 presents the mean conformity (1st P.A.) scores of the old and young subjects, averaged over the five experimental tasks, and indicates a sig nificant main effect for age (F = 24.783, p < .001). Figure 47 illustrates graphically the difference in the mean conformity (1st P.A.) scores of the two age groups. It is evident that the mean conformity scores, averaged over the five experimental tasks, increased significantly with age. Table 66 gives the mean conformity (2d P.A.) scores of the two age groups averaged over the five experi mental tasks. Again, the main effect of age was found to be significant (F = 10.809, p < .001). The difference in the mean conformity (2d P.A.) scores is presented dia- grammatically in Figare 48. It is apparent that the mean conformity scores, averaged over the five experimental 211 Mean Conformity Score 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 5 2 1 Young Old Pig. 46. Mean conformity scores (compliance) of young and old subjects (averaged over the five experimental tasks). Range of scores also indicated. TABLE 64 MEAN CONFORMITY SCORES (COMPLIANCE) OP OLD AND YOUNG SUBJECTS (AVERAGED OVER THE PIVE EXPERIMENTAL TASKS). P VALUE POR MAIN EPPECT OP AGE (TAKEN PROM ANALYSIS OP VARIANCE— TABLE 61 ). Age N Mean Score P (from Analysis of Variance) Young 60 2.750 26.725* Old 60 5.183 *p < .001 Mean Conformity Score 24 23 22 21 20 19 18 17 16 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Young Old Fig. 47. Mean conformity scores (1st private acceptance; of young and old subjects (averaged over the five experimental tasks). Range of scores also indicated. TABLE 65 MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF OLD AND YOUNG SUBJECTS (AVERAGED OVER THE FIVE EXPERIMENTAL TASKS). F VALUE FOR MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 62). Age N Mean Score F (from Analysis of Variance) Young 60 1.616 24.783* Old 60 3.433 *p < .001 213 Mean Conformity Score 24 25 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 ± Old Young Fig. 48. Mean conformity scores (2d private acceptance; of young and old subjects (averaged over the five experimental tasks). Range of scores also indicated. TABLE 66 m m GOKFOEMITY SCORES (2d PRIVATE ACCEPTANCE) OF OLD AND YOUNG SUBJECTS (AVERAGED OVER THE FIVE EXPERIMENTAL TASKS). F VALUE FOR MAIN EFFECT OF AGE (TAKEN FROM ANALYSIS VARIANCE— TABLE 65). Age N Mean Score F (from Analysis of Variance) Young 60 .855 10.809* Old 60 1.800 *p < .001 . -... 214 Mean Conformity Score 24 25 22 21 20 19 18 17 16 15 1 4 15 12 11 10 9 8 7 6 5 4 5 2 1 Old Young Exp. I Exp. II Exp. Ill Exp. IV Exp. V Fig. 49. Mean conformity scores (compliance) of young and old subjects for each of the five experiments. TABLE 67 MEAN CONFORMITY SCORES (COMPLIANCE) OF YOUNG AND OLD SUBJECTS FOR EACH OF THE FIVE EXPERIMENTS. F VALUES FOR MAIN EFFECT OF AGE (TAKEN FROM SEPARATE ANALYSIS OF VARIANCE ANALYSES). Experiment N Mean Young Score Old F (from Analysis of Variance) Experiment I 24 1.085 2.585 5.225** Experiment II 24 2.000 5.585 4.095* Experiment III 24 2.585 4.916 5.599** Experiment IV 24 5.666 7.166 4.515** Experiment V 24 4.416 7.666 15.000*** *p < .10 **p < .05 ***p < .005 215 Mean Conformity Score 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Old Young Exp. I Exp. II Exp. Ill Exp. IV Exp. V Fig. 90. Mean conformity scores (1st private acceptance; of young and old subjects for each of the five experiments. TABLE 68 MEAN GONFOEMITY SCORES (1st PRIVATE ACCEPTANCE) OF YOUNG AND OLD SUBJECTS FOR EACH OF THE FIVE EXPERIMENTS. F VALUES FOR MAIN EFFECT OF AGE (TAKEN FROM SEPERATE ANALYSIS OF VARIANCE ANALYSES). Experiment N Mean Young Score Old F (from Analysis of Variance) Experiment I 24 . 666 .999 1.081 Experiment II 24 1.083 1.833 1.680 Experiment III 24 1.416 3.000 4.056* Experiment IV 24 2.083 5.166 6.475** Experiment V 24 2.833 6.083 13.318*** *p < .10 **p < .025 ***p < .001 216 Mean Conformity Score 24 20 18 16 14 12 11 10 Old Young Exp. I Exp. II Exp. Ill Exp. IV Exp. V Eig 51* Mean conformity scores (2d private acceptance; of young and old subjects for each of the five experiments. TABLE 69 MEAE CONEOEMITY SCOPES (2d PRIVATE ACCEPTANCE) OF YOUNG AND OLD SUBJECTS FOE EACH OF THE FIVE EXPERIMENTS. F VALUES FOR MAIN EFFECT OF AGE TAKEN FROM SEPARATE ANALYSIS OF VARIANCE ANALYSES). Experiment N Mean Young Score Old F (from Analysis of Variance Experiment I 24 .083 .333 2.368 Experiment II 24 .500 .217 Experiment III 24 .300 .750 .217 Experiment IV 24 1.083 3.166 4.240 Experiment V 24 2.083 4.250 6.787 *p < .10 **p < .01 217 tasks, increased signficantly with age. Table 67 presents the mean conformity (C) scores of the young and old subjects, and the F values for the main effect of age, for each of the five experiments. The main effect of age was found to be significant in Experi ment I (F = 5*225, p < .05), Experiment III (F = 5.359, p < .05), Experiment IV (F = 4.513, P < .05), and Experi ment V (F = 13.000, p < .005). The difference in the mean conformity (C) scores of the two age groups for the five experiments is presented graphically in Figure 49. It is apparent from the figure, that mean conformity scores were higher for the older subjects on all experi mental tasks. Table 68 gives the mean conformity (1st P.A.) scores of the young and old subjects, and the F values for the main effect of age (1st P.A.), for each of the five experiments. The main effect of age was significant in Experiment IV (F = 6.475, P < .025) and Experiment V (F = 13,318, p < .005). The difference in the mean con formity (1st P.A.) scores of the two age groups for the five experiments is illustrated graphically in Figure 50. It is apparent, that although the difference between the age groups only reached significance at the .05 level of significance in Experiments IV and V, the mean conformity scores were higher for older subjects on all experimental tasks. 218 Table 69 presents the mean conformity (2d P.A.) scores of the young and old subjects and the F values for the main effect of age (2d P.A.) for each of the five experiments. The main effect of age was significant in Experiment V (F = 6.787, P < -01). The difference in the mean conformity (2d P.A.) scores of the two age groups for the five experiments is presented graphically in Figure 51- It is evident, that although the difference between the age groups only reached significance at the .05 level of significance in Experiment V, the mean conformity scores were higher for older subjects on all experimental tasks. The data from all the analyses, especially in the case of compliance, provide support for the hypothesis that older individuals are more conforming than younger individuals. The data also support the generalizability of this hypothesis as indicated by the higher conformity (C, 1st P.A., and 2d P.A.) scores for the older subjects in all experiments. Sex The general hypothesis of this investigation, in reference to sex, stated: Informational social in fluence is greater upon the judgments or opinions of females than upon the judgments or opinions of males; further, this hypothesis will hold in all five experi ment 8 . 219 The data relevant to the first portion of this hypothesis are given in Tables 70, 71, and 72 and are illustrated in Figures 52, 53, and 54-. Tables 75, 74, and 75 and Figures 53, 56, and 57 present data pertaining to the first and second portion of this hypothesis. Tables 70, 71, and 72 give the mean conformity (C, 1st P.A., and 2d P.A. respectively) scores of the male and female subjects averaged over the five experimental tasks. It is apparent that the main effect of sex was not significant in any of the three analysis of variance analyses. Figures 52, 53, and 54 illustrate the difference in the mean conformity (C, 1st P.A., and 2d P.A. respective ly) scores of the male and female subjects. It is evident from the figures, that the females were only slightly more conforming than the males when the data of the five experiments was combined. Table 73 presents the mean conformity (C) scores of the male and female subjects, and F values for the main effect of sex (C), for each of the five experiments. The main effect of sex (C) was not significant in any of the five experiments. The difference in the mean con formity (C) scores of the male and female subjects for the five experiments is presented graphically in Figure 55. It is apparent from the figure that the mean conformity scores were higher for the female subjects in Experiments I, III, and V and were higher for the male subjects in 220 Mean Conformity Score 24 25 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 Male Female Fig. 52. Mean conformity scores (compliance) of male and female subjects (averaged over the five experimental tasks). Range of scores also indicated. TABLE 70 MEAE CONFORMITY SCORES (COMPLIANCE) OF MALE AND FEMALE SUBJECTS (AVERAGED OVER THE FIVE EXPERIMENTAL TASKS). ANALYSIS OF VARIANCE— TABLE 61 ). F (from Analysis Sex N Mean Score of Variance) Male 60 5.685 1.449 Female 60 4.250 Mean Conformity Score 24 25 22 21 20 19 18 17 16 1 5. 14 15 12 11 10 9 8 n ( 6 —— 5 4 5 2 1 Î T Male Female Fig. 53* Mean conformity scores (1st private acceptance) of male and female subjects (averaged over the five experimental tasks). Range of scores also indicated. TABLE 71 MEAFT COUFOHMITT SCORES (1st PRIVATE ACCEPTAiaCE) OF MALE AND FEMALE SUBJECTS (AVERAGED OVER THE FIVE EXPERIMENTAL TASKS). F VALUE FOR MAIN EFFECT OF SEX (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 62). Sex Mean Score F (from Analysis of Variance Male 60 2.185 5.506* Female 60 2.866 *p < .10 Mean Conformity Score 24 25 22 21 20 19 18 17 16 15 14 15 12 11 10 9 8 7 6 5 4 5 2 1 Male Female Fig. 54. Mean conformity scores (2d private acceptance) of male and female subjects (averaged over the five experimental tasks). Eange of scores also indicated. TABLE 72 MEAN CONFOEMITT SCOEES (2d PEIVATE ACCEPTANCE) OF MALE AND FEMALE SUBJECTS (AVEEAGED OVEE THE FIVE EXPEEIMENTAL TASKS). F VALUE FOE MAIN EFFECT OF SEX (TAKEN FEOM ANALYSIS OF VAEIANCE— TABLE 65). F Sex N Mean Score (from Analysis of Variance) Male Female 60 60 1.116 1.516 1.850 Mean Conformity Score Exp. I Exp. II Exp. Ill Exp. IV Exp. V Fig. 55* Mean conformity scores (compliance) of male and female subjects for each of the five experi ments. TABLE 75 MEAN CONFOEMITY SCOEES (COMPLIANCE) OF MALE AND FEMALE SUBJECTS FOR EACH OF THE FIVE EXPERIMENTS. F VALUES FOR MAIN EFFECT OF SEX (TAKEN FROM SEPARATE ANALYSIS OF VARIANCE ANALYSES). Experiment N Mean Male Score Female F (from Analysis of Variance) Experiment I 24 1.585 2.085 .580 Experiment IT 24 5.166 2.416 .918 Experiment III 24 5.916 5.585 .110 Experiment IV 24 4.249 6.585 2.006 Experiment V 24 5.555 6.750 2.470 224 Mean Conformity Score 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 5 2 1 emale Male Exp. I Exp. II Exp. Ill Exp. IV Exp. V Fig. 56. Mean conformity scores (1st private acceptance; of male and female subjects for each of the five experiments. TABLE 74 MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF MALE AND FEMALE SUBJECTS FOR EACH OF THE FIVE EXPERIMENTS. F VALUES FOR MAIN EFFECT OF SEX (TAKEN FROM SEPARATE ANALYSIS OF VARIANCE ANALYSES). Experiment N Mean Male Score Female F (from Analysis of Variance) Experiment I 24 .750 .916 .270 Experiment II 24 1.58:5 1.333 .186 Experiment III 24 2.083 2.333 .101 Experiment IV 24 2.749 4.500 2.086 Experiment V 24 ."750 5.166 2.530 225 Mean Conformity Score 24 25 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 5 2 1 Exp. I Exp. II Exp. Ill Exp. IV Exp. V Fig. 57* Mean conformity scores (2d private acceptance; of male and female subjects for each of the five experiments. TABLE 75 MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OF MALE AND FEMALE SUBJECTS FOR EACH OF THE FIVE EXPERIMENTS. F VALUES FOR MAIN EFFECT OF SEX (TAKEN FROM SEPARATE ANALYSIS OF VARIANCE ANALYSES). Experiment N Mean Male Score Female F (from Analysis of Variance) Experiment I 24 .083 .353 2.368 Experiment II 24 .500 .333 .217 Experiment III 24 .750 .500 .217 Experiment IV 24 1.750 2.500 .549 Experiment V 24 2.499 3.833 2.570 226 Experiment II. The mean conformity (2d P.A.) scores of the male and female subjects, and the P values for the main effect of sex for the five experiments, are presented in Table 75* The main effect of sex was not significant in any of the experiments. The difference in mean con formity (2d P.A.) scores of the male and female subjects for the five experiments is illustrated in Figure 57. It is clear from the figure that the mean conformity (2d P.A.) scores were higher for the female subjects in Ex periments I, III, and V and were higher for the male subjects in Experiments II and IV. The data from all the analyses generally do not provide support for the hypothesis that females are more conforming than males. The main effect of sex was not significant at the .05 level, in any of the separate experiments, nor was it significant in the general analyses when the data from the individual experiments was combined. It was also found, in a number of cases, that it was the male subjects who were slightly more conforming. Task Difficulty The general hypothesis of this investigation, in reference to task difficulty, stated: Informational social influence is greater upon the judgments or opinions of the subjects as the stimulus of the task becomes more ambiguous (i.e., as the task becomes more difficult); further, this hypothesis will hold in each of the five experiments. 227 The data relevant to the first portion of this hypothesis are given in Tables 76, 77, and 78 and are illustrated in Figures 58, 59, and 60. Tables 80, 81, and 82 and Figures 61, 62, and 65 present data pertaining to the first and second portion of this hypothesis. Table 76 gives the mean conformity (C) scores for the three levels of task difficulty, averaged over the five experimental tasks, and indicates that the main effect of task difficulty was highly significant (F = 159.451, p < .001). The difference in mean conformity (C) scores for the three levels of task difficulty, combining the data of the five experiments, is illustrated in Figure 58. It is clear that the mean conformity scores increased significantly as the task becomes more diffi cult. Table 77 presents the mean conformity (1st P.A.) scores of the three levels of task difficulty, averaged over the five experimental tasks, and indicates that the main effect of task difficulty was significant (F = 84.054, p < .001). Figure 59 illustrates graphically the dif ference in the mean conformity (1st P.A.) scores of the three levels of task difficulty. It is evident that the mean conformity scores, based on the combined data of the five experiments, increased significantly as the task became more difficult. Table 78 gives the mean conformity (2d P.A.) scores of the three levels of task difficulty, averaged over the five experimental tasks, and indicates 228 8 7 6 Mean Conformity Score t Low Task Medium Task High Task Difficulty Difficulty Difficulty Fig. 58. Mean conformity scores (compliance) of the three levels of task difficulty (averaged over the five experimental tasks). Range of scores also indicated. TABLE 76 MEAN CONFORMITY SCORES (COMPLIANCE) OF THE THREE LEVELS OF TASK DIFFICULTY (AVERAGED OVER THE FIVE EXPERIMENTAL TASKS). F VALUE FOR MAIN EFFECT OF TASK DIFFICULTY (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 61). Level of Difficulty Mean Score F (from Analysis of Variance) LTD MTD HTD 'p < .001 .485 1.150 2.555 159.151* 229 8 7 6 Mean Conformity Score 5 ± Low Task Difficulty Medium Task Difficulty High Task Difficulty Fig. 59. Mean conformity scores (1st private acceptance) of the three levels of task difficulty (averaged over the five experimental tasks). Range of scores also indicated. TABLE 77 MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) OF THE THREE LEVELS OF TASK DIFFICULTY (AVERAGED OVER THE FIVE EXPERIMENTAL TASKS). F VALUE FOR MAIN EFFECT OF TASK DIFFICULTY (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 62). Level of Difficulty Mean Score F (from Analysis of Variance) LTD MTD HTD 'p < .001 .285 .641 1.600 84.054' 230 Mean Conformity Score 8 7 6 5 4 - 5 2 1 Low Task Medium Task High Task Difficulty Difficulty Difficulty Pig. 60. Mean conformity scores (2d private acceptance; of the three levels of task difficulty (averaged over the five experimental tasks). Eange of scores also indicated. TABLE 78 MEAH GONPOEMITY SCOEES (2d PEIVATE ACCEPTANCE) OP THE THEEE LEVELS OP TASK DIPPICULTY (AVEEAGED OVEE THE PIVE EXPEEIMENTAL TASKS). P VALUE POE MAIN EPPECT OP TASK DIPPICULTY (TAKEN PEOM ANALYSIS OP VAEIANCE— TABLE 63). Level of Difficulty LTD MTD HTD Mean Score .150 .558 .808 P (from Analysis of Variance) 51.595* 'p . ( ) ( ) 1 231 that the main effect of task difficulty was significant (P = 31.395, p < 001). The difference in mean conformity (2d P.A.) scores for the three levels of task difficulty, combining the data of the five experiments, is illustrated in Pigure 60. It is evident that the mean conformity scores increased as the difficulty of the task increased. Having found a significant main effect for task difficulty (G, 1st P.A. and 2d P.A.) on the combined data of the five experiments, Scheffe's (1959) method of multiply comparisons was used to determine the source of the effect in each case. The results of the comparisons made are given in Table 79. Comparison 1 was carried out to determine if the mean conformity scores (C, 1st P.A., and 2d P.A.) of low task difficulty and high task difficulty were significantly different. It can be seen from Table 79 that this comparison was significant in the case of compliance (P = 155*71, p < .001), the first test of private acceptance (P = 79*18, p < .001) and private acceptance after one week (P = 30.17, P < .001). Compari son 2 was conducted in order to ascertain if the mean conformity scores (C, 1st P.A. and 2d P.A.) of medium task difficulty and high task difficulty were significantly different. It is apparent from Table 79 that this com parison was significant in the case of compliance (P = 31.98, p < .001), the first test of private acceptance (P = 4-1.86, p < .001) and private acceptance after one 232 TABLE 79 COHPASISOUS BETWEEH levels OE task DIFEICULTT ON THE COMBINED DATA OE THE EIVE EXPERIMENTS Comparison Compliance Pirst Private Acceptance Second Private Acceptance 1 P = 135.71 P = 79.18 P = 30.17 LTD— HTD p < .001 p < .001 p < .001 2 P = 51.98 P = 41.86 P = 14.46 MTD— HTD p < .001 p < .001 p < .001 5 p = 19.4-0 p = 5.86 P = 2.86 LTD— MTD p < .001 p < .005 p < .10 233 week (P = 14.45, p < .001). Comparison 3 was made in order to determine if the mean conformity scores (C, 1st P.A., and 2d P.A.) of low task difficulty and medium task difficulty were significantly different. It is clear from Table 79 that this comparison was significant in the case of compliance (P = 19*40, p < .001) and the first test of private acceptance (P = 5*86, p < .003). This comparison was not significant for the second test of private ac ceptance (P = 2.86, p < .10) at the .03 level of signi ficance. It can be concluded that the significant main effects obtained for task difficulty (C, 1st P.A., and 2d P.A. ) on the combined data, was due to significant differences between each of the levels of task difficulty. Table 80 presents the mean conformity (0) scores of the three levels of task difficulty, and the P values for the main effect of task difficulty (C), for each of the five experiments. The main effect of task difficulty was found to be significant in Experiment I (P = 20.317, p < .001), Experiment II (P = 17.296, p < .001), Experi ment III (P = 21.346, p < .001), Experiment IV (P = 32,386, p < .001), and Experiment V (P = 34-.841, p < .001). The difference in the mean conformity (0) scores of the three levels of task difficulty for the five experiments is presented graphically in Pigure 61. It is apparent from the figure that mean conformity scores increased, as task difficulty increased, on all experimental tasks. Mean Conformity Score Exp. I Exp. II Exp. Ill Exp. IV Exp. V Eig. 61. Mean conformity scores (compliance) of the three levels of task difficulty for each of the five experiments. TABLE 80 MEAN CONFOEMITY SCOEES (COMPLIANCE) OE THE THEEE LEVELS OE TASK DIEEICULTY FOE EACH OE THE FIVE EXPEEIMENTS. E VALUES POE MAIN EPPECT OP TASK DIPPICULTY (TAKEN PEOM SEPAEATE ANALYSIS OP VAEIANCE ANALYSES). Experiment Low Task Difficulty Mean Score Medium Task Difficulty High Task Difficulty P Experiment I .123 .438 1.230 20.517* Experiment II .230 -750 1.791 17.996* Experiment III .416 1.166 2.166 21.546* Experiment IV .750 1.585 5.085 52.586* Experiment V .875 1.791 5.575 34.841* 'p < .001 253 Table 81 presents the mean conformity (1st P.A.) scores of the three levels of task difficulty, and the P values for the main effect of task difficulty (1st P.A.), for each of the five experiments. The main effect of task difficulty was found to be significant in Experiment I (P = 15.424, p < .001), Experiment II (P = 15.287, p < .001), Experiment III (P = 15.042, p < .001), Experiment XV (P = 51.033, P < .001), and Experiment V (P = 19.887, p < .001). The difference in the mean conformity (1st P.A.) scores of the three levels of task difficulty for the five experiments is presented graphically in Pigure 52. It is evident from the figure that mean conformity scores increased, as task difficulty increased, on all experi mental tasks. Table 82 gives the mean conformity scores (2d P.A.) of the three levels of task difficulty, and the P values for the main effect of task difficulty (2d P.A.), for each of the five experiments. The main effect of task difficulty was found to be significant in Experiment II (P = 5.505, p < .03), Experiment III (P = 4.000, p < .03), Experiment IV (P = 14,541, p < .001) and Experiment V (P = 9.946, p < .001). The difference in the mean con formity scores (2d P.A.) of the three levels of task difficulty for the five experiments is illustrated in Pigure 65. It is clear from the figure that mean con formity scores increased, as task difficulty increased, on 236 Mean Conformity Score 8 7 6 5 4 5 HTD 2 MTD LTD 1 Exp. I Exp. II Exp. ill Exp. IV Exp. V Ei( acceptance 62.. Mean conformity scores (1st private of tiie three levels of task difficulty for each of the five experiments. TABLE 81 MEAH COHEOEMITY SCOEES (1st PEIVATE ACCEPTAHjCE) OP THE THEEE LEVELS OP TASK DIPPICULTY POE EACH Ofi THE PIVE EXPEEIMENTS. P VALUES POE MAIN EPPECT OP TASK DIPPICULTY (TAKEN PEOM THE SEPAEATE ANALYSIS OP VARIANCE ANALYSES). Experiment Low Task Difficulty Mean Score Medium Task Difficulty Difficulty P Experiment I .000 .123 .708 13.424» Experiment II .083 .29M 1.083 13.287* Experiment III .208 .300 1.300 13.CXL2* Experiment IV .438 1.041 2.123 $1,033* Experiment V . 666 1.230 2.341 19.887* 'p < .001 237 8 7 Mean Conformity Score 4 3 2 1 Exp. I Exp. II Exp. Ill Exp. IV Exp. V Fig. 63. Mean conformity scores (2d private acceptance; of the three levels of task difficulty for each of the five experiments. TABLE 82 MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) OF THE THREE LEVELS OF TASK DIFFICULTY FOR EACH OF THE FIVE EXPERIMENTS. F VALUES FOR MAIN EFFECT OF TASK DIFFICULTY (TAKEN FROM THE SEPARATE ANALYSIS OF VARIANCE ANALYSES). HTD Experiment Low Task Difficulty Mean Score Medium Task Difficulty High Task Difficulty F Experiment I .000 .041 .166 2.600 Experiment II .000 .123 .291 3.303* Experiment III .041 .208 .373 4.000* Experiment IV .230 .383 1.291 14.641** Experiment V .438 .833 1.873 9.946** 'P < .03 * * p < .001 238 all experimental tasks. The data from all the analyses provide support for the hypothesis that conformity increases as the level of difficulty increases. The data also support the generaliz- ahility of this hypothesis as indicated by the higher con formity scores obtained at higher levels of difficulty in all the experiments. Having obtained the significant F values for the main effect of task difficulty within each experiment, the data were analyzed in each case to find the source of the effect. This was accomplished by employing Scheffe’s (1939) method of multiple comparisons. The results of the comparisons made in Experiments 1, II, III, IV, and V are presented in Table 83. Comparison 1 was aimed at determining if the mean conformity scores (C, 1st P.A., and 2d P.A.) of low task difficulty and high task difficulty were significantly different. It can be seen from Table 85 that this com parison was significant in every, experiment. Comparison 2 was conducted in order to ascertain if the mean con formity scores (C, 1st P.A., and 2d P.A.) of medium task difficulty and high task difficulty were significantly different. This comparison was significant in almost every experiment for all three conformity measures. Comparison 3 was made in order to determine if the mean conformity scores (C, 1st P.A., and 2d P.A.) of low task [ > - • o C \ J T- • o OJ OJ pH P K Û ' T * • S C \ J o C \ J o- • V LTNO o o f a n P Pq P pqPi KN 00 •9 vOJ • o ( \ j CM CM ^5 ( T O IIV PH A C Q CM •H ^5 00 O •9 V CM VD V •O vO •9 CO o • H CM [ > - L T \ V ^•8 O H 1 —I M 5 M -P •H •H •H •H 240 difficulty and medium task difficulty were significantly different. It is apparent from Table 83 that this comparison was generally significant for compliance, but it was not generally significant at the .05 level of sig nificance for the tests of private acceptance. Generally, it can be concluded that the difference between low task difficulty and high task difficulty, and the difference between medium task difficulty and high task difficulty, contributed to the overall significant ? values obtained for task difficulty (C, 1st P.A., and 2d P.A.) in the five experiments. It can also be con cluded that the difference between low task difficulty and medium task difficulty only contributed to the significant P values obtained for task difficulty, in the five experi ments, in the case of compliance. Nature of the Task The general hypothesis, in reference to the nature of the experimental task, stated: Informational social influence is greater upon the judgments or opinions of the subjects as the nature of the tasks proceed from an objective to a social frame of reference (as the tasks increase in ambiguity). That is, there will be an increase in conformity from Experiment I to Experiment V. The data relevant to this hypothesis are given in Tables 84, 85, and 86 and presented graphically in Figures 64, 65, and 66. 241 Mean Conformity Score 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 t Pig. Exp. I Exp. II Exp. Ill Exp. IV Exp. V 64. Mean confonmity scores (compliance) of the five experimental tasks (averaged over age, sex, and task difficulty). Range of scores also indicated. TABLE 84 MEAM COEPOHMITÏ SCORES (GOMELlARGE) OP THE PIPE EXPERIMENTAI. TASKS (AVERAGED OVER AGE, SEX AND TASK DIPPIGIILTI). P VALDE FOR MAIN EFPEGT OF EXPERIMENTAL TASK (TAKEN FROM ANALYSIS OF VARIANCE— TABLE 6i). Exp e riment al Task N Mean Score F (from Analysis of Variance) Exp. Task I 24 1.833 Exp. Task II 24 2.791 EXp. Task III 24 3.750 11.181* Exp. Task IV 24 5.416 Exp. Task V 24 6.041 *p < . 001 242 Mean Conformity Score Pi acceptance Exp. I Exp. II Exp. Ill Exp. IV Exp. V 65. Mean conformity scores (1st private of the five experimental tasks (averaged over age, sex, and task difficulty). Range of scores also indicated. TABLE 85 MEAN CONPORMITY SCORES (1st PRIVATE ACCEPTANCE) OP THE PIVE EXPERIMENTAL TASKS (AVERAGED OVER AGE, SEX AND TASK DIPPICULTI). P VALUE POR MAIN EPPECT OP EXPERIMENTAL TASK (TAKEN PROM ANALYSIS OP VARIANCE— TABLE 62). P Experimental Task N Mean Score (from Analysis of Variance) Task I Exp. Task II Exp. Task III Exp. Task IV Exp. Task V *p < .001 24 24 24 24 24 .833 1.500 2.208 3 .(52)4. 4.458 13.457 243 Mean Conformity Score 24 23 22 21 20 19 18 % 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Exp. I Exp. II Exp. Ill Exp. IV Exp. V Eig. 66. Mean conformity scores (2d private acceptance) of the five experimental tasks (averaged over age, sex, and task difficulty). ^ of scores also indicated. TABLE 86 MEAN CONPORMITY SCORES (2d PRIVATE ACCEPTANCE) OP THE PIVE EXPERIMENTAL TASKS (AVERAGED OVER AGE, SEX AND TASK DIPPICULTY). P VALUE POR MAIN EPPECT OP EXPERIMENTAL TASK (TAKEN PROM ANALYSIS OF VARIANCE— TABLE 63). Experimental Task N Mean Score P (from Analysis of Variance) Exp. Task I 24 .249 Exp. Task II 24 -416 Exp. Task III 24 .624 15.043* Exp. Task IV 24 2.124 Exp. Task V 24 3.1(3(5 *p < .001 244 Table 84 presents the mean conformity (C) scores of the five experimental tasks, averaged over age, sex and task difficulty, and indicates that the main effect of experimental task was significant (F = 11.181, p < .001). The difference in the mean conformity (C) scores of the five experimental tasks is illustrated in Figure 64. It is clear that the mean conformity scores increased from Experiment I to Experiment V. Table 85 gives the mean conformity (1st P.A.) scores of the five experimental tasks averaged over age, sex, and task difficulty, and indicates that the main effect of experimental task was significant (F = 15.457, P < .001). Figure 65 presents graphically the difference in the mean conformity (1st P.A.) scores for the five experimental tasks. It is evident that the mean conformity scores increased from Experiment I to Experiment V. Table 85 gives the mean conformity scores (2d P.A.) of the five experimental tasks, averaged over age, sex and task difficulty, and indicates that the main effect of experimental task was significant (F = 15*043, p < .001). The difference in the mean conformity (2d P.A.) scores of the five experimental tasks is illustrated in Figure 66. It is apparent that the mean conformity scores increased from Experiment I to Experiment V. Having found a significant main effect for experimental task in reference to compliance and both tests of private acceptance, Scheffe*s (1959) method of multiple 245 comparisons was employed to determine possible sources of the effect, and to determine if there was a significant increase in conformity between the experimental tasks. The results of the comparisons made are presented in Table 8?. Comparison 1 was aimed at determining if the mean conformity (C, 1st P.A., and 2d P.A.) scores of Experi mental Task I and Experimental Task II were significantly different. It can be seen from Table 87, that this difference was not significant in the case of compliance or for either test of private acceptance. Comparison 2 was made to determine if the mean conformity (C, 1st P.A., and 2d P.A.) scores of Experimental Task II and Experi mental Task III were significantly different. It is apparent from Table 87, that this difference was also not significant in the case of compliance or either tests of private acceptance. Comparison 5 was conducted to see if there was a significant difference between the mean con formity (C, 1st P.A., and 2d P.A.) scores of Experimental Task III and Experimental Task TV. This difference was significant in reference to compliance (E = 7*75, P < .001 ), initial private acceptance (E = 9*18, p < .001) and private acceptance after one week (F = 15.62, p < .001). Compari son 4 was directed at determining if the mean conformity (C, 1st P.A., and 2d P.A.) scores of Experimental Task IV and Experimental Task V were significantly different. The 246 < D < D m +3 m u < D % 5 0 Ü o O C M O L f\ CN \ r d C \J C D T - LT\Lf> C D L f\ < c ~ O T - C D +3 A * • o LTNO \ r • • O i > - o * • O 00 o o . • • O O O C M • 0 Ü II II Il V il V II Il V Il V Ü <4 Pa Pa pq p pq A pq pq A pq A j 0 4 - o 00 KN C D d O C M \ r V V Lf> K \ V LT\ v ro • O • O • O • o • O • o • O +3 C \J C \J C M C M o > o KN V C D o 0 ^ 0 v~ o A • • • • • L f\ . C M • 0 o -S Il V Il V Il V Il V Il V Il V Il V An A Pa A Pa A Pa A pq A pq A pq A 0 o O O L f\ - d " T- O c r \ d L f\ L f\ 0 -Ü > O Lf\Lf> C M V C M V 3 • o • o • O • • O • O • O ♦H C M \ r C M V i > - o c ^ o K N O 00 o f— 1 • • • • ^ • C M • Il V Il V Il V II Il V Il V Il V O O Pa A Pa A Pa A pq pq A pq A pq A 0 C Q 5 1 O |T \ KN E H E H P4 H + + k n] C M C M C M L f\ E H E H E H E H E H E H E H H M H + C M + |T \ + m • • • C M T - A A A A E H E H E H K H H H H M M M V C M -d " V *Lf\ E H E H E H E H E H E H E H • • • • • • • S I ' s g \r C \J | T \ L f\ VD CN P i a > d a o W)'H < D fd +3 a p s o b O T D <D »d +3 * r - D c d *d d rH +3 fH c d p< c d <D d A (D P4 d +3 & Ü A ^ O bO'H H î > CQ iH O o a + 3 *H I I I g >111 ! H H t> M H H H & Q ) I •H +3 " c d c d •H Ü 0 02 1 |i ^4 CQ CQ CQ CQ CQ Cd c d Cd cd c d E H E H E H E H E H I —Il—Il—Il—Il—I c d c d c d c d c d if if ü d d d d d < D 0 < D C D < D a a a a a •H *H "H *H tH A A A A A 0 0 0 0 0 247 difference was only significant for the second test of private acceptance (F = 7-50, P < .005). Comparison 5 was directed at determining if the mean conformity (C, 1st P.A., and 2d P.A.) score of Experimental Task I was significantly different from the average of the mean con formity (C, 1st P.A., and 2d P.A.) scores of Experimental Task II and III. This comparison was significant in the case of compliance (F = 7.51, p < .005) and in the first test of private acceptance (F = 6.36, p < .005). This comparison was not significant in the second test of private acceptance. Comparison 6 focused on ascertaining if the average of the mean conformity (C, 1st P.A., and 2d P.A.) scores of Experimental Tasks, I, II, and III differed significantly from the mean conformity (C, 1st P.A., and 2d P.A.) score of Experimental Task V. This difference was found to be significant in the case of compliance (F = 43.20, p < .001), initial private ac ceptance (F = 59.53, P < .001), and private acceptance after one week (F = 78.57, P < .001). Comparison 7 was concerned with determining if the average of the mean conformity (C, 1st P.A., and 2d P.A.) scores of Experi mental Task I, II, and III differed significantly from the mean conformity (C, 1st P.A., and 2d P.A.) score of Experimental Task TV. This difference proved to be sig nificant for compliance (F = 28.29, P < .001), initial private acceptance (F = 21.59, P < .001) and private 248 acceptance after one week (F = 30.01, p < .001). Other Significant Effects The combined analyses of the five experiments also indicated a significant age by task difficulty interaction effect for compliance (F = 21.724, p < .001— Table 61), initial private acceptance (F = 13.431, p < .001— Table 62), and private acceptance after one week (F = 4.897, p < .005— Table 63). That is, with increases in task difficulty conformity (C, 1st P.A., and 2d P.A.) scores increased for both young and old subjects; however, the increase at each level of difficulty was greater for the I ! old subjects. I I The combined analyses of the five experiments indicated a significant task difficulty by experimental task interaction effect for compliance (F = 2.634, p < .01 — Table 61), initial private acceptance (F = 2.037, p < .05— Table 62) and private acceptance after one week (F = 4.395, P < .001— Table 63). That is, with increases in task difficulty conformity (C, 1st P.A., and 2d P.A.) scores increased for all experimental tasks; however, the increase in conformity at each level of task difficulty was greater from Experiment I (visual perceptual judgment) to Experiment V (social attitude). The combined analyses of the five experiments also indicated a significant age by experimental task 249 interaction effect for the first test of private acceptance (F = 2.583, p < .05— Table 62) and for private acceptance after one week (F = 2.591, P < #05— Table 63). That is, there was an increase in conformity from Experimental Task I to Experimental Task V for both age groups; however, this increase in conformity was greater for the older subjects. In order to obtain a clearer picture of the effects operating in each of the five experiments, the mean con formity scores (C, 1st P.A., and 2d P.A.) for the two age groups and for the sexes at each level of task difficulty are given in Tables 96 through 110 in the Appendix, pp. 309-523. Postexperiment Questionnaire The postexperiment questionnaire (presented in the Appendix, p. 294) yielded qualitative data on the subjects participating in each of the five experiments. This data was combined and is presented in Table 88. The questions that exhibited different responses from the two age groups, or that are of interest in themselves, are shown in Table 88 along with the frequency of the various responses. The male and female subjects responded similarly to the questions. However, the frequency of the responses of the male and female subjects will also be shown for the questions that demonstrated an age difference. Questions which required a rating on a scale, which demonstrated an age difference, are also presented in 250 TABLE 88 RESPONSES GIVEN IN THE POSTEXPERIMENT QUESTIONNAIRE (DATA COMBINED FROM THE FIVE EXPERIMENTS) Question 1 ; How confident of your ratings were you? Age Sex Response young old male female Extremely confident 22 11 15 18 Somewhat confident 50 58 55 55 Not confident 8 11 10 .. ..2.. . Question 5: When you gave a response that differed from the other members of the group, who do you suppose was right? Response Age young old Sex male female I was always right 26 12 20 18 I was sometimes right 50 41 54 57 I was never right 4 7 6 5 Question 4: Did you ever respond opposite to your true feelings? Age Sex Response young old male female yes 18 21 17 22 no 42 59 45...... _ 58 Question 5: Did you consider the ratings of the other group members in making your rating? Age Sex Response young old male female yes 25 27 52 28 no 57 55 58 52 Question 6: Did you notice a divergence between your responses and those of the group? Age Sex Response young old male female yes 60 60 60 60 no 0 0 0 0 TABLE 88 (continued) 251 Question 9 Question 10 Question 11 Question 17 Question 20 Explain what you think the experiment was about— Response Age Sex male female Not aware 56 45 58 41 Somewhat aware 24 17 22 19 Completely aware 0 0 0 0 How would you rate your parents on their attempts to dominate you as a child? Parents never dominate 1, Parents extremely dominating 10. Age young old mean score 4.15 7.52 How would you rate your parents in regard to letting you solve your own problems as a child? No restraints 1, Extremely restraining 10. Age ___________young old mean score 5.97 7.41 How would you rate your parents on their attitude toward you? Accepting 1, Rejecting 10. Age ___________young old mean score 5.25 6.41 How would you rate yourself on cautiousness? Not cautious at all 1, Very cautious 10. Age ___________young old mean score 2.77 7.29 252 TABLE 88 (continued) Question 21 : How would you rate yourself on the way you approach problems? Actively 1, Passively 10. young old sc“ e 5.18 6.22 255 Table 88. Responses to Question 1 were classified as "ex tremely confident," "somewhat confident," and "not con fident." "Extremely confident" responses were those in which the subject expressed complete assuredness of the accuracy of his responses. "Somewhat confident" responses were those in which the subject expressed some doubt as to the accuracy of his responses. "Not confident" responses were those in which the subject expressed un certainty as to the accuracy of his responses. Responses to Question 5 are self-explanatory. It is apparent from Table 88, that the responses to Questions 1 and 3 of the older subjects were more often self-doubting than those of the younger subjects. These results indicate that the older individuals were less certain of the accuracy of their responses than were the younger individuals. It is evident from the responses given to Questions 4 and 5, that both age groups generally did not acknowledge their conformity behavior. The difference in response to these questions, between the two age groups, was relatively small compared to the difference in conformity expressed by the old and young subjects. Therefore, it can be con cluded that the older subjects were even less willing to admit to their conformity, or less aware of their con formity, than were the younger subjects. 254 Eespouse8 to Question 9 were classified as "Not aware as to the nature of the experiment," "somewhat aware,^ and "completely aware." "Not aware" responses were those in which the subject did not express any awareness as to ! what the experiment was actually concerned about, or the deception procedures employed. "Somewhat aware" responses were those in which the subject expressed a general aware ness as to the purpose of the experiment, but did not realize that there was any deception actually taking place. "Completely aware" responses would have been those where the subject not only had an awareness of the nature of the experiment, but also where the subject was aware of the deception procedures. It is clear from Table 88, indicated by the responses to Questions 5 and 9, that the deception pro cedure, the apparatus and the situation were found to be credible by the subjects of both age groups. The responses to Question 9 indicated that the older subjects were somewhat more deceived and completely unaware of the nature of the experiment and the deception procedures. Responses to Questions 10, 11, 17, 20, and 21 were on a 10-point scale. It can be seen from the responses given to Questions 10, 11, and 17 that the mean scores for these questions were substantially higher for the older subjects. It is apparent that the older subjects rated their parents more dominating, restraining and rejecting 255 than the younger subjects rated their parents. It is clear from the responses given to Question 20 that the older subjects considered themselves to be more cautious. The responses given to Question 21 indicate that the older subjects felt that they approach problems more passively, whereas, younger subjects felt they approached problems more actively. CHAPTER IX DISCUSSION In each of the five experiments of this investi gation, one outcome consistently recurred. When an individual was confronted by social influence, his judg ments and opinions were affected. A subject may have had an unequivocal judgment about an object when tested alone; but as soon as a group and its concomitant social influence were present he ceased to behave solely on the basis of his own perceptions or opinions. A more specific assertion can be made about this responsiveness to social influence: in different ways and in varying degrees, the majority of subjects tested on the various experimental tasks employed showed a "shift toward the group position. " The major hypothesis of this investigation was that I jolder individuals are more conforming than younger indi viduals. It was further posited that this hypothesis would hold in all five of the experiments involving a variety of tasks. When conformity was viewed in terms of compliance, older individuals were found to be significantly more conforming than younger individuals in four of the five experiments conducted. In the remaining experiment, the 256 257 difference was nearly significant at the .05 level of significance. The experiments demonstrated a disparity ■between the compliant behavior of the young and old indi viduals on a variety of experimental tasks. Collectively, the results of this investigation argue against the position that greater conformity by the older individual is "task specific." Rather, the results of this investi gation lend support to the position that there is a "general tendency" for older individuals to be more con forming than younger individuals regardless of the experi mental task. The results of this investigation, pertaining to the tests of private acceptance, require a somewhat dif ferent interpretation. The older subjects were not found to be significantly more conforming than the younger sub jects on the two tests for private acceptance in Experi ments I, II, and III. In Experiment IV, the older subjects were found to be significantly more conforming than the younger subjects on the initial test for private acceptance. In Experiment V, the older subjects were found to be more conforming than the younger subjects on the initial test of private acceptance, and on the test of private acceptance one week after the initial testing. It should be pointed out that the older individuals were found to be more con forming on both tests of private acceptance, in every experiment, although this difference did not always reach 258 statistical significance. Therefore, these results, al though not to the same degree as the results on compliance, lend support to the position that there is a general tendency for the older individual to be more conforming than the younger individual regardless of the experimental task. The results indicate that on tasks of perception when the group was no longer present, and when the social I influence was removed, the age difference in conformity ' behavior was not as marked. It may have been the un ambiguity of the perceptual tasks that caused a relatively small amount of private acceptance by both age groups. In the past other investigations utilizing an unambiguous ! ! task have generally found only a minimal degree of private acceptance (Asch 1956; Luchins & Luchins 1955). However, there have been a few investigations that have reported 1 private acceptance on perceptual tasks (Bovard 1955; Luchins & Luchins 1961). The results of this investigation indicate that on tasks of problem solving and social attitudes, the age difference in conformity behavior is still very apparent even after the social influence has been alleviated. It appears that when the stimulus of the task is sufficiently ambiguous to allow a variety of responses, or where there is no "correct” response, it is very unlikely that only compliance occurs. Past experimentation has found that on 259 unambiguous tasks there is usually private acceptance in addition to compliance (Bovard 1948; Hardy 1957; Israel 1965; Rohrer et al. 1954; Sherif 1955). Implications of age findings.— If a high value is to be placed on the ability of an individual to resist social influence which is in opposition to his individual judgments or opinions, then the implications of the afore- stated results are not especially optimistic. It appears that regardless of the task, the older individual is more allowing than the younger individual to be influenced by others. This tendency, for the old to show a greater susceptibility to social influence, was found to be most visible in the case of social attitudes. It is in this domain that the practical implications of conformity behavior are most salient. Therefore, the data obtained from this investigation, and from data obtained through future research on conformity, should be examined for practical implications. For example, having found that the older individual is more conforming, future investi gations should be directed at determining if some of the behavior patterns attributed to older individuals (e.g., disengagement) are due to their conformity behavior. In addition, perhaps the older individual’s conformity behavior could be modified when it is found to be in conflict with his needs, interests, rights and desires. In this way, the older person would cease to be targets 260 for advertisement, demagoguery and con games. Through future research, it may be found that the greater conformity by the older individual accounts for the disparity between what the older person wants and needs for his environment and society and what he does about obtaining these things. If this were found to be true, ways of encouraging the older individual to be less conforming and more active in striving for his needs and rights could be developed. The question arises, therefore, as to the methods one would employ to decrease the older individual's susceptibility to social influence. The available experi mental evidence lends itself to the interpretation that one of the primary reasons for susceptibility and con formity is inadequate individual knowledge and understand ing. An inference from this, is that resistance to social influence can be attained by insuring that the older individual is well informed and educated with respect to facts, opinions, et cetera, where he has a concern. Coleman et al. have said, "... susceptibility to con formity pressures can be decreased by increasing an individual's ability to make a competent selection of responses." The present investigation has provided a great deal of data germane to the question— are older individuals more conforming than younger individuals? Future research 261 should be devoted to answering the questions pertaining to the aforementioned social issues which have broad social implications. are older individuals more conforming?— The question arises as to why older individuals appear to be more conforming than younger individuals. Both young and old subjects had identical direct access to the stimulus material and only indirect knowledge of the experiences of others. When they faced a task alone, in any of the ex periments, they performed similarly; neither group expressing the slightest need for corroboration of their judgments or opinions. Why then was there a divergent trend in the conformity behavior of the two age groups? A multiplicity of causes may serve to answer this question, The answer is complex, but despite its complexity a better understanding of why older individuals were found to be j more conforming can be attained by considering some of the I theoretical explanations of conformity behavior. One explanation of conformity behavior can be derived from Festinger's (1957) theory of cognitive dissonance or the theories of cognitive consistency (Heider 1946, 1958; Osgood & Tannenbaum 1955)* That is, disagreement with others creates dissonance (Festinger 1957; Brehm & Cohen 1962) or imbalance (Heider 1958). Theoretically, the states of imbalance or dissonance create tensions within the individual which generates 262 forces to reduce dissonance or restore balance. One of the ways to reduce this dissonance or to restore balance is to change one's opinion in the direction of that held by the group. The possibility exists that the older individual is more disturbed by states of dissonance or imbalance. If this were true, then the greater conformity by older indi viduals would be a logical result. That is, in a con formity situation, the older individual experiencing a stronger motivation to reduce dissonance or restore balance, would be more likely to conform in order to obtain the more comfortable state. Another explanation of conformity behavior centers around the socialization process and learning theory. During the period the child is experiencing the sociali zation process, he is rewarded for imitating others and is frequently punished for trying to be different (Kidd &, Campbell 1955)- As the individual continues through ele mentary school, high school, college and beyond school a great deal of his behavior is likewise shaped by punishment and reward. Through stimulus generalization and secondary reinforcement, imitative behavior may become rewarding in and of itself. Bandura (1962, 1965) has taken a parallel view. He presents an abundance of evidence indicating that imitative behavior may be learned according to the postulates of learning theory. A number of investigators (Estes 1944; Dinsmoor 1954^ 1955; Howrer I960) have 265 considered the punishment situation as a type of avoidance learning. They have postulated that the punishing stimulus is a negative reinforcer which produces anxiety on the part of the subject. Campbell (1961) points out that the more an individual is rewarded for nonconformity, the less he will conform; and the more he is punished for nonconformity, the more he will conform. According to Howrer*s theory, in a conformity situation the individual who has been punished for nonconformity would conform. That is, he would inhibit nonconformity tendencies in order to avoid anxiety that would accompany expressing this type of behavior. The older individual of today may have experienced a greater degree of reward for conformity behavior and punishment for nonconformity behavior than today's youth. Therefore, the older individual would be more likely to conform in a present conformity situation. For example, today's older individual lived through the depression of the 1950's during which time he had to conform if he was to retain I ! his job. Today, stress is placed on being individualistic. This interpretation implies that the differences found in each of the conformity experiments of this investigation may have actually been a cohort or generational difference. Although an eclectic approach has been taken here, this explanation presently appears to be most promising. One of the objectives of future research, in the area of age and conformity, should be devoted to this question (Klein 264 1970). Another explanation of conformity behavior stems from considering the side effects of a conformity situ ation. There is evidence that a centrally aroused state, as measured by a variety of physiological indicators, and in part perhaps reflecting a state of anxiety in the individual, is created by social influence situations. There is also evidence that conforming in the social influ ence situation serves to decrease the centrally aroused state (Allen & Crutchfield 1965; Bogdonoff et al. 1961; Hoffman 1957)» Perhaps the older individual experiences a greater degree of anxiety in a conformity situation than I the younger individual. If this were the case, then by whatever means the conformity responses elicited by the older individual would become strengthened since they would serve to reduce the anxiety being experienced. This view would explain the general tendency of the older indi vidual to be more conforming than the younger individual regardless of the experimental task. That is, it is the older individual's motivation to reduce the anxiety experi enced in any conformity situation that causes him to be more conforming. The literature which indicates that the older individual is more cautious than the younger individual (Korchin & Basowitz 1956; Rees & Botwinick 1971; Silverman 1965 ; Wallach & Kogan 1961; et cetera) could also serve to 26^ explain the greater conformity expressed by the older individual. That is, their cautiousness might dispose them to conformity in order to avoid errors. Conformity is the safe and more cautious mode of behavior than is acting independently. Weiner (1958, 1959) has indicated that there is a significant positive relationship between un certainty and conformity. Mann (1965) has illustrated six studies which demonstrated a high relationship between conservatism and conformity. The older individual, due to his cautiousness and uncertainty, may turn to others for cues as to how to respond, which would lead to conformity. Another explanation of conformity, that could account for the age differences found in this investi gation, relates to a developmental theory of small groups. A developing theory in social psychology considers the "economic man." That is, when two or more people interact, each will behave in a manner which will maximize their rewards (payoffs) and minimize their punishments (costs) (Homans 1961). Thibaut and Kelley (1959) maintain that almost any behavior is both rewarding and costly. They emphasized the point that it is the balance of reward and cost for a given behavior, in comparison with the reward- cost balance for a potential alternative behavior, that determines which behavior will be expressed. In a con formity situation, a conformity response may be both rewarding and costly. The primary cost of conforming would 266 be responding opposite to how one actually believes. The rewards of conforming would include feeling a part of the group, not appearing strange or different, et cetera. Therefore, the older individual, having higher affiliation needs (Koponen 1957; Spangler & Thomas 1962), would be expected to conform more readily in order to attain these rewards (i.e., satisfy his affiliative needs). In other words, whereas the costs of conforming appear to be the same for young and old, the rewards for conforming seem to be greater for the older individual (Klein 1970). The psychoanalytic literature offers another possible explanation of conformity behavior. The origin of conformity may rest in the individual’s childhood experi ences with parents who are punitive, strict and dominating. The child’s experiences and attempts to gratify pleasurable impulses are punished. He is prevented from exploring the environment at his own rate. Bather, he is made to behave in the way in which his parents feel appropriate, even though it may be foreign to his own needs and beyond the scope of his ability. This frustrating parental behavior causes the child to make aggressive responses, whether directed toward the parents or displaced. These responses are punished in such a manner as to create anxiety and feelings of guilt over the aggressive impulses. This guilt is nurtured by the fact that it is these frustrating parents who care for the child and who are his main source 267 of support. To avert these unpleasant guilt feelings the child represses his aggressive impulses. The individual develops modes of behavior designed to foster repression and thus alleviate this threat. Conformity may result as one such behavior mode. That is, the individual may develop a very positive attitude toward, and a very strong need to conform to all forms of authority. If authority can be generalized to include the norms of the immediate group in a conformity situation, there exists a basis of a guilt-motivated need to conform (Hoffman 1953)• Based on the results of the questionnaire administered in this investigation, it was the older subjects who appeared to be reared in the authoritarian manner described above. That is, the older subjects considered their parents more restricting and dominating than the younger subjects con sidered their parents. Therefore, according to the psycho analytic explanation of conformity behavior, it would follow that the older individual would be more conforming. This explanation also leads to the interpretation that ; the age differences found in this investigation may actually be due to cohort or generational differences. A final explanation of conformity behavior can be derived from the theory of social comparison, first pro posed by Hestinger (195^)* There is a proposed "motivation to be correct." The judgment of others is taken as evidence to be considered in arriving at one’s own judgment. 268 According to Secord and Backman (1964), the need to vali date one's opinion is a basic human requirement. They point out that, although clear information from the physical environment contributes to satisfying this need, the expressed behavior of other individuals also provides a source of validation. Secord and Backman (1964) point out that. Particularly in situations where he is confused— where he does not know how to react— a person can turn to the behavior of other persons to observe a stable world. This social reality provides him with a reference point for his own behavior. The conformity situations of this investigation typified this type of situation, especially when the stimuli were highly ambiguous (High Task Difficulty situation). Eosow (1967) suggests that older individuals are more disturbed by ambiguity than younger individuals. Therefore, it could be inferred from the above argument that older per sons, as compared to younger persons, would seek the judgments of others more readily in an ambiguous conformity situation. This would lead to greater conformity by the older individual. The result found in a majority of the experiments conducted, which indicated that the increase in conformity scores as task difficulty increased was disproportionately greater for the older subjects, supports this interpretation. The studies concerning an increase in cautiousness with age, mentioned earlier, also support the above explanation of conformity behavior. Also the 269 results of the questionnaire administered in this investi gation, which indicated that the older subjects were less confident of their responses, would lend support to this explanation. In an ambiguous situation, the older indi vidual being more cautious and uncertain would seek more information in order to reduce his hesitancy toward responding. In this case, the information available is the judgments of the other subjects. The question also arises as to why the older indi vidual appears to be more privately accepting of the group norm. The answer may come from a consideration of an interesting variable in studies of social influence. This variable pertains to the subject's "commitment" to a particular position. When an individual makes a judgment or expresses an opinion when confronted by social influ ence, he has made a decision. That is, he has either yielded to the group or has remained independent. Whichever alternative he chooses, he may feel committed to that decision. Cognitive dissonance theory (Hestinger 1957) would posit that after a decision is made, the attractive ness of the chosen alternative (e.g., to conform) increases and the attractiveness of the unchosen alternative de creases. Gerard (1965) stated, in reference to the issue of commitment, that "The initial decision, then, commits the subject, although not irrevocably, to yield to the group or be independent whenever the same two alternatives 270 present themselves on succeeding trials." In sum, this reasoning implies that a group member who has committed himself to a given position may maintain that position for some time. In this investigation the older subject was found to make the decision of "compliance" more often than the younger subject. Therefore, based on^the above dis cussion of commitment, it was not surprising to find that the older subject generally privately accepted the group norm to a greater degree than the younger subject. Sex It was a hypothesis of this investigation that females would be more conforming than males. It was further posited that this hypothesis would hold in each of the five experiments on a variety of tasks. The results of each of the experiments conducted did not reveal a significant difference between the conformity behavior (either com pliance or private acceptance) of males and females, as expected. These results are in opposition to earlier findings (Applezweig & Moeller 1955; Asch 1956; Beloff 1958; Crutchfield 1955; Tuddenham, MacBride & Zahn 1958; and others) that found females to be significantly more conforming than males. These results do support the findings of Phelps (1966) and others, who did not find a significant sex difference in conformity. Contrary to expectations, males were found to be more conforming than 271 females in Experiment II and III of this investigation. This result supports the finding of the field study of Kahanna and Coe (1969) who also found males to he more conforming. It is interesting that although the sex dif ference was not significant in any of the experiments, it was in the experiments concerned with auditory tasks that resulted in the male subjects being slightly more conforming. The results of this investigation do not provide support for the view that there is a general tendency for females to be more conforming than males. Bather, these results substantiate the position that there is not a significant difference between the conformity behavior of males and females regardless of the task employed. This result could have various implications when considered together with the explanations of conformity presented above. For example, males and females may not be dif ferentially disturbed by states of dissonance or imbalance created in a conformity situation (Festinger 1957; Brehm & Cohen 1962); the act of conforming, by males and females, may not result in a differential decrease in the centrally aroused state created by a conformity situation (Allen & Crutchfield 1965; Bogdonoff et al. 1961 ; Hoffman 1957); the childhood experience of individuals, that may lead to a later life behavior pattern of conformity (Hoffman 1955), may not have been different for males and females ; and 272 based on the theory of social comparison (Festinger 1954; Secord & Backman 1964), males and females may not posses differing needs to validate their opinions in uncertain situations. Task Difficulty It was a hypothesis of this investigation that there is greater conformity when the difficulty of the task increases. It was further posited that this hypothesis would hold in each of the five experiments on all of the experimental tasks employed. The results of each of the five experiments, pertaining to both compliance and private acceptance, confirmed this hypothesis. These results corroborate the earlier observations of Asch (1951), Chipman (1966), Hunter (1968) and others, who have found conformity to increase as the stimulus of the task becomes more ambiguous. Generally, these results indicate that if an individual is certain of the correct answer, he is more able to resist influence being exerted by being able to respond on the basis of internal cues. It is worth noting, that these results fit well with Festinger's (1954) theory of social comparison. The theory postulates that we turn to others to validate our opinions. The more, difficult the task, the less able we are to rely on the physical reality; consequently the more likely we should turn to the opinions of others in order to validate our own. It is interesting to note, that the effect of task difficulty reached the 275 highest level of significance of all the effects tested in each of the five experiments. It also reached the highest level of significance in the analyses based on the combined data of the five experiments. Nature of the Task The hypothesis of this investigation, that there is an increase in conformity as the nature of the tasks proceed from an objective to a social frame of reference, was given support from the results which combined the data of the five experiments. It was found that conformity in creased from Experiment I to Experiment V. The subjects tested on the tasks that were concerned with perceptual judgments were found to be most resistant to conformity pressure; whereas, the subjects tested on the task of social attitudes were found to be most subject to social influence. This finding can also be explained by Eestinger’s (1954) social comparison theory. The theory postulates that individuals are "motivated to be correct." However, on the task that involved a social frame of reference (Experiment 7), as opposed to the tasks involving an objective frame of reference (Experiments I, II, and III), there is no way to be "correct." The issue of right or wrong does not pertain to social attitudes, but rather "correctness" is defined by others. It was not surprising, therefore, that on this task the subjects turned to others more readily in order to validate their 274 opinions. The results of the post hoc comparisons made between various combinations of experimental tasks provided interesting outcomes. It was found that although the degree! of conformity expressed by the subjects increased from Experiment I to Experiment II, there was not a significant difference in conformity between these two perceptual tasks. This indicated that tasks that are similar in nature (e.g., judgment tasks) produce similar degrees of conformity. It was also found that there was not a sig nificant difference between the auditory tasks (Experi ments II and III). However, there was a significant dif ference between the average of the auditory tasks and the visual task. This indicated that the sense modality with which a task is concerned can result in varying degrees of conformity. It also indicated that tasks that differ in nature (e.g., in sense modality) produce different degrees of conformity. The task of problem solving was found to be significantly different from the auditory detection task and also significantly different from the average of the three perceptual tasks. In the task of problem solving, the correct answer is never presented to the subject; whereas, in the perceptual tasks the true state of affairs is presented. Significant results were found when comparing the social attitudes task to the average of the three perceptual tasks. It is clear that 275 was important in reference to the degree of conformity expressed by the subjects. There was also a significant difference between Experiment IV (problem solving) and Experiment V (social attitudes) on the test of private acceptance after one week. It appears that subjects tested on tasks with a social frame reference are not only most subject to social influence, but they continue to retain the group's norm even after the influence has been alleviated over a period of time. Extremeness of the Norm In Experiment II and IV, the discrepancy of the contrived background report was investigated as a tangential issue. It was found that conformity (compliance and pri vate acceptance) increased significantly as the reported responses of the contrived group approached the correct answers. In Experiment II (auditory judgment) conformity was greatest when the contrived report of the group was discrepant by one click from the actual number of clicks sounded. Conformity decreased linearly as the contrived report of the group became discrepant by two, three, and four clicks. These results generally confirm the results of studies in the past which were concerned with this issue. Hovland and Pritzer (1957) and Pestinger (1957) have hypothesized that subjects may discredit influence- which is too extreme. This position has been supported by the findings of Asch (1956) and Johnson (1964). Festinger 276 (195^) states, "The tendency to compare oneself with some other specific person decreases as the difference between his opinion or ability and one's own increases." There have been some contradictory findings, however. 01mstead and Blake (1955), using a metronome click counting task, varied the contrived group norm by one, two, or three clicks from the actual answer. Generally, it was found that there was most conformity when the contrived report was discrepant by one click. However, the lowest amount of conformity was found at the middle level of discrepancy (two clicks). This result was not confirmed by the results of an investigation by Schroder and Hunt (1958) or by the results of Experiment II of this investigation. Goldberg (1954) and Tuddenham (1961) have found greater conformity to highly discrepant group norms than to group norms that are moderately discrepant. The results of the studies, including the present one, that have found more conformity as the group norm approaches the actual answer, can perhaps be explained in terms of "assimilation" and "contrast" effects (Hovland, Harvey & Sherif 1957). That is, when the group norm is slighltly discrepant, subjects tend to see their own position as closer to that of the others than it is; when the group norm is highly discrepant, subjects tend to see their own position as more distant from the others than it is. It may also be that conformity to near and distant 277 group norms involve different psychological processes. That is, when the contrived discrepant reports are small, the subjects' concern and conformity may be primarily related to "obtaining the correct answer." However, when the contrived discrepant reports are distant from the actual answer, the subjects' concern and conformity may be related to "not appearing different from the group members." Olmstead and Blake (1955) and Schroder and Hunt (1958) have considered this possibility. General Tendency vs. Trait The hypothesis of this investigation, that older individuals are more conforming than younger individuals regardless of the experimental task, was confirmed (re ported above). That is, the older subjects of this investigation were found to be more conforming (compliant and privately accepting) than the younger subjects in each of the five experiments including a variety of tasks. These results lend support to the studies that have found a generality of conformity (Blake, HeIson & Mouton 1956; Harper & Tuddenham 1964; Nakamura 1958; Rosner 1957; Sears 1965; Vaughn 1964), and they argue against those studies that have posited the specificity of conformity (Goldberg 1954-; Hollander & Willis 1967; Linton 1955)» The question arises as to whether these results demonstrate that the older individual possesses a personality "trait" of high conformity. The issue is remindful of the previous one 278 about leadership attributes (Cowley 1928; Mann 1959; Stogdill 1948; etc.). Social psychology was concerned with searching for the general traits of leaders. This search proved to be generally unavailing, which led to an arrant réévaluation of thought on the topic. Situational determinants of leadership— the group's task, its structure and other contextual features— then received the great majority of consideration for a period of time (Carter, Haythorn, Shriver & Lanzetta 1951 ; Cartwright & Zander I960; Hemphill 1949; Hollander & Julian 1967)» Presently, a more moderate position has been adopted by the majority of social psychologists. That is, stress continues to be placed on the importance of the situation, however, personal attributes (temperamental, physical, and intellectual) also receive recognition. Regardless of the issue, leadership or conformity, it may be better under stood by considering the interaction between the character of the individual and the situation. Hunt (1965) has said. Thus, it is neither the individual differences among subjects, nor the variations among situations that produce the variations in behavior. It is, rather, the interactions among these which are important. Although this investigation found the older indi vidual to be more conforming in five separate situations, there may be one or more situations, yet to be tested, that would find the young more conforming. To speak in terms of a "trait" of the old to be highly conforming__ 279 is perhaps premature. It seems more appropriate to consider these results as supportive of a "general tendency" for older individuals to be more conforming than younger individuals. Another reason for avoiding the use of "trait," at this point in research on conformity and age, is due to the myriad of settings outside the laboratory in which the relationship between age and conformity have yet to be investigated. Still another reason to temper the view of "traits" of conformity, based on the results of this investigation, stems from the possibility that the age differences found here may actually be better explained in terms of generational or cohort differences (this possibility was mentioned above.) Future Re se arch This investigation was an inquiry into a relatively small area of a wide region of social influence. Although the fact of social influence and a general tendency for the old to be more conforming than the young is beyond doubt, the understanding of the responsible processes has not been completely accomplished. One aim of future research, in this area, would be to explore the ways group behavior become forces in the psychological field of individuals. Further, to study the forces within individuals that cooperate with or withstand those actuated by the group environment. 280 The results of this investigation found that the young and old subjects not only complied to the group norm, but also privately accepted it on a number of occasions. The question as to what factors contributed to the adoption of the group norm will have to be answered by future research. One factor might be the negative evaluation, by the group members and the subject himself, for being incon sistent in his responses. Another factor may be that consistently responding as one did initially helps diminish post-decision dissonance. Even if these two factors, or other factors, were found to underly private acceptance, future investigation will have to consider if there is an interaction between these factors and age. Another area of future concern should center around the conditions of the conformity situation under which the response is given. It would be of special interest to determine if there is an interaction between age and the response conditions of the conformity situation. For example, Deutsch and Gerard (1955) studied conformity in two situations varying in amount of privacy. In one condition, subjects announced their judgments aloud; in another condition subjects responded by pushing a button and were prevented from seeing each other because of partitions. More conformity was found in the less private condition than in the more anonymous condition. Similar results have been found by Asch (1956), Mouton et al. (1956) 281 ' and Olmstead and Blake (1955)* It should be a concern of future research to ascertain whether the anonymity of the {conformity condition would have an effect on the relation ship between conformity and age consistently found in this investigation. In all five of the experiments of this investigation the response conditions were basically pri vate. That is, there were partitions between subjects and the subjects did not respond verbally. In a less pri vate situation, the age differences in conformity may prove to be similar or discrepant from those obtained here. The attraction the group holds for the individual is an interesting variable that has been investigated in conformity studies during the past thirty years. Generally, the results of these investigations have found greater con formity where the group is attractive to the subjects (Festinger 1950; Festinger 1952; Gerard 1954; Jackson & Saltzstein 1958; Lott & Lott 1961; Scott 1956). There have been a few investigations, however, that have not found attractiveness of the group to be a significant factor (Harper 1961 ; Wilson I960). The issue that arises for future investigation concerns the interaction between attractiveness of the group and age. That is, would varying the attractiveness of the group affect the conformity behavior of young and old individuals differently? In the present investigation no attempt was made to vary the attractiveness of the group. It remains a task for future 282 research to determine if the age differences found in Experiments I through V, of this investigation, would increase, decrease, or remain the same if group attractive ness was experimentally varied. Another issue, that has been empirically investi gated by a few researchers, deals with the size of the group in the conformity situation opposing an individual. Some investigators have found conformity to increase from one to three opponents, with this trend either reaching a plateau or decreasing with further additions to the size of the opposing group (Asch 1956; Rosenberg 1961). Other investigators have found the trend of increased conformity, with an increase in group size, to continue to five opponents (Gerard, Wilhelmy, & Conolley 1968). Still others have not found a significant relation between group size and conformity. It is clear that future research is warranted to clarify these contradictory findings. More germane to the primary issue of this investigation, the ifollowing question arises : What would happen to the age ! I relationship found in this investigation if the size of I the opposing group were varied? This is a question that remains to be answered through future research. Other variables that have been investigated in reference to conformity, but have yet to be investigated in terms of the interaction between these variables and age in relation to conformity, include: status of the group 285 members, self concept of the group members, cohesiveness of the group, competitiveness of the group, interdependence of the group, et cetera. Future research should ascertain the effect manipulation of these variables would have on the age-conformity relation observed in this investigation. Technological advancement and refinement in the measurement of conformity is another target area at which future research should be directed. This would include devising new types of conformity apparatus, experimental techniques and procedures and conformity situations. In order to further the work of pioneering investigators, such as Sherif, Asch and Crutchfield, future investigators will have to concern themselves with the development of innovations in the area of conformity research. Future research should also investigate, in greater detail, the processes underpinning conformity behavior. Cognitive processes, in particular, should receive the majority of this attention. For example, it would be of interest to determine if there is an age difference in the degree to which conformity is due to a process of cognitive dissonance reduction. The role of the experimenter is another aspect of conformity investigation that deserves future investi gation. The issue of "demand characteristics" (Orne 1959, 1962) in a social influence situation is one that has received little attention. The conformity found in social 284 influence situations may be partially attributed to the subject's perception (whether correct or not), of pressure being exerted by the experimenter. It would be of special interest to determine if "demand characteristics" are more or less intertwined in the conformity behavior of older individuals than in the conformity behavior of younger individuals. Finally, it should be the goal of future research to determine if the general tendency for the older indi vidual to be more conforming than the younger individual occurs across situations outside the laboratory. While a great deal of future research should continue in the laboratory, as discussed above, there also needs to be research which offers the experimenter the opportunity of controlling and manipulating variables under realistic operating circumstances. Ethical Concerns The question of ethics must be considered. There are definite ethical issues involved in psychological research when deception procedures are employed. The deception used in experimentation often results in mis informing the subjects, embarrassing them and even, per haps, harming them. Generally, there is a sound reason for deceiving subjects. In many cases if the subject knew the purpose of the investigation, the experimenter would obtain a distorted picture of the phenomena he was 285 concerned with. This would have been the case in the present experiment if the minimal degree of deception was not employed. However, even in these situations the ethical issues cannot be avoided. Kelman (1966) has stated, "How can we strike a proper balance between the interests of science and the considerate treatment of people who make themselves available as the raw material of research?" In the present investigation it was felt that alternative procedures would not have been as effective in the attempt to study conformity. Therefore, the problem was to obtain the "proper balance" to which Kelman referred. After the investigation was over, each subject in this investigation was fully informed regarding what actually took place and concerning the purpose of the experiment. These expla nations were designed to be meaningful and instructive. Not a single subject when questioned, after being informed of the deception, felt he had been ethically mistreated during the experiment. The majority of the subjects felt that the experiment was a valuable and educative experience. CHAPTER X SUMMARY AND CONCLUSION This investigation focused on determining whether the greater degree of conformity expressed by older sub jects in a visual perceptual judgment situation (Klein 1970) was unique to that situation, or whether it is a general tendency for the older person to be more con forming regardless of the experimental task. This investi gation also studied sex differences in conformity behavior and variation in conformity behavior due to ambiguity of the stimulus in a search for general tendencies of con formity behavior. As a secondary concern, this investi gation was directed at determining if the degree of conformity expressed varies due to the nature of the experimental task. In order to answer these questions, a series of five conformity experiments were conducted on a variety of experimental tasks. A review of the literature on conformity, directly related to each of the independent variables, was pre sented. Pour hypotheses were tested in the general investigation by measuring degree of social conformity in terms of compliance and private acceptance. The first portion of hypotheses 1, 2, and 5 were also separately tested in each of the five experiments of the general 286_ 287 inve stigation: The hypotheses were; 1) Informational social influence is greater upon the judgments or opinions of older individuals than upon the judgments or opinions of younger individuals. Purther, this hypothesis will hold in each of the sub-experiments on all five tasks employed. 2) Informational social influence is greater upon the judgments or opinions of females than upon the judg ments or opinions of males. Purther, this hypothesis will hold in each of the sub-experiments on all five tasks employed. 5) Informational social influence is greater upon the judgments or opinions of the subjects as the stimulus of the task becomes more ambiguous (i.e., as the task becomes more difficult). Purther, this hypothesis will hold in each of the five sub-experiments on all five tasks employed. 4) Informational social influence is greater upon the judgments or opinions of the subjects as the nature iof the experimental tasks proceed from an objective to a ! social frame of reference (i.e., there will be an increase in conformity from Experiment I to Experiment V. A group of sixty young subjects (17 to 24 years) were compared to sixty old subjects (60 to 81 years) re garding susceptibility to informational social influence. There were twenty-four subjects (half young and half old, half male and half female) in each of the five sub-experi ment s. Using one of three types of conformity apparatus (Klein 1970, Mathie 1959, Deutsch & Gerard 1955), the subjects in each of these experiments were subjected to contrived group pressures toward erroneous judgments or opinions. Experiments I through V were all interested in age, sex, and task difficulty as predictors of social 288 conformity. Below are listed the five conformity experi ments including the task performed in each. Experiment I: Visual perceptual judgment task— the subject's task was to judge which of two circular discs was largest in size. Experiment II. Auditory perceptual judgment task— the subject's task was to judge the number of metronome clicks heard. Experiment III. Auditory signal detection task— the subject's task was to report whether he heard a pure tone plus noise, or noise alone in his right ear while sometimes receiving additional information in his left ear. Experiment IV; Problem-solving task— the subject's task was to solve one-operation arithmetic problems. Experiment V; Social attitude task— the subject's task was to rate statements on nationalism as to whether he agreed or disagreed with them. The five experiments followed an identical research design, which provided measures of both compliance and private acceptance as two distinct measures of social conformity. The following describes the research design for each of the sub-experiments : First, individual per formance was evaluated under "alone" conditions. The subject was then tested in a "conformity" situation where he experienced contrived social influence. The subject's behavior was then again measured under the alone condition after the conformity pressures were relieved. Finally, the subject's behavior was again measured under the alone condition one week later. Analysis of variance designs were used in each experiment, and in the analyses combining the data of the 289 five experiments, to determine if there were significant effects present. The primary results of the analyses of each experiment, and of the combined analyses, are pre sented below: Experiment I: (a) Older subjects conformed (com pliance) significantly more often than the younger sub jects; (b) there was not a significant sex difference in conformity behavior; (c) conformity occurred significantly more frequently as task difficulty increased (compliance, 1st test of private acceptance). Experiment II: (a) Older subjects were more con forming than younger subjects, although, the difference was not quite statistically significant; (b) there was not a significant sex difference in conformity behavior; (c) conformity occurred significantly more frequently as task difficulty increased (compliance, 1st and 2d test of private acceptance); (d) conformity occurred significantly more often as the contrived report of the group approached the correct answer (compliance, 1st and 2d test of private acceptance). Experiment III: (a) Older subjects conformed sig nificantly more often than younger subjects (compliance); (b) there was not a significant sex difference in con formity behavior; (c) conformity occurred significantly more frequently as task difficulty increased (compliance, 1st and 2d test of private acceptance). Experiment IV: (a) Older subjects conformed signi ficantly more often than younger subjects (compliance, 1st test of private acceptance); (b) there was not a significant sex difference in conformity behavior; (c) conformity occurred significantly more frequently as task difficulty increased (compliance, 1st and 2d test of private acceptance); (d) conformity occurred significantly more often as the contrived report of the group approached the correct answer (compliance, 1st and 2d test of private acceptance). Experiment V: (a) Older subjects conformed signi ficantly more often than younger subjects (compliance, 1st and 2d test of private acceptance); (b) there was not a significant sex difference in conformity behavior; (c) conformity occurred significantly more frequently as task difficulty increased (compliance, 1st and 2d test of private acceptance). 290 Combined Analyses: (a) Older subjects were found to conform significantly more often than younger subjects when the data of the five experiments were combined, and older subjects were more conforming in each experiment (compliance, 1st and 2d test of private acceptance); (b) there was not a significant sex difference in conformity? based on the combined data; (c) conformity was found to occur significantly more often as task difficulty increased when the data of the five experiments were combined (com pliance, 1st and 2d test of private acceptance); (d) con formity occurred more frequently as the nature of the task preceded from an objective to a social frame of reference (i.e*, conformity increased from Experiment I through Experiment V) (compliance, 1st and 2d test of private acceptance). Some implications of the general tendency for older individuals to be more conforming than younger individuals were discussed. The question as to why older individuals are more conforming was raised. An attempt was made to better understand this finding by considering "age" in relation to some of the theoretical explanations of con formity behavior. Generational differences regarding reward and punishment and concomitant anxiety of conformity and nonconformity, was considered the most promising explanation. Other explanations included: conformity as a strategy of reducing the dissonance or "imbalance” created in a conformity situation; conformity as a strategy of decreasing the centrally aroused state and concomitant anxiety which is created in a social influence situation; a consideration of the "rewards" and "costs" of conforming as opposed to nonconforming; a psychoanalytic explanation; and a social comparison theory explanation where the individual turns to others to validate his opinions. 291 The use of traits versus general tendencies of conforming was discussed. The position was taken, that to speak in terms of a "trait" for the old to be highly con forming, would perhaps inappropriately finalize the re lationship. The results were considered to support the view that there is a general tendency for older individuals to be more conforming than younger individuals. Although this investigation demonstrated a general tendency for older individuals to be more conforming than younger individuals, age differences on a number of issues within the rubric of social conformity remain to be in vestigated. Suggestions for future research were offered including: determing if there is an interaction between age and the response conditions of the conformity situ ation, the attractiveness of the group, the size of the group, status of the group members, self concept of the group members, cohesiveness of the group, and inter dependence of the subjects in relation to conformity; the development of technological innovations in the area of conformity research; investigations of the processes, especially cognitive, underlying conformity behavior; determining if "demand characteristics" are more or less intertwined in the conformity behavior of older individuals than in the conformity behavior of younger individuals; and determining if the general tendency for the older individual to be more conforming occurs across situations 292 outside the laboratory. The ethical considerations of employing deception procedures in psychological research were discussed. Emphasis was placed on maximizing scientific output while minimizing the negative effects deception may have on the subjects. 293 APPENDIX POSIEXPEEJMENO? QUESTlONNAIEE (GIVEN IN EACH EXPERIMENT) Age_____ Sex 1. How confident of your ratings were you? 2. If something important depended on your ratings, if this were a matter of practical consequence, would have you responded in the way you did? 5. When you gave a response that differed from the other members of the group, who do you suppose was right? 4. Did you ever respond opposite to your true feelings? 5. Did you consider the ratings of the other group members in making your ratings? 6. Did you notice a divergence between your responses and those of the group? 7- Did you always keep the instructed task in mind? 8. Were you at all puzzled, or confused during any part of the experiment? If yes, in what way? 9. Explain what you think the experiment was about— On the questions below you will be asked to make ratings. Please make an X on the line in the place that best applies to your ratings. 10. How would you rate your parents on their attempts to dominate you as a child? parents never parents attempted -------------------- extremely to dominate dominating 295 11. How would you rate your parents in regard to letting you solve your own problems as a child? no restraints -------------------- ^Straining 12. Compared with individuals of your age and circum stance, how would you rate yourself on success in life? successful------------------------ unsuccessful 13. How would you rate your childhood family atmosphere? friendly and unfriendly and tolerant-------------------------- intolerant 14. How would you rate yourself on shyness? not shy extremely at all---------------------------- shy 15. How satisfied are you with your way of life? extremely extremely satisfied ■ -..- .—---— ....... dissatisfied 16. How happy a person do you consider yourself to be? an unusually an extremely happy person----------- unhappy persor. 17. How would you rate your parents on their attitude toward you? accepting --------------------- rejecting 18. How would you rate yourself on self consciousness? not self very conscious --------------------- self at all conscious 19. In general, how would you rate yourself as an individual? favorably --------------------- unfavorably 296 20. How would you rate yourself on cautiousness? not cautious very at all --------------------- cautious 21. How would you rate yourself on the way you approach problems? actively --------------------- passively 297 Description of Attitude Scale I The attitude scale utilized was the Worldmindedness I Scale developed by Sampson and Smith (1957). It is a 'Likert-type scale consisting of 32 items. The scale is I designed to measure nationalistic-internationalistic I attitudes. Sampson and Smith referred to it as a "social I attitudes questionnaire." The items appearing in the I final scale were selected from a pool of 60 items on the j basis of an item analysis. The items are arranged so that I I every eighth item pertains to the same dimension in the i following order: religion, immigration, government, I I economics, patriotism, race, education, and war. I The reliability of this scale is exceptionally I high. Based upon the responses of 56 college students, split-half reliabilities were found to be .93. A test- retest reliability of .93 was obtained for the students I I I after a twenty-eight day interval (Sampson and Smith 1937). This scale has also been used successfully by j Garrison (1961), Allman (1961) and by Smith (1935). I Sampson and Smith suggested that the internal consistency I I of the scale argued for its validity. However, they did provide some empirical evidence. Scale scores correlated “.71 with an 11-item form of the Ethnocentrism Scale (Levinson 1949). 298 TABLE 89 CONDITIONS OE VISUAL PERCEPTUAL JUDGMENT TASK— EXPERIMENT I Trial Level of Difficulty Actual Largest PiKure Eigure Reported as Largest 1 hedium-TD Eight Right 2 Low-TD Left Left *5 High-TD Right Left *4 Low-TD Left Right 5 High-TD Right Right *6 High-TD Right Left *7 Low-TD Right Left *8 High-TD Left Right 9 Medium-TD Eight Right ♦10 Medium-TD Left Right ♦11 Medium-TD Right Left 12 High-TD Left Left *13 High-TD Left Right ♦14 High-TD Left Right 15 Low-TD Left Left ♦16 Low-TD Left Right ♦17 Low-TD Right Left ♦18 Medium-TD Right Left 19 High-TD Right Right ♦20 Medium-TD Left Right ♦21 Low-TD Right Left ♦22 Medium-TD Left Right 25 Medium-TD Left Left ♦24 Low-TD Left Right ♦25 High-TD Right Left 26 Low-TD Right Right ♦27 High-TD Left Right 28 High-TD Left Left ♦29 Low-TD Right Left ♦50 Medium-TD Left Right ♦51 Medium-TD Right Left 32 Medium-TD Left Left *33 High-TD Right Left ♦54 Low-TD Left Right 35 Low-TD Right Right *56 ^ rnirs ^ A Medium-TD Right >/I T «I T c nrUi <n 4- Left 4 d 4-m4 n et /ivt which the contrived peer judgments were incorrect. All other trials were innocuous, where contrived peer judg ments were correct. Low-TD=radii of the circular discs in the ratio of 16 to 10« Medium-TD=radii of the circular discs in the ratio of 16 to 14. High-TD=radii of the circular discs in the ratio of 16 to 15. 299 Tektronix Type iG; C D -P CQ cd H U CQ Cd I—IA Tec. Type 161 < D -P CQ cd I—IA Q ) 4^ CQ Cd I—I A Noise Gen. Model 901B Gen. Radio ■ #1^13 Grason Stadler Model 829E Tek. Type 502 A Amp ex Ag 4-40 Figure 67 Equipment Set-Up Used in Making Auditory Stimulus— Experiment III Noise o s c Waveform Generator Electronic Switch Electronic Switch Tape Recorder Oscilloscope TABLE 90 CONDITIONS OF THE AUDITORY JUDGMENT TASK— EXPERIMENT II 500 Trial Level Difficulty Actual Clicks Sounded Contrived Majority Report Degree of Discrepancy 1 Medium-TD 22 22 0 2 Low-TD 26 26 0 *5 High-TD 25 21 2 *4 Low-TD 24 27 3 High-TD 28 28 0 *6 High-TD 27 25 2 *7 Low-TD 29 25 4 *8 High-TD 24 27 3 9 Medium-TD 24 24 0 *10 Medium-TD 24 27 3 *11 Medium-TD 29 25 4 12 Medium-TD 22 22 0 *15 High-TD 25 21 4 *14 High-TD 28 31 3 15 Low-TD 28 28 0 *16 Low-TD 28 31 3 *17 Low-TD 25 21 2 *18 Medium-TD 27 25 2 19 High-TD 26 26 0 *20 Medium-TD 26 27 1 *21 Low-TD 22 25 1 *22 Medium-TD 22 25 1 25 Medium-TD 28 28 0 ♦24 Low-TD 25 21 4 *25 High-TD 29 25 4 26 Low-TD 22 22 0 *27 High-TD 22 25 1 28 High-TD 24 24 0 *29 Low-TD 26 27 1 *50 Medium-TD 28 31 3 *51 Medium-TD 25 21 2 32 Medium-TD 26 26 0 *33 High-TD 26 27 1 *54 Low-TD 27 25 2 35 Low-TD 24 24 0 Medium-TD 25 4 ♦These designate "critical" trials ÿhat IS, trials on which the contrived peer judgments were incorrect. All other trials were innocuous, where contrived peer judgments were correct. Low-TD=Rate of click presentation was 158 per minute. Medium-TD=Rate of click presentation was 184 clicks per minute. High-TD=Eate of click presentation was 224 clicks per minute. TABLE 91 CONDITIONS OE THE AUDITORY TASK— EXPERIMENT III DETECTION 501 Trial Level of Difficulty Actual Auditory Presentation Contrived Majority Report 1 Medium-TD Noise + Tone Noise + Tone 2 Low-TD Noise Noise *3 High-TD Noise Noise + Tone *4 Low-TD Noise + Tone Noise 5 High-TD Noise + Tone Noise + Tone *6 High-TD Noise Noise + Tone *n Low-TD Noise Noise + Tone *8 High-TD Noise + Tone Noise 9 Medium-TD Noise + Tone Noise + Tone *10 Medium-TD Noise + Tone Noise *11 Medium-TD Noise Noise + Tone 12 High-TD Noise Noise *15 High-TD Noise + Tone Noise *14 High-TD Noise + Tone Noise 13 Low-TD Noise Noise *16 Low-TD Noise + 'Tone Noise *17 Low-TD Noise Noise + Tone *18 Medium-TD Noise Noise + Tone 19 High-TD Noise + Tone Noise + Tone *20 Medium-TD Noise + Tone Noise *21 Low-TD Noise Noise + Tone *22 Medium-TD Noise + Tone Noise 25 Medium-TD Noise Noise *24 Low-TD Noise + Tone Noise *25 High-TD Noise Noise + Tone 26 Low-TD Noise + Tone Noise + Tone *27 High-TD Noise + Tone Noise 28 High-TD Noise Noise *29 Low-TD Noise Noise + Tone *30 Medium-TD Noise + Tone Noise *51 Medium-TD Noise Noise + Tone 32 Medium-TD Noise Noise *33 High-TD Noise Noise + Tone *54 Low-TD Noise + Tone Noise 35 Low-TD Noise + Tone Noise + Tone *36 Medium-TD Noise Noise + Tone That is, trials on which the contrived peer judgments were incorrect. All other trials were innocuous, where contrived peer judgments were correct. Low-TD=No information presented in the left ear of the subject. Medium-TD=Digits presented to the left ear of the subject. The subject is instructed to ignore the digits. Hi^-TD=Digits presented to the left ear of the subject. The subject is instructed to pay attention to the digits. 502 TABLE 92 CONDITIONS OP THE PROBLEM SOLVING TASK (POP THE YOUNG SUBJECTS)— EXPERIMENT IV Level of Actual Reported Degree of Trial Problem Difficulty Answer Answer Discrepancy 1 7.4+9.9 Medium-TD 17.30 17.30 none 2 8.9-4.5 Low-TD 4.60 4.60 none *5 17.29+14.82 High-TD 52.11 51.11 mild *4 2.9-1.7 Low-TD 1.20 1.90 mild 5 82/4 High-TD 20.50 20.50 none *6 17.84-16.99 High-TD .85 .75 mild *7 4.8-5.2 Low-TD 1.60 2.60 high *8 92.4/7 High-TD 15.20 11.20 mild 9 24.2-7.1 Medium-TD 17.10 17.10 none *10 16x11 Medium-TD 176.00 193.00 high *11 148/4 Medium-TD 37.00 35.00 mild 12 21.4x8.2 High-TD 175.48 175.48 none *15 9.47+28.86 High-TD 38.53 45.33 high *14 40.5/4.5 High-TD 9.00 12.50 high 15 13x9 Low-TD 117.00 117.00 none *16 12x7 Low-TD 84.00 86.00 mild *17 98/7 Low-TD 14.00 15.00 mild *18 12.1+7.9 Medium-TD 20.00 21.00 mild 19 24.25+11.47 High-TD 35.70 35.70 none *20 18x12 Medium-TD 216.00 224.00 Mild *21 14x4 Low-TD 56.00 60.00 high *22 252/8 Medium-TD 29.00 24.00 high 25 29x5 Medium-TD 145.00 145.00 none *24 57/3 Low-TD 19.00 15.70 high *25 14.72-11.99 High-TD 2.73 4.88 high 26 4.5+2.5 Low-TD 6.60 6.60 none *27 37.14 High-TD 518.00 418.00 high 28 25.21-21.52 High-TD 3.89 3.89 none *29 1.7+2.5 Low-TD 4.00 5.00 high *50 15.8-8.9 Medium-TD 4.90 4.70 mild *51 19.4+7.8 Medium-TD 27.20 58.20 high 32 48.6/5 Medium-TD 16.20 16.20 none *33 18x15 High-TD 254.00 254.00 mild *54 1.7+9.2 Low-TD 10.90 11.70 mild 35 36/4 Low-TD 9.00 9.00 none *36 18.2-7.1 Medium-TD 11.10 10.50 mild *These designate "critical" trials. That is, trials on which the contrived peer judgments were incorrect. All other trials were innocuous. where contrived peer judgments were correct. Low-TD=:Problem answered correctly by more than 75% of the young standardizing group. Medium-TD=Problem answered correctly by less than 75% but by more than 25% of the young standardizing group. High-TD=Problem answered correctly by less than 25% of the young standardizing group. 305 TABLE 93 CONDITIONS OE THE PROBLEM SOLVING TASK (FOR THE OLD SUBJECTS)---EXPERIMENT IV Trial Problem Level of Difficulty Actual Answer Reported Answer Degree of Discrepancy 1 7.4+9.9 Medium-TD. 17.30 17.30 none 2 8.9'*“4« 5 Low-TD 4.60 4.60 none *3 8.8+7.6 High-TD 16.40 18.80 mild *4 2.9-1.7 Low-TD 1.20 1.90 mild 5 81/5 High-TD 27.00 27.00 none *6 11.8-2.9 High-TD 8.90 9.80 mild *7 4.8-5.2 Low-TD 1.60 2.60 high *8 96/6 High-TD 19.00 16.00 mild 9 24.2-7.1 Medium-TD 17.10 17.10 none *10 18x5 Medium-TD 90.00 60.00 high *11 112/8 Medium-TD 14.00 15.00 mild 12 16x11 High-TD 176.00 176.00 none *13 6.7+18.9 High-TD 25.60 52.80 high *14 148/4 High-TD 37.00 45.00 high 15 9x8 Low-TD 72.00 72.00 none *16 12x7 Low-TD 84.00 86.00 mild *17 64/4 Low-TD 16.00 18.00 mild *18 12.1+7.9 Medium-TD 20.00 21.00 mild 19 25.8+4.6 High-TD 28.40 28.40 none *20 13x9 Medium-TD 117.00 127.00 mild *21 14x4 Low-TD 56.00 60.00 mild 22 98/7 Medium-TD 14.00 22.00 high 25 29x5 Medium-TD 145.00 145.00 none *24 57/3 Low-TD 19.00 25.00 high *25 25.8-16.9 High-TD 6.90 10.00 high 26 4.5+2.5 Low-TD 6.60 6.60 none *27 18x12 High-TD 216.00 240.00 high 28 28.2-15.7 High-TD 12.50 12.50 none *29 1.7+2.5 Low-TD 4.00 5.00 high *30 15.8-8.9 Medium-TD 4.90 4.70 mild *31 19.4+7.8 Medium-TD 27.20 58.20 high 32 252/8 Medium-TD 29.00 29.00 none *33 18x15 High-TD 254.00 254.00 mild *54 1.7+9.2 Low-TD 10.90 11.70 mild 35 36/4 Low-TD 9.00 9.00 none *56 18.2-7.1 Medium-TD 11.10 10.50 mild which the contrived peer judgments were incorrect. All other trials were innocuous, where contrived peer judgments were correct. Low-TD=Problem answered correctly by more than 75% of the old standardizing group. Medium-TD=Problem answered correctly by less than 75% but by more than 25% of the old standardizing group. High-TD=Problem answered correctly by less than 25% of the old standardizing group. 504 THE WORLDMINDEDNESS SCALE (SAMPSON AND SMITH 1957) USED IN EXPERIMENT V 1. Our country should have the right to prohibit certain racial and religious groups from entering it to live. 2- Immigrants should not be permitted to come into our country if they compete with our own workers. 5. It would be a dangerous procedure if every person in the world had equal rights which were guaranteed by an international charter. 4. All prices for exported goods and manufactured goods should be set by an international trade committee. 5. Our country is probably no better than many others. 6. Race prejudice may be a good thing for us because it keeps many undesirable foreigners from coming into this country. 7. It would be a mistake for us to encourage certain racial groups to become well educated because they might use their knowledge against us. 8. We should be willing to fight for our country without questioning whether it is right or wrong. 9. Foreigners are particularly obnoxious because of their religious beliefs. 10. Immigration should be controlled by an international organization rather than by each country on its own. 11. We ought to have a world government to guarantee the welfare of all nations irrespective of the rights of any one. 12. Our country should not cooperate in any international trade agreements which attempt to better world economic conditions at our expense. 15. It would be better to be a citizen of the world than of any particular country. 14. Our responsibility to people of other races ought to be as great as our responsibility to people of our own race. 305 15* An international committee on education should have full control over what is taught in all countries about history and politics. 16. Our country should refuse to cooperate in a total disarmament program even if some other nations agreed to it. 17. It would be dangerous for our country to make inter national agreements with nations whose religious beliefs are antagonistic to ours. 18. Any healthy individual, regardless of race or religion, should be allowed to live wherever he wants to in the world. 19. Our country should not participate in any inter national organization which requires that we give up any of our national rights or freedom of action. 20. If necessary, we ought to be willing to lower our standard of living to cooperate with other countries in getting an equal standard for every person in the world. 21. ¥e should strive for loyalty to our country before we can afford to consider world brotherhood. 22. Some races ought to be considered naturally less intelligent than others. 25. Our schools should teach the history of the whole world rather than of our own country. 24. An international police force ought to be the only group in the world allowed to have armaments. 25. It would be dangerous for us to guarantee by inter national agreement that every person in the world should have complete religious freedom. 26. Our country should permit the immigration of foreign peoples even if it lowers our standard of living. 27. All national governments ought to be abolished and replaced by one central world government. 28. It would not be wise for us to agree that working conditions in all countries should be subject to international control. 506 29- Patriotism should he a primary aim of education so our children will believe our country is the best in the world. 50. It would be a good idea if all the races were to intermarry until there was only one race in the world. 51. We should teach our children to uphold the welfare of all people everywhere even though it may be against the best interest of our own country. 52. War should never be justifiable even if it is the only way to protect our national rights and honor. 307 TABLE 94 CONDITIONS OE THE SOCIAL ATTITUDE TASK (EOE THE YOUNG SUBJECTS)— EXPERIMENT V Level of Majority Report of Contrived Report Item Attitude Trial Difficulty Standardizing Group Given Ss Scale 1 Medium-TD Disagree Disagree - j 2 High-TD Disagree Disagree 2 3 High-TD Disagree Disagree 3 *4 High-TD Disagree Agree 4 *5 High-TD Agree Disagree 5 *6 Low-TD Disagree Agree 6 *7 Low—TD Disagree Agree 7 *8 Low-TD Disagree Agree 8 *9 Low-TD Disagree Agree 9 *10 Low-TD Disagree Agree 10 *11 High-TD Disagree Agree 11 12 High-TD Disagree Disagree 12 13 High-TD Disagree Disagree 13 *14 Medium-TD Agree Disagree 14 *15 Low-TD Disagree Agree 15 *16 High-TD Disagree Agree 16 *17 Medium-TD Disagree Agree 17 *18 Medium-TD Agree Disagree 18 19 Medium-TD Agree Agree 19 *20 High-TD Disagree Agree 20 *21 High-TD Agree Disagree 21 22 Low-TD Disagree Disagree 22 *25 Medium-TD Agree Disagree 23 *24 Medium-TD Disagree Agree 24 *25 Medium-TD Disagree Agree 25 *26 High-TD Disagree Agree 26 *27 Low-TD Disagree Agree 27 *28 High-TD Agree Disagree 28 *29 Medium-TD Disagree Agree 29 30 Medium-TD Disagree Disagree 30 *51 Low-TD Agree Disagree 31 *32 Medium-TD Disagree Agree 32 which the contrived peer judgments were incorrect. All other trials were innocuous, where contrived peer judgments were correct. Low-TD=90% (or more) of the standardizing group agreed or disagreed with the attitude statement. Medium-TD=Between 75% and 90% of the standardizing group agreed or disagreed with the attitude statement. High-TD=Between 55% and 75% of the standardizing group agreed or disagreed with the attitude statement. TABLE 95 CONDITIONS OE THE SOCIAL ATTITUDE TASK (EOR THE OLD SUBJECTS)--EXPERIMENT V 508 Contrived ïtem of Level of Majority Report of Report Attitude Trial Difficulty Standardizing Group Given Ss Scale 1 High-TD Disagree Disagree 1 2 High-TD Disagree Disagree 2 5 High-TD Agree Agree 5 *4 Low-TD Disagree Agree 4 *5 High-TD Disagree Agree 5 *6 Low-TD Disagree Agree 6 *7 Low-TD Disagree Agree 7 8 High-TD Disagree Disagree 8 *9 Low-TD Disagree Agree 9 *10 Low-TD Disagree Agree 10 *11 Medium-TD Disagree Agree 11 12 High-TD Disagree Disagree 12 *15 Low-TD Disagree Agree 15 *14 Medium-TD Agree Disagree 14 *15 Low-TD Disagree Agree 15 *16 High-TD Disagree Agree 16 *17 Medium-TD Disagree Agree 17 *18 Medium-TD Agree Disagree 18 *19 High-TD Agree Disagree 19 *20 High-TD Disagree Agree 20 21 High-TD Disagree Disagree 21 *22 High-TD Disagree Agree 22 *25 Medium-TD Agree Disagree 25 24 High-TD Disagree Disagree 24 *25 Medium-TD Disagree Agree 25 *26 High-TD Disagree Agree 26 *27 Low-TD Disagree Agree 27 *28 High-TD Agree Disagree 28 *29 High-TD Agree Disagree 29 *50 Medium-TD Disagree Agree 50 51 High-TD Disagree Disagree 51 Medium-TD Disagree Agree 52 * The se designate critical" trials. That is, trials on which the contrived peer judgments were incorrect. All other trials were innocuous, where contrived peer judgments were correct. Low-TD=90% (or more) of the standardizing group agreed or disagreed with the attitude statement. Me d ium-TD=Be tween 75% and 90% of the standardizing group agreed or disagreed with the attitude statement. High-TD=Between 55% and 75% of the standardizing group agreed or disagreed with the attitude statement. 309 TABLE 96 EXPERIMENT I MEAN CONPOEMITÏ SCORES (COMPLIANCE) POR THE YOUNG, OLD, MALE AND PEMALE SUBJECTS AT EACH OP THE THREE LEVELS OF TASK DIFFICULTY. (SCORES AT EACH LEVEL OF DIFFICULTY COULD RANGE FROM 0 TO 8.) Young Male Female Old Male Female Low Task Difficulty 0.000 0.166 0.166 0.166 Medium Task Difficulty 0.555 0.555 0.500 0. 666 High Task Difficulty 0.655 0.666 1.500 2.166 510 TABLE 97 EXPERIMENT I MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) FOR THE YOUNG, OLD, MALE AND FEMALE SUBJECTS AT EACH OF THE THREE LEVELS OF TASK DIFFICULTY. (SCORES AT EACH LEVEL OF DIFFICULTY COULD RANGE FROM 0 TO 8.) Young Male Female Old Male Female Low Task Difficulty 0.000 0.000 0.000 0.000 Medium Task Difficulty 0.166 0.166 0.000 0.166 High Task Difficulty 0.500 0.500 0.855 1.000 311 TABLE 90 EXPERIMENT I MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) FOR THE YOUNG, OLD, MALE AND FEMALE SUBJECTS AT EACH OF THE THREE LEVELS OF TASK DIFFICULTY. (SCORES AT EACH LEVEL OF DIFFICULTY COULD RANGE FROM 0 TO 8.) Young Male Female Old Male Female Low Task Difficulty 0.000 0.000 0.000 0.000 Medium Task Difficulty 0.000 0.000 0.000 0.166 High Task Difficulty 0.000 0.166 0.166 0.533 312 TABLE 99 EXEEHIMENT II MEAN COHEOEMITT SCORES (COMPLIANCE) FOR THE YOUNG, OLD, MALE AND FEMALE SUBJECTS AT EACH OF THE THREE LEVELS OF TASK DIFFICULTY. (SCORES AT EACH LEVEL OF DIFFICULTY COULD RANGE FROM 0 TO 8.) Young Male Female Old Male . Female Low Task Difficulty 0.533 0.166 0.533 0.166 Medium Task Difficulty 0.666 0.500 1.000 0.853 High Task Difficulty 1.333 1.000 2.666 2.166 313 TABLE 100 EXPERIMENT II MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) FOR THE YOUNG, OLD, MALE AND FEMALE SUBJECTS AT EACH OF THE THREE LEVELS OF TASK DIFFICULTY. (SCORES AT EACH LEVEL OF DIFFICULTY COULD RANGE FROM 0 TO 8. ) Young Male Female Old Male Female Low Task Difficulty 0.000 0.000 0.166 0.166 Medium Task Difficulty 0.533 0.533 0.533 0.166 High Task Difficulty 0.853 0.666 1.500 1.333 514 TABLE 101 EXPERIMENT II MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) FOR THE YOUNG, OLD, MALE AND FEMALE SUBJECTS AT EACH OP THE THREE LEVELS OF TASK DIFFICULTY. (SCORES AT EACH LEVEL OF DIFFICULTY COULD RANGE FROM 0 TO 8. ) Young Male Female Old Male Female Low Task Difficulty 0.000 0.000 0.000 0.000 Medium Task Difficulty 0.166 0.000 0.166 0.166 High Task Difficulty 0.533 0.166 0.533 0.533 515 TABLE 102 EXPERIMENT III MEAN CONEOEMITr SCOEES (COMPLIANCE) FOE THE YOUNG, OLD, MALE AND FEMALE SUBJECTS AT EACH OF THE THEEE LEVELS OF TASK DIFFICULTY. (SCOEES AT EACH LEVEL OF DIFFICULTY COULD EANGE FEOM 0 TO 8.) Young Male Female Old Male Female Low Task Difficulty 0.333 0.355 0.500 0.500 Medium Task Difficulty 0.835 0.666 1.666 1.500 High Task Difficulty 1.300 1.500 3.000 2.666 316 TABLE 103 EXPERIMENT III MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) FOR THE YOUNG, OLD, MALE AND FEMALE SUBJECTS AT EACH OF THE THREE LEVELS OF TASK DIFFICULTY. . (SCORES AT EACH LEVEL OF DIFFICULTY COULD RANGE FROM 0 TO 8. ) Young Male Female Old Male Female Low Task Difficulty 0.000 0.166 0.166 0.500 Medium Task Difficulty 0.355 0.500 0.666 0.500 High Task Difficulty 0.835 1.000 2.166 2.000 $17 TABLE 104 EXPERIMENT III MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) FOR THE YOUNG, OLD, MALE AND FEMALE SUBJECTS AT EACH OF THE THREE LEVELS OF TASK DIFFICULTY. (SCORES AT EACH LEVEL OF DIFFICULTY COULD RANGE FROM 0 TO 8. ) Young Male Female Old Male Female 0.000 0.000 0.166 0.000 0.166 0.166 0.53$ 0.166 0.355 0.55$ 0.500 0.533 Low Task Difficulty Medium Task Difficulty High Task Difficulty 318 TABLE 103 EXEESIMENT IV KEAN CONFOBMITT SCOEES (GOKELIANCE) FOE THE TOÜHG, OLD, KALE AHD EEKALE SUBJECTS AT EACH OF THE THEEE LEVELS OF TASK DIFFICULTY. (SCOEES AT EACH LEVEL OF DIFFICULTY COULD EANGE FEOM 0 TO 8.) Young Male Female Old Male Female Low Task Difficulty 0.553 0.853 0.853 1.000 Medium Task Difficulty 0.666 1.333 1.666 2.666 High Task Difficulty 1.500 .266 3.300 4.666 319 TABLE 106 EXPERIMENT IV MEAN CONFORMITY SCORES (1st PRIVATE ACCEPTANCE) FOR THE YOUNG, OLD, MALE AND FEMALE SUBJECTS AT EACH OF THE THREE LEVELS OF TASK DIFFICULTY. (SCORES AT EACH LEVEL OF DIFFICULTY COULD RANGE FROM 0 TO 8.) Young Male Female Old Male Female Low Task Difficulty 0.166 0.533 0.500 0.853 Medium Task Difficulty 0.555 1.000 1.000 1.853 High Task Difficulty 1.000 1.333 2.500 5.666 520 TABLE 107 EXPERIMENT IV MEAN CONFORMITY SCORES (2d PRIVATE ACCEPTANCE) FOR THE YOUNG, OLD, MALE AND FEMALE SUBJECTS AT EACH OF THE THREE LEVELS OF TASK DIFFICULTY. (SCORES AT EACH LEVEL OF DIFFICULTY COULD RANGE FROM 0 TO 8.) Young Male Female Old Male Female Low Task Difficulty 0.166 0.166 0.533 0.533 Medium Task Difficulty 0.166 0.500 0.666 1.000 High Task Difficulty 0.500 0.666 1.666 2.533 521 TABLE 108 EXPERIMENT V MEAN CONFORMITY SCOEES (COMPLIANCE) POE THE YOUNG, OLD, MALE AND FEMALE SUBJECTS AT EACH OP THE THEEE LEVELS OP TASK DIFFICULTY. (SCOEES AT EACH LEVEL OF DIFFICULTY COULD EANGE FEOM 0 TO 8.) Young Male Female Old Male Female Low Task Difficulty 0.855 0.855 0.855 1.000 Medium Task Difficulty 1.166 1.500 1.853 2.666 High Task Difficulty 2.000 2.500 4.000 5.000 TABLE 109 EXPERIMENT V MEAN CONFORMITY SCOEES (1st PRIVATE ACCEPTANCE) FOR THE YOUNG, OLD, MALE AND FEMALE SUBJECTS AT EACH OF THE THREE LEVELS OF TASK DIFFICULTY. (SCORES AT EACH LEVEL OF DIFFICULTY COULD RANGE FROM 0 TO 8.) Young Male Female Old Male Female Low Task Difficulty- 0.500 0.555 0.853 1.000 Medium Task Difficulty 0.666 1.000 1.333 2.000 High Task Difficulty 1.333 1.853 2.853 4.166 325 TABLE 110 EXPERIMENT V MEAN CONFORMITY SCOEES (2d PRIVATE ACCEPTANCE) FOR THE YOUNG, OLD, MALE AND FEMALE SUBJECTS AT EACH OF THE THREE LEVELS OF TASK DIFFICULTY. (SCORES AT EACH LEVEL OF DIFFICULTY COULD RANGE FROM 0 TO 8.) Young Male Female Old Male Female Low Task Difficulty 0.553 0.533 0.500 0.666 Medium Task Difficulty 0.500 0.566 0.853 1.333 High Task Difficulty 0.853 1.500 2.000 5.166 BIBLIOGRAPHY Abelson, R. P., & Lesser, G. S. The measurement of per- suasibility in children. In I. L. Janis & C. I. Hovland (Eds.), Personality and persuasibility. New Haven: Yale XJniver. ptess, 1959. Allen, V. L., & Crutchfield, R. S. Generalization of experimentally reinforced conformity. J. abnorm. soc. Psychol., 1965, 67, 526-553. Allman, E. W. A study of the social attitudes of college students. J. soc. Psychol., 1961, 53, 33-31. Applezweig, M. H., & Moeller, G. Conforming behavior and personality variables. Tech. Rep. No. 8, Contract No. NR 99^ (02), Connecticut College, New London, Conn., 1938. Asch, S. E. The doctrine of suggestion, prestige and imitation in social psychology. Psychol. Rev., 1948, 53, 250-277. Asch, S. E. Effects of group pressure upon the modificatioE and distortion of judgment. In H. Guetzkow (Ed.), Groups, leadership, and men. Pittsburgh: The Carnegie Press, 195^. Asch, S. E. Social Psychology. Englewood. Cliffs, New Jersey: Prentice Hall, 1952. Asch, S. E. Studies of independence and conformity. A minority of one against a unanimous majority. Psycho1. Monogr., 1936, 70, No. 9 (Whole No. 416). Atkin, S. Discussion of the paper by M. R. Kaufman, Old age and aging: the psychoanalytic point of view. Amer. J. Orthopsychiat., 1940, 10, 79-85. Back, K. ¥., & Davis, K. E. Some personal and situational factors relevant to the consistency and prediction of conforming behavior. Sociometry, 1965, 28, 227-240.--------------------------- Bandura, A. Social learning through imitation. In M. R. Jones (Ed.), Nebraska symposium on motivation. Lincoln: University of Nebraska, 1962. 325 526 Bandura, A. Influence of model's reinforcement contin gencies on the acquisition of imitative responses. J. pers. soc. Psychol., 1965, 1, 589-395. Barch, A. M., Trumbg, D., & Nangle, J. Society setting and conformity to a legal requirement. J. abnorm. soc. Psychol., 1937, 33, 596-598. Barron, P. Some personality correlates of independence of judgment. J. Pers., 1953, 21, 287-297. Basowitz, H., & Korchin, S. J. Age differences in the perception of closure. J. abnorm. soc. Psychol., 1937, 34, 93-97. Bass, B. Conformity, deviation, and a general theory of interpersonal behavior. In I. Berg, & B. Bass (Eds.3, Conformity and deviation. New York: Harper, 1961. Beloff, H. Two forms of social conformity: Acquiescence and conventionality. J. abnorm. soc. Psychol., 1938, 36, 99-105. Bendig, A. W. Age differences in the interscale factor structure of the Guilford-Zikmerman Temperament Survey. J. Consult. Psychol., I960, 24, 154-158. Berenda, R. W. The influence of the group on the judgments of children! New York: King's Crown Press, 1930. Berg, I., & Bass, B. (Eds.), Conformity and deviation. New York: Harper, 1961. Berkowitz, L., & Cottingham, D. R. The interest value and relevance of fear arousing communications. _J. abnorm. soc. Psycho1., I960, 60, 57-45. Birren, J. E. Toward an experimental psychology of aging. Amer. Psychol.. 1970, 25, 124-155. Blake, R. R., Berkowitz, A., Bellamy, R. Q., & Mouton, J. S. Volunteering as an avoidance act. J. abnorm. soc. Psychol.. 1936, 53, 154-156. Blake, R. R., & Brehm, J. W. The use of tape recording to stimulate a group atmosphere. J. abnorm. soc. Psychol.. 1954, 49, 511-313. 327 Blake, R. R., Helson, H., & Mouton, J. S. The generality of conformity behavior as a function of factual anchorage, difficulty of task, and amount of social pressure. J. Pers., 1936, 2,5, 194-505- Blake, R. R., & Mouton, J. S. Conformity, resistance and conversion. In I. Berg & B. M. Bass (Eds.), Conformity and deviation. New York: Harper, 1961. Bogdonoff, M. D., Klein, R. P., Estes, E. H., Jr., Shaw, B. M., & Back, K. ¥. The modifying effect of con forming behavior union lipid responses accompanying CNS arousal. Clin. Res.. 1961, 9, 153 (Abstract). Botwinick, J. Research problems and concepts in the study of aging. Gerontologist. 1964, 4, 121-129- Botwinick, J. Cautiousness in advanced age. J. Gerontol.. 1966, 21, 547-533. Botwinick, J., Brinley, J. P., & Robbin, J. S. The inter action effects of perceptual difficulty and stimulus exposure time on age differences in speed and accuracy of response. Gerontologia. 1938, 2, 1-10. Bovard, E. W., Jr. Social norms and the individual. J. abnorm. soc. Psychol.. 1948, 45, 62-69. Bray, D. ¥. The prediction of behavior from two attitude scales. J. abnorm. soc. Psychol., 1930, 45, 64-84. Brehm, J. ¥., & Cohen, A. R. Exp1orat ions in cognitive dissonance. New York: Wiley, 1962. Broadbent, D. E., & Gregory, M. Division of attention and the decision theory of signal detection. Proc. Roy Soc. B.. 1965, 158, 222-251 - Brown, R. Social psychology. New York: Free Press, 1965- Burt, H. E. Sex differences in the effect of discussion. J. exp. Psychol.. 1920, 5, 590-593* Campbell, D. T. Conformity in psychology's theories of acquired behavioral dispositions. In I. Berg & B* Bass (Eds.), Conformity and deviation. New York: Harper, 1961. Cardno, J. A. The notion of attitude : An historical note. Psychol. Rep.. 1933, 1, 545-532. 528 Carter, L. F., Haythorn, W,, Shriver, E., & Lanzetta, J. The behavior of leaders and other group members. J. abnorm. soc. Psychol.. 193^, 46, 389-393- Cartwright, D. C., & Zander, A. (Eds.), Group dynamics: Research and theory (2nd éd.), Evanston, 111. : Sow,’ Peterson, 1960. Champney, H. The variables of parent behavior. J. abnorm. soc. Psychol.^ 1941, 56, 523-542. Chipman, A. Conformity as a differential function of social pressure and judgment difficulty. J. soc. Psychol.. 1966, 70, 299-311- Christie, R., & Lindauer, Florence. Personality structure. In P. R. Farnsworth (Ed.), Annual review of psychology. Vol. 14, Palo Alto, Calif-: Annual Reviews, 1965, 201-250. Clark, H. The crowd. Psychol. Monogr.. 1916, 21, 26-56. jCoch, L., & French, J. R. P., Jr. Overcoming resistance to change. Human Relations. 1938, 11, 41-55- Coffin, T. E. Some conditions of suggestion and sus ceptibility. Psychol- Monogr., 1941, 53, No. 4 (Whole No. 241TT Cohen, A. R. Some implications of self-esteem for social influence. In I. L. Janis & C. I. Hovland (Eds.), Personality and persuasibility. New Haven: Yale TJniver. Pres s, 1939 - Coleman, J. F,, Blake, R. R., & Mouton, J. S. Task diffi culty and conformity pressures. J. abnorm. soc. Psychol.. 1938, 37, 120-122. Converse, P., & Campbell, A. Political standards in secondary groups. In D. Cartwright, & A^ Zander (Eds.), Group dynamics. New York: Haiper and Row, 1968. Cowley, W. H. Three distinctions in the study of leaders. J- abnorm. soc. Psychol.. 1928, 25, 144-157- Craik, F. I. M. The effects of age and the experimental situation on confidence behaviour. Bull. Brit. Psychol. Soc.. 1962, 47, (abstract). 329 Craik, F. I. M. An observed age difference in responses to a personality inventory. British Journal of Psychology. 1964, 55, 453-462. Craik, F. I. M. Age differences in confidence and decision processes. Unpublished doctoral dissertation. University of Liverpool, 1965- Crutchfield, R. S, Conformity and character. Amer. Psychologist. 1955, 10, 191-198. Cummings, E., & Henry, W. E. Growing old. New York: Basic Books, 1961. Barley, J. M. Fear and social comparison as determinants of conformity behavior. J. pers. soc. Psychol.. 1966, 4, 75-78. Beutsch, M., & Gerard, H. B. A study of normative and informational social influence upon individual judg ment. J. abnorm. soc. Psychol.. 1955, 51, 629-656. Dinsmoor, J. A. Punishment : I. The avoidance hypothesis. Psychol. Rev.. 1954, 61, 54-46. Dinsmoor, J. A. Punishment: II. An interpretation of empirical findings. Psychol. Rev.. 1955, 62, 96-105* Dittes, J. E. & Kelley, H. H. Effects of different conditions of acceptance upon conformity to group norms. J. abnorm. soc. Psychol.. 1956, 53, 100-107* Di Vesta, F. J. Effects of confidence and motivation on susceptibility to informational social influence. J. abnorm. soc. Psychol.. 1959, 59, 204-209. Bunker, K. Experimental modifications of children's food preferences through social suggestion. J. abnorm. soc. Psychol.. 1938, 53, 489-507* Eisdorfer, C., Axelrod, S., & Wilkie, F. L. Stimulus exposure time as a factor in serial learning in an aged sample. J. abnorm. soc. Psychol.. 1965. 67. 594-600. Endler, N. S. Social conformity in perception of the auto- kinetic effect. J. abnorm. soc. Psychol.. I960. 61, 489-490. 330 Endler, N. S. The effects of verbal reinforcement on conformity and deviant behavior. J. soc. Psychol., 1965, 66, 147-154. Estes, W. K. An experimental study of punishment. Psycho1. Monogor.. 1944, 57, ^o. 5. Ferguson, L. ¥. An analysis of the generality of suggestibility to group opinion. Charact, and Pers., 1944, 12, 257-245. Festinger, L. Informal social communication. Psychol. Rev.. 1950, 57, 271-292. Festinger, L. An analysis of compliant behavior. In M. Sherif & M. 0. Wilson (Eds.), Group relations at the crossroads. New York: Harpe*r^ Î953I Festinger, L., Gerard, H. B., Hymovitch, G., Kelley, H. G., & Raven, B. The influence process in the presence of extreme deviates. Human Relations. 1952, 5, 527-346. Festinger, L. A theory of social comparison processes. Human Relations. 1954, 7, 117-140. Festinger, L. A theory of cognitive dissonance. Stanford, Calif.: Stanford University Press, 1957* Festinger, L., & Thibaut, J. Interpersonal communication in small groups. J. abnorm. soc. Psychol.. 1951, 46, 92-99. Fisher, S., Williams, H. L., & hubin, A. Personal pre dictors of susceptibility to social influence. Amer. Psychologist. 1957, 12, 560 (Abstract). Foskett, J. M. Social structure and social participation. Amer, sociol. Rev.. 1955, 20, 451-458. Freed, A., Chandler, P. J., Mouton, J. S., & Blake, R. R. Stimulus and background factors in sign violation. J. Pers.. 1955, 25, 499. Freeman, J. L., & Fraser, S. C. Compliance without pressure : the foot-in-the-door technique. J. pers. soc. Psychol.. 1966, 4, 195-202. French, J. R. P., Jr. Organized and unorganized groups under fear and frustration. Studies in topological and vector psychology. III. Studies in Child Welfare. Iowa City: University of~Iowa, 1944. 331 French, J. R. P., Jr., & Raven, B. H. The bases of social power. In D. Cartwright (Ed.), Studies in social power. Ann Arbor, Mich.: Institute for Social Research, 1959. Frye, R., & Bass, B. M. Social acquiescence and behavior in groups. In I. Berg & B. Bass (Eds.), Conformity and deviation. New York; Harper, 1961. Gardner, J. W. Self-renewal. New York: Harper, 1965. Garrison, K. C. Worldminded attitudes of college students in a southern university. J. soc. Psychol.. 1961, 54, 147-153. Gavron, E. F. Changes in Edwards personal preference schedule needs with age and psychiatric status. J. Clin. Psychol.. 1965, 21, 194-196. Gerard, H. B. The anchorage of opinions in face-to-face groups. Human Relations. 1954, 7, 313-326. Gerard, H. B., Wilhelmy, R. A., & Conolley, E. S, Con formity and group size. J. pers. soc. Psychol.. 1968, 8, 79-82. Giedt, F. H., & Lehner, G. F. J. Assignment of ages on the Draw-a-Person Test by male neuropsychiatrie patients. J. Pers., 1951, 19, 440-448. Goldberg, S. C. Three situational determinants of conformity to social norms. J. abnorm. soc. Psychol., 1954, 49, 325-329. Gordon, R. L. Interaction between attitude and the definition of the situation in the expression of opinion. Amer, sociol. Rev., 1952, 17, 50-58. Graham, D. Experimental studies of social influence in simple judgmental situations. J. soc. Psychol., 1962, 56, 245-269. Green, D. M., & Swets, J. A. Signal detection theory and psychophysics. New York: Wiley, 1966. Grosser, D., Polansl^, N., & Lippitt, R. A laboratory study of behavioral contagion. Human Relations, 1951. 4, 115-142. 332 Hardy, K. R. Determinants of conformity and attitude change. J. abnorm. soc. Psychol., 1957, 54, 245-269. Harper, P. B. V., & Tuddenham, R. D. The sociometric composition of the group as a determinant of yielding to a distorted norm. J. Psychol.. 1964, 58, 307-311. Heider, P. Attitudes and cognitive organization. J. Psychol.. 1946, 21, 107-112. Heider, P. The psychology of interpersonal relations. New York.: Wiley, 1958. Helson, H., Blake, R. R., Mouton, J. S., & Olmstead, J. A. Attitudes as adjustments to stimulus background and residual factors. J. abnorm. soc. Psychol., 1956, 52, 514-522. Hemphill, J. K. Situational factors in leadership. Educ. Res. Monogr.. Ohio State Univer., 1949, No. 52. Henry, W. E., & Gumming, E. Personality development in adulthood and old age. J. proj. Tech., 1959, 25, 585-590. Hilgard, E. R., Bauer, Lillian W., & Melei, Janet P. Acquiescence, hypnotic susceptibility & the MMPI. J. consult. Psychol., 1965. Hochbaum, G. M. The relation between group members' self- confidence and their reactions to group pressures to uniformity. Amer, sociol. Rev., 1954, 19, 678-687. Hoffman, M. L. Some psychodynamic factors in compulsive conformity. J. abnorm. soc. Psychol., 1953, 48, 385-393. Hoffman, M. L. Conformity as a defense mechanism and as a form of resistance to genuine group influence. J. Pers., 1957, 25, 412-424. Hollander, E. P. Leaders, groups, and influence. New York: Oxford Univer. Press, 1964. Hollander, E. P. Principles and methods of ycial psychology. New’ToSl Oxford^ Univer. Press, 1967. Hollander, E. P., & Julian, J. W. Leadership. In E. P. Borgatta & W. W. Lambert (Eds.), Handbook of personality theory and research. Chicago : Rand McNally, 1967. 333 Hollander, E. P. & Willis, R. H. Some current issues in the psychology of conformity and nonconformity. Psychol. Bull., 1967, 68, 62-76. Homans, G. C. Social behavior: Its elementary forms. New York : Harcourt Brace, and World, 1961. Horwitz, M-, Piana, G. M., Goldman, D. M., & Lee, P. J. Veridicality of attitudes toward authority and effects on learning. Amer. Psychologist, 1955, 10, 336 (Abstract). Hovland, C. I., Harvey, 0. J., & Sherif, M. Assimilation and contrast effects in reactions to communication and attitude change- J. abnorm. soc. Psychol., 1957, 55, 244-252. Hovland, C. I., Janis, I. L., & Kelley, H. H. Communication and persuasion. New Haven, Conn.: Yale Univer. Press, 1953. Hovland, C. I., & Pritzker, H. A. Extent of opinion change as a function of amount of change advocated. J. abnorm. soc. Psychol., 1957, 54, 257-261. Hunt, J. McV. Traditional personality theory in the light of recent evidence. American Scientist., 1965, 53, 60-96 Hunter, K. R. The effects of insecurity, task ambiguity, and sex on conformity. Dissertation Abstr., 1968, 29 (6-B). Insko, C. A., Arkoff, A., & Insko, V. M. Effects of high and low fear-arousing communications upon opinions toward smoking. J. exp. soc. Psychol., 1965, 1, 256-266. Israel, J. Experimental change of attitude using the Asch-effect. Acta Sociol., 1965, 7, 95-104. Jackson, J. H., & Saltzstein, H. D. The effect of person-group relationships on conformity processes. J. abnorm. soc. Psychol., 1958, 57, 17-24. Jahoda, M. Conformity and independence. Human Relations, 1959, 12, 99-120. Janis, I. L. Anxiety indices related to susceptibility to persuasion. J. abnorm. soc. Psychol., 1955, 51, 665—667. 334 Janis, I. L., & Field, P. B. Sex differences and per sonality factors related to persuasibility. In I. L. Janis & C. I. Hovland (Eds.), Personality and persuasibility. New Haven, ConnTl Yale tmiver. Press, 1959• Janis, I. L., & Hovland, C. I. (Eds.), Personality and persuasibility. New Haven, Conn.: Tale Univer. Press, 1959. Jenness, A. The role of discussion in changing opinions regarding a matter of fact. J. abnorm. soc. Psychol., 1932, 27, 279-296. Johnson, H. H. Some effects of discrepancy level on responses to adverse information about one's self. Unpublished Ph.D. dissertation, Univer. of Illinois, 1964. Kahana, E., & Coe, M. Dimensions of conformity: a multi disciplinary view. J. Gerontol., 1969, 24, (1), 76-81. Kanareff, V. F., & Lanzetta, J. F. The acquisition of imitative and opposition responses under two conditions of instruction induced set. J. exp. Psychol., 1958, 56, 516-528. Kanareff, 7. F., & Lanzetta, J. F. Effects of task definition and probability of reinforcement upon the acquisition and extinction of imitative responses. J. exp. Psychol.. i960, 60, 540-548. Kassarjian, W. M., & Kassarjian, H. H. Conformity of judgment in a group situation. Psychol. Pep., 1962, 10, 491- 494. Kaufman, M. R. Old age and aging: the psychoanalytic point of view. Amer. J. Orthopsychiat., 1940, 10, 73- 79. Kelley, H. H., & Lamb, T. W. Certainty of judgment and resistance to social influence. J. abnorm. soc. Psychol., 1957, 55, 137-139. Kelley, H. H., & Shapiro, M. M. An experiment on conformity to group norms where conformity is detri mental to group achievement. Amer, sociol. Rev., 1954, 19, 667-677. 335 Kelley, H. H. & Volkart, E. H. The resistance to change of group-anchored attitudes. Amer. Sociol. Rev., 1952, 17, 453- 465. Kelman, H. C. Effects of success and failure on "suggestibility” in the autokinetic situation. J. abnorm. soc. Psychol., 1950, 45, 267-285. Kelman, H. C. Processes of opinion change. Publ. Opin. Quart., 1961, 25, 57-78. Kelman, H. C. Deception in social research. Transaction, 1966, 5, 20-24. Kidd, J. S., & Campbell, D. T. Conformity to groups as a function of group success. J. abnorm. soc. Psychol., 1955, 51, 390- 593. I I Kiesler, C. A. Conformity. Philippines: Addison-Wes1er Pub. Co., 1969. Kiesler, C. A., Zanna, M., & DeSalvo, J. Deviation and conformity: opinion change as a function of commit ment, attraction, and presence of a deviate. J. pers. soc. Psychol., 1966, 5, 458-467. Kirkpatrick, C. An experimental study of the modification of social attitudes. Amer. J. Sociol., 1936, 41, 649-656. Klein, R. L. Age, sex, and task difficulty as predictors of social conformity. Unpublished master's thesis. University of Southern California, 1970. Knower, P. H. Experimental studies of changes in atti tudes I. A study of the effect of oral arguments on changes of attitudes. J. soc. Psychol., 1935, 6, 315-347. Koponen, A. The influence of demographic factors on response to the Edwards Personal Preference Schedule. Unpublished doctoral dissertation, Columbia University, 1957. Korchin, S. J., & Basowitz, H. Age differences in verbal learning. J. abriuxm. soc. Psychol., 1957, 54, 64-69. Krebs, A. M. Two determinants of conformith, age of independence training and n achievement. J. abnorm. soc. Psychol., 1958, 56, 150-151. 336 Krech, D., Crutchfield, R. S., & Ballachey, E. L. Individual in society. New York; McGraw-Hill, 1962. Kuhlen, R. G. Aging and life-adjustment. In J. E. Birren (Ed.) Handbook of aging and the individual. Chicagol Univer. Chicago Press, 1959. Kuhlen, R. G. Developmental changes in motivation during the adult years. In J. E. Birren (Ed.), Relations of development and aging. Springfield, 111.: Charles C. Thomas, 1964• Kuhlen, R. G., & Johnson, G. H. Changes in goals with increasing adult age. J. Consult. Psychol., 1952, 16, 1-4. iLeBon, G. Psychologie des Foules. Paris: Felix Alcan, 1895. (English translation, London: T. Fisher Unwin, 1896.) League, B. J., & Jackson, D. N. Conformity, veridicality and self-esteem. J. pers. soc. Psychol., 1964, 68, 115-115. Lehner, G. F. J., & Gunderson, E. K. Height relationships on the Draw-a-Person Test. J. Pers., 1953, 21, 392-599. Lehner, G. F. J., & Silver, H. Age relationships on the Draw-a-Person Test. J. Pers., 17, 199-209. Leventhal, H., & Niles, Patricia. A filed experiment on fear arousal with data on the validity of question naire measures. J. Pers., 1964, 459-479. Leventhal, H., Singer, R. P., & Jones, S. H. The effects of fear and spcificity of recommendation upon attitudes and behavior. J. pers. soc. Psychol., 1965, 2, 20-29. Levinson, D. J. An approach to the theory and measure ment of ethnocentric ideology. J. Psychol., 1949, 28, 19-39. Lewin, K. Field theory in social science ; selected theoretical papers. New Ÿork: ïïaxpers, 1951. Linton, Harriet. Dependence on external influence: correlates in perception, attitudes, and adjustment. J. abnorm. soc. Psychol., 1955, 51, 502-507. 337 Linton, Harriet, & Graham, Elaine. Personality correlates of persuasibility. In I. L. Janis & C. I. Hovland (Eds.), Personality and persuasibility. New Haven, Conn.: Yale Univer. Press, 1959* London, P., & Lim, H. Yielding reason to social pressure : task complexity and expectation to conformity. J. Pers.. 1964, 52, (l), 73-89. Lott, A. J., & Lott, B. E., Group cohesiveness, communi cation level, and conformity. J. abnorm. soc. Psy chol. , 1961, 62, 408-412. Luchins, A. S. On agreement with anothers judgments. J. abnorm. soc. Psychol. 1944, 97-111. Luchins, A. S. Social influence on perception of complex drawings. J. abnorm. soc. Psychol., 1945, 21, 257- 274. Luchins, A. S. A variational approach to social influence on perception. J. soc. Psychol., 1955, 42, 115-119- Luchins, A. S., & Luchins, Edith H. On conformity with true and false communications. J. soc. Psychol., 1955, 42, 285-304. MacBride, P. D. The influence of confidence upon resistance to perceptual judgments to group pressure. Tech. Kept. 10, Contract NR 170-159, Univer. of California, 1958. Mahler, I. Yeasayers and naysayers: a validity study. J. abnorm. soc. Psychol., 1962, 64, 517-318. Mann, R. D. A review of the relationships between personality and performance in small groups. Psychol. Bull., 56, 241-270. Mason, Evelyn P. Some correlates of self-judgments of the aged. J. Gerontol., 1954, 9, 224-257- Mathie, J. Conformity apparatus. In S. A. Smith, Conformity in cooperative and competitive groups. J. soc. Psychol., 1965, 65, 357-350* Mausner, B. The effect of one partner's success in a relevant task on the interaction of observer pairs. J. abnorm. soc. Psychol., 1954, 49, 557-560. 338 Mayo, S. C. Age profiles of social participation in rural areas of Wake County, North Carolina. Rural Sociol., 1950, 15, 242-251. McClelland, D. C. The achievement motive. New York: Appleton-C entury-Oro ft s, 1953. McConnell, J. V., & Blake, R. R. A methodological study of tape recorded synthetic group atmosphers. Amer. Psychologist, 1953, 8, 595 lAhstract). McBavid, J. W. Personality and situational determinants of conformity. J. abnorm. soc. Psychol., 1959, 58, 241-246. McBavid, J. W. Effects of ambiguity of environmental cues upon learning to imitate. J. abnorm. soc. Psychol., 1962, 65, 581-586. McBavid, J. W. Effects of ambiguity of imitative cues upon learning by observation. J. soc. Psychol., 1964, 62, 165-174. Mead, M. Cooperation and competition among primitive peoples. New Yoik: McGraw-Hill, l939. Milgram, S. Nationality and conformity. Scientific American, 1961, 205, (6), 45-51. Miller, G. R., & Tiffany, W. T. The effects of group pressure on the judgments of speech sounds. _J. speech hear. Res., 1965, 6, 149-156. Milner, A. D., Walker, 7. J., & Beech, H. R. Obsessional behavior and decision-making— a signal-detection approach. Cited by P. I. M. Craik, Paper presented to the N.A.T.O. advanced study meeting on decision making and age, Thessaloniki, August, 1967. Moeller, G., & Applezweig, M. H. A motivational factor in conformity. J. abnorm. soc. Psychol., 1957, 55, 114-120. Mouton, J. S., Blake, R. R., & Olmstead, J. A. The relationship between frequency of yielding and the disclosure of personal identity. J. Pers., 24, 339-347. 339 Mowrer, 0. H. Authoritarianism vs. "self-government” in the management of children’s aggressive (anti-social) reactions as a preparation for citizenship in a democracy. J. soc. Psychol., 1939, 10, 121-126. Murphy, G., Murphy, Lois G., & Newcomb, T. M. Experimental social psychology. New York; Harper, 1937« Mussen, P. H., & Kagan, J. Group conformity and per ception of parents. Child Development, 1958, 29, 57-60. Musterberg, H. Beitragezur. Experimentellen psychologie, in P. W. Berenda (Ed.), The influence of the group on the .judgments of children. New York; Kings Crown Press, 1950. Nakamura, C. Y. Conformity and problem solving. J. abnorm. soc. Psychol., 1958, 56, 315-320. Neugarten, Bernice L., & Gutman, D. L. Age-sex roles in personality in middle age: a thematic apperception study. Psychol. Monogr., 1958, 72, No. A70. Newcomb, T. M., Turner, E. H., and Convers, P. E. Social psychology. New York: Holt, Rinehart, and Winston, 1965. Norman, R. D. Concealment of age among psychologists: evidence for a popular stereotype. J. Psychol., 1949, 30, 127-135. Olmstead, J. A., & Blake, R. R. The use of simulated groups to produce modifications in judgments. J. Pers., 1955, 23, 335-345. Olsen, I. A., & Elder, J. H. A word-association test of emotional disturbance in older women. J. Gerontol., 1958, 13, 305-3O8. Orne, M. T. The nature of hypnosis : artifact and essence. J. abnorm. soc. Psychol., 1959, 58, 277- 299. Orne, M. T. On the social psychology of the psychological experiment: With particular reference to demand characteristics and their implications. Amer. Psychologist, 1962, 17, 776-783. Packard, V. P. The status seekers. New York: McKay, 1959. 34-0 Phelps, P. E., & Meyer, E. Personality and conformity, and sex differences. Psychol. Ken., 1966, 18 (5), 750. Powell, M., & Ferraro, C. P. Sources of tension in married and single women teachers of different ages. J. Educ. Psychol., i960, , 92-101. Haven, B. H. Social influence on opinions and the communi cation of related content. J. abnorm. soc. Psychol., 1959, 58, 119-128. Haven, B. H., & Hietsema, J. The effects of varied clarity of group goal and group path upon the individual and his relation to his group. Human Helations, 1957, 10, 29-4-5. Hees, J. N., & Botwinick, J. Detection and decision factors in auditory behavior of the elderly. J. Gerontol, 1971, 26, 135-156. Heisman, D. The lonely crowd. New Haven, Conn.: Yale Univer. Press, 1950. Hiopelle, A. J. Complex processes. In H. H. Waters, D. A. Hethlingshafer, & W. E. Caldwell (Eds.), Principles of comparative psycholo^y. New York : McGraw-Hill, i960. Hohrer, J. H., Baron, S. H., Hoffman, E. L., & Swander, D. V. The stability of autokinetic judgments. J. abnorm. soc. Psychol., 1954-, 4-9 , 595-597* Hosen, J. L., & Neugarten, Bernice L. Ego functions in the middle and later years: a thematic apperception study of normal adults. J. Gerontol., I960, 15, 62-67* Hosenbaum, M. E. The effect of stimulus and background factors on the volunteering response* J. abnorm. soc. Psychol., 1956, 55, 118-121. Hosenberg, L. A. Group size, prior experience, and con formity. J. abnorm. soc. Psychol., 1961, 63, 4-36- 437* Hosner, S. Consistency of response to group pressures. J. abnorm. soc. Psychol., 1957, 55, 14-5-14-6. Hosow, I. Social integration of the aged. New York: The Free Press, 1967* 341 Samelson, F. Conforming behavior under two conditions of conflict in the cognitive field. J. abnorm. soc. Psychol., 1957, 55, 181-187. Sampson, D. L., & Smith, H. P. A scale to measure world- minded attitudes. J. soc. Psychol., 1957, 43, 99- 106. Sampson, E. E. (Ed.), Approaches, contests, and problems of social psychology. New Jersey: Prentice-Hall, 1964. Schacter, S., & Hall, P. Group-derived restraints and audience persuasion. Human Eolations, 1952, 5, 397-402. Schaie, I. W., and Strother, C. R. The psychological functioning of superior normal old adults. In S. M. Chown, & K. Riegel (Eds.), Psychological functioning in the normal aging and senile aging. Basle: Karger, 1968. Scheffe, H. The analysis of variance. New York: Wiley, 1959. Schonbar, R. A. The interaction of observer-pairs in judging visual extent of movement. Arch. Psychol., 1945, 41, No. 299. Schroder, H. H., & Hunt, D. E. Dispositional effects upon conformity at different levels of discrepancy. J. Pers., 1958, 26, 244-258. Scott, S. A. Factors affecting the learning of personal values through social reinforcement. Amer. Psy chologist, 1956, 11, 407-408, (Abstract). Sears, R. R. Dependency motivation. In M. R. Jones (Ed.), Nebraska symposium on motivation. Lincoln: Univer. of Nebraska Press, 1963* Secord, P. F., & Backman, C. W. Social psychology. New York : McGraw-Hill, 1964. Seward, K. Age and mental ability in superior man. Amer. J. Psychol., 1945, 58, 443-470. Shanksmith, G. A. Conformity behavior on a Zebra crossing: an activity sampling analysis. Papers in Psychology, 1967, 1 (2), 56-58. 342 Shanan, J., & Sharon, M. Personality and cognitive functioning of Israeli males during the middle years. Human Dev., 1965, 8, 2-15* Sherif, M. A study of some social factors in perception. Arch. Psychol., 1955, 27, No. 187* Sherif, M. An outline of social psychology. New York: Harper and Brothers, 1$48. Silverman, I. Age and the tendency to withhold a response. J. Gerontol., 1963, 18, 372-375* Smith, H. P. Do intercultural experiences affect atti tudes? J. abnorm. soc. Psychol., 1955, 51, 469-477* Spangler, D. P., & Thomas, C. W. The effects of age, sex, and physical disability upon manifest needs. J. Couns. Psychol., 1962, 9 (4), 513-319* Stogdill, P. M. Personal factors associated with leader ship. J. Psychol., 1948, 25, 55-71* Swets, J. A., Tanner, ¥. P., & Birdsall, T. G. Decision processes in perception. Psychol. Pev., 1961, 68, 301-340. Tallent, N., & Lucas, L. Socialization of the aged. Geriatrics, 1958, 11, 266-272. Tarde, G. Les Lois de 1 * limitation. Paris: Pelix Alcan, 1890. C English transiat ion, Rew York : Holt, 1903*) Thibaut, J. W., & Kelley, H. H. Kie social psychology of groups. New York: Wiley, 1959* Thibaut, J. W., & Strickland, L. H. Psychological set and social conformity. J. Pers., 1956, 25, 115-129* Thorndike, E. L. The effect of discussion upon the correctness of group decisions when the factor of majority influence is allowed for. J. soc. Psychol., 1958, 9, 543-362. Thrasher, J. Interpersonal relations and gradations of stimulus structure as factors in judgmental variation: an experimental approach. Sociometry, 1954, 17, 228-541. 543 Tuddenham, R. D. The influence upon judgment of grossly distorted norm. Tech. Rept. 2, Contract NR 170-159* Univer. of California, l957* Tuddenham, R. D. The influence of a distorted group norm upon individual judgment. J. Psychol., 1958, 46, 227-241. Tuddenham, R. D. The influence upon judgment of the apparent discrepancy between self and others. _J. soc. Psychol., 1961, 55, 69-79* Tuddenham, R. D., & Mac Bride, P* D. The yielding experi ment from the point of view of the subject* Tech. Rept., 9 Contract NR 170-159, Univer. of California, 1958. Tuddenham, R. D., MacBride, P. D., & Zahn, V. The sex composition of the group as a determinant of yielding to a distorted norm. Tech Rept., 4, Contract NR 170-159, Univer. of California, 1958* Vaughn, C. M. The trans-situational aspect of conformity behavior. J. Pers., 1964, 32, 355-554. Veroff, J., Atkinson, J. ¥., Feld, Sheila, & Gurin, G. The use of thematic apperception to assess motivation in a nationwide interview study. Psychol. Monogr., 74 (12, Whle No. 499), I960. Vidulich, R. N., & Kaiman, I. P. The effects of information source status and dogmatism upon conformity behavior. J. abnorm. soc. Psychol., 1961, 65, 639-642. Vagner, E. E. Differences between old and young executives on psychological test variables. Gerontol., I960, 15, 296-299* Walker, E. L., & Heyns, R. W. An anatomy for conformity. New Jersey: Fhentice-Hall, 1062. Wallach, M. A., & Kogan, N. Aspects of judgment and decision-making : interrelationships and changes with age. Behav. Sci., 1961, 6, 23-56. Walters, R. H. & Karol, P. Social deprivation and verbal behavior. J. Pers.. I960, 28, 89-107* Walters, R. H., Marshall, W. S., & Shooter, J. R. Anxiety, isolation, and susceptibility to social influence. J. Pers.. I960, 28, 518-529* 544 Whyte, W. H., Jr. The organization of man. New York : Simon and Schuster, 1956. Wiener, M. Uncertainty of judgment as a determinant of conformity behavior. Amer. Psychologist, 1956, 11, 407, (abstract). Wiener, M. Certainty of judgment as a variable in con formity behavior. J. soc. Psychol., 1958, 48, 257- 265. Wiener, H. Some correlates of conformity responses. J. soc. Psychol.. 1959, 49, 215-221. Wiener, M., Carpenter, Janet, & Carpenter, B. External validation of a measure of conformity behavior. J. abnorm. soc. Psychol., 1956, 52, 421-422. Welford, A. T. Skill and age : an experimental approach. London: Oxford Univer. Press, 1951• Willis, P. H. Descriptive models of social response. Tech. Pep., 1964, Washington University, Nonr 816(12), Office of Naval Research. Wilson, R. S. Personality patterns, source attractive ness and conformity. J. Pers., i960, 28, 186-199* Zborowski, M., & Eyde, L. E. Aging and social partici pation. J. Gerontol.. 1962, 17, 424-450.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Age, Sex, And Task Difficulty As Predictors Of Social Conformity: A Search For General Tendencies Of Conformity Behavior

PDF

Personality and initial interaction behavior as predictors of readiness to form a friendship relation

PDF

Age comparisons of the effects of cognitive and behavioral mood elevating tasks

PDF

Relevance in public administration: An identification and analysis of relevant behavior

PDF

The search for structure: A comparison of systems and social learning theory models of family interaction

PDF

A constraint-based approach to phonology and morphology

PDF

Age and sex differences in personal needs and the nature of love: A study of happily married young, middle-aged and older adult couples

PDF

The effects of systematic conceptual training using modeling techniques on intellectual performance, modeling, attention to teachers, and classroom behavior of preschool children from middle clas...

PDF

Triadic concepts in religious thought: (A selected study of Platonic, Indian, and Chinese sources)

PDF

A quantitative study of behavioral development: The relationship between age and tonal patterns in isolated roller canaries

PDF

A study of Puranic myth from the viewpoint of depth psychology

PDF

An analysis method for direct sequence systems employing noncoherent reception with applications to fast frequency hopping, multiple access, jamming and soft limit diversity combining

PDF

Mood and age

PDF

The relationship between co-operation and competition in social life; a general systems theory analysis

PDF

Intellectual deficit and the formation of learning sets

PDF

Dual coding in memory as a function of age, sex, and intelligence

PDF

Therapist and client in-session behaviors and their relationship to client between-session behaviors and outcome in marital and family therapy: A microlevel analysis

PDF

The Type A behavior pattern: Articulated thoughts and attributions during failure and success

PDF

The effect of social evaluations upon behavior: Secondary reinforcement versus cognitive interpretations

PDF

Predictors of subjective experience in schizophrenia

Asset Metadata

Creator Klein, Ronald Lawrence (author)

Core Title Age, sex, and task difficulty as predictors of social conformity: A search for general tendencies of conformity behavior

Degree Doctor of Philosophy

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,Psychology

Format application/pdf (imt)

Language English

Contributor Digitized by ProQuest (provenance)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c36-860731

Unique identifier UC11250192

Identifier DP30477.pdf (filename),usctheses-c36-860731 (legacy record id)

Legacy Identifier DP30477.pdf

Dmrecord 860731

Document Type Dissertation

Format application/pdf (imt)

Rights Klein, Ronald Lawrence

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA