University of Southern California Dissertations and Theses

Anxiety, Physiologically And Psychologically Measured, And Its Consequences On Mental Test Performance

(USC Thesis Other)

Anxiety, Physiologically And Psychologically Measured, And Its Consequences On Mental Test Performance

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content This dissertation has been microfilmed exactly as received 68-5 8 5 9 CHAMBERS, Alma Clyde, 1917- ANXIETY, PHYSIOLOGICALLY AND PSYCHOLOGICALLY MEASURED, AND ITS CONSEQUENCES ON MENTAL TEST PERFORMANCE. University of Southern California, Ph. D., 1967 Education, psychology University Microfilms, Inc., Ann Arbor, Michigan © COPYRIGHT BY ALMA CLYDE CHAMBERS 1968 ANXIETY, PHYSIOLOGICALLY AND PSYCHOLOGICALLY MEASURED, AND ITS CONSEQUENCES ON MENTAL TEST PERFORMANCE by Alma Clyde Chambers A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Educational Psychology) August 1967 UNIVERSITY O F SOUTHERN CALIFORNIA T H E G R A D U A T E S C H O O L U N IV E R S IT Y PA R K L O S A N G E L E S . C A L IF O R N IA 9 0 0 0 7 This dissertation, written by ................................ .ALMA. CLYDE. C H E E R S ................................... under the direction of h.^....Dissertation Com mittee, and approved by all its members} has been presented to and accepted by the Graduate School, in partial fulfillment of requirements for the degree of D O C T O R OF P H IL O S O P H Y <3. Dean. Date. September^JL967 The research reported herein was performed pursuant to a contract with the United States Department of Health, Education, and Welfare, Office of Education, under the provision of the Cooperative Research Program: Contract Number: OE-5-10-324, Bureau Number: 5-8119- 2-12-1. Refer to Educational Research Information Center, Rm. .3-0-083, 400 Maryland Avenue, S. W., Wash ington, D. C. 20202. ACKNOWLEDGMENT S ! | The investigator wishes to express grateful appreciation for the invaluable and competent guidance of Dr. Kenneth D. Hopkins, who sponsored this study for a Federal grant and served as chairman of her doctoral committee until he accepted a call to the University of Colorado, and to Dr. Richard M. Wolf, who assumed the chairmanship upon Dr. Hopkin's de parture . Appreciation is extended to the Federal Government, for the research grant that supported this study. The investigator wishes to express indebted ness to the department of Pharmacology at Loma Linda University, School of Medicine for providing use of | the Grass polygraph and other equipment. Appreciation is also extended to Dr. M. G. Hardinge and to the other members of the department of Pharmacology for their assistance with the pilot studies and with the in vestigation itself. | Appreciation is also extended to the University i ! of Southern California School of Medicine for the use of their tape recorder and pneumograph, and to Dr. J. P. Meehan of the School of Medicine and Mr. Tad Uno for professional counsel on the study. The investigator is especially indebted to i her parents, other members of her family and friends for encouragement and especially to her mother for assistance with the preparation of the data for the analysis. Appreciation is also extended to her sister and brother-in-law, Dr. and Mrs. William H. I Philpott for financial assistance. Appreciation is extended to Dr. Bobby R. ! Hopkins for the computer analysis and to the Western Data Processing Center for the Processing of the data. i Appreciation is also extended to Dr. John H. Hull, superintendent of the Torrance Unified School j District; Dr. Robert R. Ford, principal, and Mr. Paul i | F. Hawkins, assistant principal of West High School; | and tp Mr. Arnold C. Plank, Mr. Roland L. Glover, and | Mr. Merrill C. Louden at West High School for their j cooperation in releasing students from their govern- i ment classes during the collection of the data, j Appreciation is extended to the Educational Testing Service for permission to modify and use the i Cooperative Academic Ability Test, i iii To all these, whose contributions have helped make the fruition of this study possible, the investi gator wishes to express grateful acknowledgment. TABLE OF CONTENTS LIST OF LIST OF Chapter I. II. III. Page TABLES. . . ■ . » ... ... . . . . . . . . . vii FIGURES . ................. ........ xix PROBLEM Statement of the Problem Background of the Study Operational Definitions Hypotheses Limitations of the Study Significance of the Study Preview REVIEW OF THE LITERATURE ON ANXIETY . . . 15 Anxiety Anxiety and Fear Anxiety as a Constructive Force Anxiety: A Learned Concept Anxiety as a Drive Research on the Instruments Used Psychological Measures of Anxiety Physiological Measures of Anxiety Summary ANXIETY AND ITS RELATIONSHIP TO OTHER VARIABLES ...... ........ .... 67 Anxiety and Performance on an Ability Test Anxiety and Achievement Experimental Studies with Psychological Measures of Anxiety Sex Difference Summary v Chapter IV. METHODOLOGY . • . . ....... ....... The Sample Place and Conditions for Testing Instruments Procedures Analysis of Data Summary V. RESULTS . . . . . . ... . . . .... ... . Hypothesis One Hypothesis Two Hypothesis Three Hypothesis Four Hypothesis Five Hypothesis Six Hypothesis Seven Hypothesis Eight Hypothesis Nine Hypothesis Ten Hypothesis Eleven Summary VI. SUMMARY.- RECOMMENDATIONS, AND CONCLUSIONS ......... ....... Problem Procedure Recommendations REFERENCES. . . „ .... .... . . . . . . . . . APPENDICES................................ APPENDIX A. . ... , ... ......... . Letter for Approval to the Parents Cooperative Academic Ability Test The Student Survey Parts I, II, III, IV Treatment Instructions Page 34 108 161 I [ | .184 | | 199 | 200 ! APPENDIX B; Tables of the Analysis of the Data.......... . . . . 216 LIST OF TABLES Table 1. Results of Analysis of Variance for the Total, Verbal, and Mathematical Scores for the Academic Ability Test................ 2. Summary of Results of Analysis of Variance for the Academic Ability Test and Summary of Results of Analys^is of Covariance of the Academic Ability Test with the Pretest Anxiety Level Held Constant . . ............. 3. Summary of Results of Analysis of Covariance for Sex X Instruction X Sex Interaction for Total, Verbal, and Mathematical Scores When the Pretest Level for Each of the Physio logical Measures Was Held Constant. . . . . . . 4. Summary of Results of Analysis of Variance for Proficiency Level with the Mean Physiological Measures for the Three Instruc tion Groups and the Control Group ........... 5. Summary of Results for Analysis of Variance for Sex Differences for Physiological Measures. . . . . . . . . . . . . . . . . . . 6. Comparison of Means for Pretest Anxiety and Test Anxiety as Measured by Physio logical Measures for the Three Instruction Groups and the Control Group. . . . ......... 7. Correlational Matrix of Physiological Measures of Anxiety for the Group with Anxiety Reducing Instructions ...... ........... 8. Correlational Matrix of Physiological Measures of Anxiety for the Group With Neutral Instructions.. ............... ....... Page 109 113 115 120 121 127 130 131 vii Table Page 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. Correlational Matrix of the Physiological Measures of Anxiety for the Group with Anxiety-Producing Instructions. . . . . . . . Correlations Among Psychological Measures for the Total Group „ . . .......... . . Summary of Analysis of Variance for the Psychological Test According to Sex ...... Correlations, Means and t Values for the AACL Administered Two Weeks Prior to the Major Testing and Administered Again After the Anxiety Instructions..................... Correlations between Physiological and Psychological Measures of Anxiety for the Group with Anxiety-Reducing Instructions. . . Correlations between Physiological and Psychological Measures of Anxiety for the Group with Neutral Instructions ....... Correlations between Physiological and Psychological Measures of Anxiety for the Group with Anxiety-Producing Instructions . . Correlations between Psychological Measures of Anxiety and Scores on the Academic Ability Test with Anxiety-Reducing Instructions . . „ Correlations between Psychological Measures of Anxiety and Scores on the Academic Ability Test with .Neutral Instructions. ....... Correlations between Psychological Measures of Anxiety and Scores on the Academic Ability Test with Anxiety-Producing Instructions. . . Correlations between Physiological Measures of Anxiety and Scores on the Academic Ability Test with Anxiety-Reducing Instructions . . . 132 136 137 139 143 144 146 148 149 150 154 viii Table Page 20- 21- 22. 23. 24. 25. 26. 27. 28. 29. Correlations between Physiological Measures of Anxiety and Scores on the Academic Ability Test with Neutral Instructions. . ...... Correlations between Physiological Measures of Anxiety and Scores on the Academic Ability Test with Anxiety-Producing Instructions. . . Multiple Correlation for Physiological Measures Used as Predictors of Ability Test Scores. . . . . . . . . . a . . . o . . .. . . Results of the Analysis of Variance for the Date the Letter of Approval was Received. Results of the Analysis of Variance for the Period of the Day for Testing the Students... Results of the Analysis of Variance for the Day of Testing for Each Student . . . . . . „ . Means and Standard Deviations for the Total, Verbal and Mathematical Scores on the Academic Ability Test » . . . . . . . , „ Results of the Analysis of Variance for the Pretest Systolic Blood Pressure for the Three Instruction Groups and the Control Group for Physiological Measures. ......... Results of the Analysis of Variance for the Pretest Diastolic Blood Pressure for the Three Instruction Groups and the -Control for Physiological Measures. ..... ... ... Results of the Analysis of Variance for the Initial Pulse Pressure for the Three Instruction Groups and the Control Group for Physiological Measures. . „ . . . ... . .. 155 156 217 218 219 224 225 2 28 229 230 ix Table Page 30. Results of the Analysis of Variance for Pretest Oral Temperature for the Three Instruction Groups and- the Control Groups for Physiological Measures ...... 231 31. Results of the Analysis of Variance for the Initial Face Temperature for the Three Instruction Groups and the Control Group for Physiological Measures... . . . . . 232 32. Results of the Analysis of Variance for the Pretest Finger Temperature for the Three Instruction Groups and the Control Group for Physiological Measures. . . . . . . 233 33. Results of the Analysis of Covariance for the Total Test Score with the Initial Respiration Rate Held Constant. .. ..... 234 34. Results of the Analysis of Covariance for the Verbal Score with the Initial Respiration Rate Held Constant. . . . . . . . 235 35. Results of the Analysis of Covariance for the Mathematical Score with the Initial Respiration Rate Held Constant. . . . 236 36. Results of Analysis of Covariance for the Total Test Scores with Differences Between Pretest Reading and Test Reading Respiration Depth Held Constant . . . . . ... 237 37. Results of Analysis of Covariance for the Verbal Scores with Differences Between the Pretest Reading and Test Reading Respiration Depth Held Constant . . . 238 * 38. Results of Analysis of Covariance for the Mathematical Scores with Differences Between Pretest Reading and Test Reading Respiration Depth Held Constant . ...... . 239 x Table Page 39. Results of the Analysis of Covariance for the Total Test Score with the Initial Heartbeat Rate Held Constant. . . . . 240 40. Results of the Analysis of Covariance for the Verbal Score with the Initial Heartbeat Rate Held Constant.... . . . . . , 241 41. Results of the Analysis of Covariance for the Mathematical Score with the Initial Heartbeat Rate Held Constant. . . . . 242 42. Results of Analysis of Covariance for the Total Test Scores with, Differences Between Pretest Reading and Test Reading GSR Held Constant . ............... 243 43. Results of Analysis of Covariance for the Verbal Scores with Differences Between Pretest Reading and Test Reading GSR Held Constant.......... ... . . . . . . 244 44. Results of the Analysis of Covariance for the Mathematical Scores with Differences Between Pretest Reading and Test Reading GSR Held Constant........... . . . . . . . . 245 45. Results of the Analysis of Covariance for the Total Test Score with the Initial Systolic Blood Pressure Held Constant .... 246 46. Results of the Analysis of Covariance for the Verbal Score with Initial Systolic Blood Pressure Held Constant .... 247 47. Results of the Analysis of Covariance for the Mathematical Score with the Initial Systolic Blood Pressure Held Constant..................... 248 48. Results of the Analysis of Covariance for the Total Test Score with the Initial Diastolic Blood Pressure Held Constant. . . . 249 xi Table Page 49. Results of the Analysis of Covariance for the Verbal Score with the Initial Diastolic Blood Pressure Held Constant. . . . 250 50. Results of the Analysis of Covariance for the Mathematical Score with the Initial Diastolic Blood Pressure Held Constant. ..... ......................... 251 51. Results of the Analysis of Covariance for the Total Score with the Initial Pulse Pressure Held Constant. . . . . . . . . 252 52. Results of the Analysis of Covariance for the Verbal Score with the Initial Pulse Pressure Held Constant............. 253 53. Results of the Analysis of Covariance for the Mathematical Score with the Initial Pulse Pressure Held Constant. .... 254 54. Results of the Analysis of Covariance for the Total Test Score with the Initial Oral Temperature Held Constant. ............. 255 55. Results of the Analysis of Covariance for the Verbal Score with the Initial Oral Temperature Held Constant. . . . . . . . 256 56. Results of the Analysis of Covariance for the Mathematical Test Score with the Initial Oral Temperature Held Constant. . 257 57. Results of the Analysis of Covariance for the Total Test Score with the Initial Face Temperature Held Constant. . . . 258 58. Results of the Analysis of Covariance for the Verbal Score with the Initial Face Temperature Held Constant. . . . . . . . 259 Table Page 59. Results of the Analysis of Covariance for the Mathematical Score with the Initial Pace Temperature Held Constant..... 260 60. Results of the Analysis of Covariance for the Total Test Score with the Initial Finger Temperature Held Constant. . . . . . . 261 61. Results of the Analysis of Covariance for the Verbal Test Score with the Initial Finger Temperature Held Constant. ...... 262 62. Results of the Analysis of Covariance for the Mathematical Test Score with the Initial Finger Temperature Held Constant. . . 263 63. Results of the Analysis of Covariance for the Total Test Score for the Three Instruction Group and the Control Group with the Mean EKG Held Constant ... . . . . . 264 64. Results of the Analysis of Covariance for the Verbal Score for the Three Instruction Groups and the Control Group with the Mean EKG Held Constant. ............. 265 65. Results of the Analysis of Covariance for the Mathematical Score for the Three Instruction Groups and the Control Group with the Mean EKG Held Constant ............. 266 66. Results of the Analysis of Covariance for the Total Test Score for the Three Instruction Groups and the Control Group with the Mean Systolic Blood Pressure Held Constant . . . ... . . . . . ...........267 67. Results of the Analysis of Covariance for the Verbal Score for the Three Instruction Groups and the Control Group with the Mean Systolic Blood Pressure Held Constant . . . . 268 Xill J Table Page 68. Results of the Analysis of Covariance for the Mathematical Score for the Three Instruction Groups and the Control Group with the Mean Systolic Blood Pressure Held Constant . . . ■ . . . . • o-• . . ■ . • 69. Results of the Analysis of Covariance for the Total Test Score for the Three Instruction Groups and the Control with the Mean Diastolic Blood Pressure Held Constant. . . . . . . . . . . . . . . . . . . 70. Results of the Analysis of Covariance for the Verbal Score for the Three Instruction Groups and the Control Group with the Mean Diastolic Blood Pressure Held Constant . . . . . . 0 . 0 ....... 71. Results of the Analysis of Covariance for the Mathematical Score for the Three Instruction Groups and the Control Group with the Mean Diastolic Blood Pressure Held Constant 72. Results of the Analysis of Covariance for the Total Score for the Three Instruction Groups and the Control Group with the Mean Oral Temperature Held Constant. ....... 73. Results of the Analysis of Covariance for the Verbal Score for the Three Instruction Groups and the Control Group with the Mean Oral Temperature Held Constant. ....... 74. Results of the Analysis of Covariance for the Mathematical Score for the Three Instruction Groups and the Control Group with the Mean Oral Temperature Held Constant. 75. Means and Standard Deviations for Pretest Respiration Rate. . . . ........ ...... xiv 269 270 272 273 274 275 276 Table Page 76. Means and Standard Deviations for Pretest Respiration Depth . . . ........ 277 77. Means and Standard Deviations for Pretest Heartbeat Rate.............. 278 78. Means and Standard Deviations for Pretest GSR . . . . . . . . . . . . . . . 279 79. Means and Standard Deviations for Pretest Systolic Blood Pressure ....... 280 80. Means and Standard Deviations for Pretest Diastolic Blood Pressure. ...... 281 81. Means and Standard Deviations for Pretest Pulse Pressure. . . . . . . . . . . . 282 82. Means and Standard Deviations for Pretest Oral Temperature............. 283 83. Means and Standard Deviations for Pretest Pace Temperature. ...............284 84. Means and Standard Deviations for Pretest Finger Temperature. . . . . . . . . . 285 85. Unadjusted Means and Adjusted Means with the Respiration Rate Held Constant for the Ability Test with the Three Types of Instruction.................................286 86. Unadjusted Means and Adjusted Means with the Respiration Depth Held Constant for the Ability Test with the Three Types of Instruction....................................287 87. Unadjusted Means and Adjusted Means with the Heartbeat Rate Held Constant for the Ability Test with the Three Types of Instruction „ .. . . . .. .. . . .. . .. 288 xv Table Page 88. 89. 90. 91. 92. 93. 94. 95. 96. i i _ _ _ _ _ _ _ Unadjusted Means and Adjusted Means with the Diastolic Blood Pressure Held Constant for the Ability Test with the Three Types of Instruction. . . . . . . . 289 Unadjusted Means and Adjusted Means with the Pulse Pressure Held Constant for the Ability Test with the Three Types of Instruction............................... . 290 Results of the Analysis of Variance for the Mean Respiration Rate for the Three Instruction Groups and the Control Group. . . 291 Results of the Analysis of Variance for the Mean Minus the Initial Respiration Depth for the Three Instruction Groups and the Control Group ..................292 Results of the Analysis of Variance for the Mean Heartbeat Rate for the Three Instruction Groups and the Control Group. . . 293 Results of the Analysis of Variance for the Mean Minus the Initial GSR for the Three Instruction Groups and the Control Group . . . . . . . . . . . . ............... 294 Results of the Analysis of Variance for the Mean Systolic Blood Pressure for the Three Instruction Groups and the Control Group...................................... 295 Results of the Analysis of Variance for the Mean Diastolic Blood Pressure for the Three Instruction Groups and the Control Group ................... 296 Results of the Analysis of Variance for the Mean Pulse Pressure for the Three Instruction Groups and the Control Group. . . 297 xvi Table Page 97. Results of the Analysis of Variance for the Mean Oral Temperature for t^e Three Instruction Groups and the Control Group. . . 298 / 98. Results of the Analysis of Variance for the Mean Face Temperature for the Three Instruction Groups and the Control Group. . . 299 99. Results of the Analysis of Variance for the Mean Finger Temperature for the Three Instruction Groups and the Control Group. . . 300 100. Means and Standard Deviations for Face Temperature................ 316 101. Means and Standard Deviations for Finger Temperature..............................317 102. Means and Standard Deviations for Systolic Blood Pressure . .. . . . . . .. . 318 103. Means and Standard Deviations for Pulse Pressure. ........ ............. 319 104. Means and Standard Deviations for Mean Diastolic Blood Pressure ............... 320 105. Means and Standard Deviations for Oral Temperature. ............... 321 106. Means and Standard Deviations for Mean Respiration Rate...........................322 107. Means and Standard Deviations for the Mean Minus the Pretest Respiration Depth.......................................... 323 108. Means and Standard Deviations for the Mean Respiration Depth............ . . . 324 109. Means and Standard Deviations for the Heartbeat Rate............ 325 110. Means and Standard Deviations for the Mean Minus the Pretest Reading for GSR ................................326 xv ii Table Page 111. Means and Standard Deviations for the G S R .................. . . . . 327 112a. Means and Standard Deviations for the Psychological Test of Anxiety 338 112b. Means and Standard Deviations for the Psychological Test of Anxiety 339 112c. Means and Standard Deviations for the Psychological Test of Anxiety ...... 340 112d. Means and Standard Deviations for the Psychological Test of Anxiety ...... 341 113. Results of Analysis of Variance for the S-R Inventory of Anxiousness................342 114. Results of Analysis of Variance for the Affect Adjective Check List Administered Prior to the Major Testing with the Instructions to Score it According to Their Usual Feelings During an Examination {Part II of the Student Survey)................. 343 115. Results of Analysis of Variance for the Test Anxiety Scale. 344 116. Results of Analysis of Variance for the Manifest Anxiety Scale....... ... 345 117. Results of the Analysis of Variance for the Affect Adjective Check List Administered After the Instruction Stimuli . . .... . . 346 118. Data for AACL Administered as a Part of the Student Survey ........... ...... 347 119. AACL Administered After the Anxiety Treatment................... 348 xviii LIST OF FIGURES Figure Page 1. Treatment Groups Design (Campbell Stanley notation) . . ................ 94 2. A Student Performing. ................... 98 3. The Grass Polygraph................. 99 4. A Portion of the Polygraph Record . . . . . 102 5. The Statistical Model Within Which The Data was Analyzed . 103 6. Means for the Total, Verbal, and Mathematical Scores on the Academic Ability Test for the Anxiety- Reducing, Neutral, and Anxiety- Producing instruction Groups, and for the Control Group for Test Performance . „ .......... ........ 220 7. Means for the Total Test Scores on the Academic Ability Test for the Anxiety-Reducing, Neutral, and Anxiety-Producing Instruction Groups and the Control for Test Performance. . . 221 8. Means for the Verbal Scores on the Academic Ability Test for the Anxiety- Reducing, Neutral, and Anxiety-Producing Instruction Groups and the Control for Test Performance for Boys and Girls . 222 9. Means for the Mathematical Scores on the Academic Ability Test for the Anxiety- Reducing, Neutral, and Anxiety-Producing Instruction Groups and the Control for Test Performance. . . 223 xix Figure 10. 11. 12. 13. 14. 15. 16. Page Unadjusted and Adjusted Means for the Total Test Scores with the Initial Respiration Rate Held Constant for the Three Types of Instructions................................ 301 Unadjusted and Adjusted Means for the Verbal Scores with the Initial Respiration Rate Held Constant for the Three Types of Instructions ..... 302 Unadjusted and Adjusted Means for the Mathematical Scores with the Initial Respiration Rate Held Constant for the Three Types of Instructions................................ 303 Unadjusted and Adjusted Means for the Total Test Scores with the Initial Respiration Depth held Constant for the Three Types of Instructions. ............ 304 Unadjusted and Adjusted Means for the Verbal Scores with the Initial Respiration Depth Held Constant for the Three Types of Instructions. . ............................305 Unadjusted and Adjusted Means for the Mathematical Scores with the Initial Respiration Depth Held Constant for the Three Types of Instructions.............. 306 Unadjusted and Adjusted Means for the Total Test Scores with the Initial Heartbeat Rate Held Constant for the Three Types of Instructions. . ......................... 307 xx Figure Page 17. Unadjusted and Adjusted Means for the Verbal Scores with the Initial Heartbeat Rate held Constant for the Three Types of Instructions ............. 308 18. Unadjusted and Adjusted Means for the Mathematical Scores with the Initial Heartbeat Rate Held Constant for the Three Types of Instructions. . ■ . . . . . . ............309 19. Unadjusted and Adjusted Means for the Total Test Scores with the - Initial Diastolic Blood Presure Held Constant for the Three Types of Instructions................... 310 ; 20. Unadjusted and Adjusted Means for the Verbal Scores with the Initial Diastolic Blood Pressure Held Constant for the Three Types of Instructions.......... ... . . . 311 : 21. Unadjusted and Adjusted Means for the Mathematical Scores with the Initial Diastolic Blood Pressure Held Constant for the Three Types of Instructions. .......... 312 22. Unadjusted and Adjusted Means for the Total Test Scores with the I Initial Pulse Pressure Held Constant for the Three Types of Instructions. .......... . . . . . . . 313 I 23. Unadjusted and Adjusted Means for the Verbal Scores with the Initial Respiration Depth Held Constant for the Three Types of Instructions. ............................ 314 xxi Figure Page 24. Unadjusted and Adjusted Means for the Mathematical Scores with the Initial Respiration Rate Held Constant for the Three Types of Instructions . . . . . . . 315 25. Mean Respiration Rate ............ 328 26. The Mean Minus the Pretest Respiration Depth .. . .............. 329 27. Mean Heartbeat Rate . 330 28. Mean Minus the Pretest Reading for Galvanic Skin Response. ....... 331 29. Mean Systolic Blood Pressure. . . . . . . . 332 30. Mean Diastolic Blood Pressure . „ . . „ . . 333 31. Mean Pulse Pressure . . .. . .. . . . 334 32. Mean Oral Temperature . . . . . , 0 . .. . 335 33. Mean Face Temperature ......................336 34. Mean Finger Temperature ..... ......... 337 xxii CHAPTER I PROBLEM Eor several decades the construct, "anxiety," as a result of the impetus given by Cannon (1929) and Freud (1936), has received attention by psychologists in clinical research with abnormal individuals. More recently, however, anxiety on the basis of theoretical considerations and empirical evidence has been investi gated for normal individuals in an examination environ ment. Teachers and psychologists have repeatedly observed capable individuals whose test performance was not commensurate with their apparent ability. For decision making, the failure of individuals to achieve their highest levels of performance poses a serious social problem. Statement of the Problem The problem in this study was to investigate the following questions: (a) To what extent does anxiety influence test performance? (b) What is the relationship between various physiological and 1 ! 2 I psychological measures of anxiety? i Background of the Study To assist in making important decisions about individuals, psychologists and educators have made manifold use of a multiplicity of tests. These tests cannot be de-emphasi^ed, since it is possible for results : I of an examination to alter a person's entire future by j causing him to be denied entrance to a chosen pursuit or by revealing to him the presence of unrecognized abilities. There are many who share the belief that anxiety affects test performance. Anastasi (1961, p. 51) pointed out that "Children who become over-anxious in a test situation are thereby handicapped in their per formance. " She concluded that "test anxiety does i ' j interfere with effective learning and test performance." < i Cowen (1957) believed test anxiety could hide an ex aminee's underlying potential. Cronbach (I960, p. 54) j expressed a similar thought: "When the subject wishes I i i to earn the best score he can, his very desire to do well j | may interfere with good performance. When one is tense, | he commits errors that he would readily detect as such i i | otherwise." According to Thorndike and Hagen (1961, p. 506), "Performance under examination pressure may fail to represent the individual's competence under more relaxed and normal life conditions." S.B. Sarason and Mandler (1952) concluded that it is questionable whether intelligence test scores adequately describe the under lying abilities of individuals who have high anxiety drive in the testing situation. Since tests play such an important role in making decisions about individuals/ the results of performance on them should be as free from contaminating influences as possible. Hopkins (1961, p. 1) writes: "If tests are to have their maximum validity, elements which cause discrepancies between obtained and true scores should be identified and corrective measures sought." If a noncognitive factor, such as anxiety, causes discrepancies in test performance, it should be investigated. Much of the research in this general area has been correlational, not experimental, and has used paper-and-pencil self-report devices as indices of anxiety. Only this type of anxiety measure has been systematically investigated as a variable concomitant with mental test performance. Paper-and-pencil self-report devices are dependent upon the voluntary answers by the examinee. These responses are based on the individual's self concept, ideal self concept, and the concept he feels | others have of him. In addition, the situation is con founded by the fact that there are some feelings and attitudes that an individual is willing to admit to others, some that he is willing to admit to only himself, and still others that he does not admit even to himself. The true validity of these scales would be difficult to determine since an individual's defense system is inextricably involved. As S. B. Sarason and Mandler (1952) indicated, two individuals might have the same strong anxiety tendency but differ in their readiness, conscious or unconscious, to reveal it. People vary in methods of defending them selves against experiences of anxiety. The extent to which individuals differ in their admission to emotions, : combined with the fact that they have no common objective frame of reference from which to indicate their feelings, limits the validity of anxiety scales. Physiological measures of anxiety are not I dependent upon the subject's introspective evaluation of [ his emotional feelings. Rush (1967, p. 423) observed i | that measurement of physiological change is the most ; sensitive and objective method of studying emotions. According to Stevens (1951, pp. 473-477), any final ; description of emotion must be in terms of a reacting mechanism, confined to the emotional behavior and its 5 underlying mechanism. Verbalization is, at most, an audible indication of inner feelings; therefore, it seems desirable to work directly with the responses of the body, rather than to depend upon the individual's verbal expression. Jost (1953, p. 3) found that when the organism was stimulated before the subject responded, there were physical changes within him. The body, by its responses, is able to answer questions concerning anxiety that the intellect might not be willing to admit. Another difficulty in research on anxiety has resulted from the use of only a single physiological measure of anxiety, as in the study of Winter and his associates (Winter, Ferreira, and Ransom, 1963). Even when multiple measures have been used, as in Smith and Wenger's (1965) investigation, the concern has been with the anxiety measures per se, not with the relation ship of anxiety and test performance. According to Ruebush (1963, p. 500), relatively little is known concerning the biological antecedents and correlates of anxiety in children. It is evident that there is need for experimental study of test anxiety using multiple physiological and psychological measures and resultant performance on an ability test. The above information shows that the relation of anxiety to test performance is not entirely clear. The present study was designed to cast further light, if possible, on this issue. In order to determine the relationship between anxiety and test performance it was necessary to measure anxiety. The present study proposed to measure anxiety by introspective responses to anxiety questionnaires that are subject to conscious or uncon scious faking, as well as physiological measures that are not dependent upon voluntary responses. The study also was designed to use multiple physiological measures of anxiety rather than one or two measures, as has been done previously. The study was concerned not only with physiological responses but also with the relationship between physiological measures of anxiety and performance on an ability test. Operational Definitions Test Anxiety Test anxiety, in this study, refers to physio logical responses during the examination on the following measures; respiration rate, respiration depth, heart beat rate, galvanic skin response, systolic blood pressure, diastolic blood pressure, pulse pressure (S-D) , oral temperature, face temperature, and finger temperature; and by the scores on the following psychological measures given prior to the examinations Endler, Hunt and Rosenstein's S-R Inventory of Anxiousness (1962) (S-RI)? Zuckerman's Affect Adjective Check List (1960) (AACL) ; Irwin G. Sarason's Test Anxiety Scale (1961) (TAS); and Bendig's (1956) short form of the Taylor Manifest Anxiety Scale (1953) (MAS)» Initial Level of Anxiety The initial level of anxiety was recorded as the pretest readings for the above physiological measures taken before the anxiety stimuli were presented, j Ability Test The ability test, used as a dependent variable, j was the Cooperative Academic Ability Test (AAT) Form J "A" published by the Educational Testing Service. j i i Mathematical, verbal, and total scores were obtained. j ! i Proficiency Levels Individuals with a low proficiency level refers j to those with scores below the mean of 56-57 on the j . i composite of test 1-8 of the Iowa Test of Educational j Development (ITED). Individuals with high proficiency j level refers to those with scores above 56-57. j 8 Experimental Treatments or Test Instructions Test instruction is the critical independent variable, and refers to the experimental test instruc tions designed to elicit differences in anxiety. Three kinds of test instructions were presenteds anxiety- reducing, neutral, and anxiety-producing. In this study the terms experimental groups, treatments, treatment groups, and instruction groups will be used interchange ably. A copy of the three types of instructions used is provided in Appendix A. Hypotheses Hypothesis One Anxiety will influence test performance in the following ways % a. Anxiety-reducing test instructions will result in highest levels of performance on an ability test, b. Neutral test instructions will result in j medium levels of performance on an ability test. c. Anxiety-producing instructions will result in lowest levels of performance on an ability test. Hypothesis Two The degree of anxiety elicited by the instructions; 9 will influence test performance in the following ways when the initial level of anxiety is held constants a. Anxiety-reducing test instructions will result in the highest level of performance on an ability test,, b. Neutral test instructions will result in medium levels of performance on an ability test, c„ Anxiety-producing instructions will result in lowest levels of performance on an ability test. Hypothesis Three The degree of anxiety elicited by the instruc tions will influence physiological responses during test performance in the following ways? a. Anxiety-reducing test instructions will result in the lowest levels of physiological responses. b. Neutral test instructions will result in medium levels of responses. c„ Anxiety-producing instructions will result in the highest levels of physiological responses. Hypothesis Four There will be no difference in physiological responding between high and low proficiency groups. Hypothesis Five Boys will have higher levels of physiological 10 ' measures than girls for face temperature, finger temperature, systolic blood pressure, diastolic blood pressure, and pulse pressure. Girls will have higher levels of these measures than boys for oral temperature, respiration rate, respiration depth, heart beat rate, and GSR, Hypothesis Six There will be significant relationships within each treatment group among the following physiological measures of anxiety? respiration rate, respiration depth, heart beat rate, GSR, systolic blood pressure, diastolic blood pressure, pulse pressure, oral tempera ture, face temperature, and finger temperature. Hypothesis Seven There will be a positive relationship between the following psychological measures of anxiety? S-R Inventory of Anxiousness, Affect Adjective Check List, I, G» Sarason's Test Anxiety Scale, and Bendig's short form of Taylor's Manifest Anxiety Scale, Hypothesis Eight Within each treatment group the Affect Adjective Check List, administered prior to the ability test, will have a positive relationship with the scores on the same measure obtained immediately following the 11 i ! anxiety instructions,, Hypothesis Nine There will be significant relationships within each treatment group between the physiological and psychological measures of anxiety. Hypothesis Ten There will be significant relationships within each treatment group between psychological measures of anxiety and scores on the Academia Ability Test. Hypothesis Eleven There will be significant relationships within each treatment group between physiological measures of anxiety and performance on the ability test. Limitations of the Study 1. This study was limited to 100 seniors in one school; however, since the school closely parallels the national parameters for the California Test of Mental Maturity (CTMM) and the Iowa Test of Educational Development (ITED), the sampling should provide a basis for valid generalization. 2. Although due to the imposed limitations of the equipment, an unnatural testing situation was necessitated these conditions should not greatly limit 12 generalization, since the conditions were common to all groupso In addition, the possible influence of this factor can be assessed due to the inclusion of one group of examinees on whom physiological measures were not included. 3. Letters of approval from the parents of those taking part in the study were required by the school. This requirement somewhat restricted the selectivity of the sample, which in turn might lead to some lack of generalizability of the findings. To the extent that approval introduced a selective factor in the subjects, the results may not be representative. Significance of the Study There are profound differences among the views of psychologists as to what is meant by the construct, anxiety. One's theoretical approach to anxiety largely determines the method used to measure it. Thus, a knowledge of the relationship of the results of the various types of assessment should help to modify and refine the theoretical concept of anxiety. Getzels and Jackson (1963, pp. 574-576) point out that defini tions of the inordinately elusive concept of personality are often contradictory, and that observations based on one definition will contradict observations based on 13 another definition,, According to them, the more common definitions fall into three main categories? behavioral, social stimulus, and depth. The present study investigated the behavioral definition of anxiety by using both the self-report (psychological) and performance (physiological) types of measures. The use of multiple measures assists in clarifying the concept of anxiety by providing a description of how bodily responses in an anxiety situation relate to the indivi dual's reports of the anxiety he experiences. The investigation also shows how these self-reports of anxiety, developed around various concepts of anxiety, relate to one another; and how performance measures of physiological responses relate to one another. Finally, the investigation attempts to demonstrate the influence of anxiety on mental test performance under varying experimental conditions” . Preview The remainder of the investigation will be organized in the following manners Chapter II presents a review of the literature related to the study under investigation. Various concepts of anxiety will be examined, and research pertaining to the instruments used will be considered. 14 Chapter III provides reports of empirical investigations showing the relationships between testing variables and measures of anxiety,, Chapter IV describes the sample, instruments, procedures, and the statistical procedures used in analyzing the data, Chapter V presents the results of the investiga tion in terms of the hypotheses. Chapter VI summarizes the study, states con clusions revealed by the results of the investigation, and presents recommendations for further study, CHAPTER II REVIEW OF THE LITERATURE ON ANXIETY The purpose of this chapter is to review briefly the literature c6ncerning various concepts of anxiety and the literature on the instruments used for the measurement of anxiety. Anxiety Theories of Anxiety The twentieth century has been called the age of anxiety (Dickel & Dixon, 1957). Americans spend over ten billion dollars a year on liquor and buy hundreds of tons of tranquilizing drugs (Coleman, 1964, ppe 2, 3), and still anxiety remains one of the most perplexing problems of our day. The concept of anxiety is nebulous. It has been studied by a great many investigators and approached from a variety of points of view; e.g., the psychoanalytical (Freud, 15 16 1936), physiological (Laeey & Lacey/ 1962; Martin, 1961; and Wenger, 1957), psychological (I. G. Sarason, 1960; Taylor, 1951), and learning theory (Mowrer, 1950). In fact, the definition of anxiety varies widely among authors, depending upon theoretical framework, empirical studies, and assumptions (Ruebush, 1963, p. 461) „ Until the time of Freud and other "depth" psychologists, the problem of anxiety resided in the domain of philosophy. According to Freud (1936, pp. 90-99) , anxiety is "a specific state of unpleasure accompanied by motor discharge along definite pathways. According to him, it arises as a response to a situa tion of danger, and will be regularly reproduced thenceforward when such a situation recurs. He re garded it as an expression of helplessness and a reaction to the perception of the absence of the love object. According to the Psychiatric Glossary (American Psychiatric Association, 1957, p. 18) anxiety is "apprehension, tension or uneasiness which stems from the anticipation of danger, the source of which is largely unknown or unrecognized.1 1 Freud (1936, pp. 149 150) also saw anxiety as the anticipation of danger, helplessness, discomfort, or as a reminder, created by 17 a present situation, of a traumatic condition previously J experiencedo He implied that the danger signal may produce an infinite variety of reactions that are un- j like the one that occurred in the actual trauma of which the signal is premonitory (Mowrer, 1950, p. 20) . Pavlov's (1928, p. 52) findings were similar to those of Freud, but according to his hypothesis, a danger j signal (the conditional stimulus) elicits essentially \ the same movement reaction that was previously pro- j | duced by actual trauma (the unconditioned stimulus) j (Pavlov, 1928, p. 52)„ Goldstein (1940, p. 91), considered anxiety to correspond on the subjective side to a condition in j which the organism's existence is in danger. The catastrophic condition and the phenomenon of anxiety, i 1 in short, have a special significance for life. This catastrophic condition is that particular danger which threatens the physical or psychological life of the individual. To one student a particular examination may not be a traumatic experience, whereas to another, whose life career depends on passing the examination, it may be a catastrophic experience. j Horney (1939, pp. 194, 201) agreed with Gold- i stein that what is menaced by a danger-provoking j anxiety is something belonging to the essence of the 18 core of the personality. She stated further that anxiety emerges because the safety, security, or an essential value of the individual is endangered. Horney pointed out that in Freud's concept of anxiety in neuroses the source of danger is in the "id" and "superego," but to her, the reason is that a safety device of vital importance is endangered. Anxiety and Fear Horney (1939, pp. 194, 195) saw anxiety and fear as emotional responses to danger, but she con tended that anxiety is characterized in contra distinction to fear by a quality of diffuseness and uncertainty. Even a concrete danger, an earthquake for instance, has something of the horror of the un known. When anxiety is produced, what is menaced by the danger is something belonging to the essence or core of the personality, thus causing a feeling of helplessness toward the danger. According to Horney, the same situation may elicit fear or anxiety. If the person is afraid, but does something in an attempt to remedy the situation, the feeling would be fear. But if he is afraid and helpless to do anything about it, the experienced emotion is anxiety. The distinction between fear and anxiety, as 19 pointed out by Coleman (1964, pp. 75, 93), is that "fear tends to protect the organism by leading it to withdraw from dangerous situations," whereas anxiety is aroused by a threat to the adequacy or worth of the self. Anxiety is often referred to as "psychic pain," and can be acutely unpleasant. Danger arouses fear, but threat arouses anxiety. Martin (1961) proposed that the construct of | anxiety is similar, perhaps identical, to the reaction of fear, the neurophysiological basis of which seems ! to involve the functions of the posterior hypothalamus | and its effects upon the sympathetic nervous system, I including the adrenal medulla, the pituitary- j adrenocortical system, and perhaps the birain stem and reticular formation. j 1 Marmor (1962) tended to agree with the position | held by Majrtin that, physiologically, #ear and anxiety j are similar or identical. However, he did point out j i that the differentiation between anxiety and fear has been a source of frequent discussion with no uniform agreement about it. He stated further: "In general, the term fear is used to refer to reactions to known, tangible and objective dangers, while the term anxiety i is reserved for reactions to unknown, intangible and subjective ones." He also stated that fear most often 20 refers to present dangers, whereas anxiety usually refers to anticipated or future ones. In addition, he ! said that, physiologically, there is no difference be tween fear and anxiety. Martin (1961) and Marmor (1962) both agreed on the point that the physiological responses of fear and anxiety are similar or identical. Contrarily, Wolf and Wolff (1943, pp. 110-119) did not find the physiological responses of fear and anxiety to be the same. In a study of gastric changes accompanying emotions of their subject, named Tom, who fed himself through a fistula, they found a marked difference in the reactions to fear and anxiety. With anxiety there was hypersecretion of the gastric juice and a bright redness of the mucous membrane of the stomach because of a superabundance of blood drawn to the area. In fear there was hyposecretion of gastric juice and a paleness of the mucosa of the stomach and a paleness of the face. According to Rogers (1961, p. 346) , sympathetic stimulation of the stomach results in inhibition of motility and secretion; but with parasympathetic stimulation there is increased motility and secretion. Smith and Wenger (1965) found a sympathetic nervous system response for nine male and two female students prior to their oral doctoral examination. 21 The investigators did not use gastric secretion as a variable, but did observe a decreased salivary output. Wolf and Wolff (1943) found an increased gastric out put to be accompanied by an increased salivary output. May (1950, pp. 203-204) distinguished between fear and anxiety by stating that "The reactions of an organism in times of fear and of anxiety may be radically different, due to the fact that these re actions occur on different psychological levels of the personality." According to May, "The capacity of the organism to react to threats to its existence and its values is, in its general and original form, anxiety." As the organism becomes neurologically and psychologi cally mature enough to differentiate specific objects of danger, the protective reactions also become specific. The emotion involving differentiated re actions to specific dangers is fear. The understanding of fear depends upon the understanding of the prior problem of anxiety. May spoke of anxiety as the gener al, original response to threat on the basic level of the personality: "It is a response to a threat to the 'core1 or 'essence' of the personality rather than to a peripheral danger." He implies that anxiety de creases with maturity. This seems to contradict Cronbach's (1960, p. 54) statement concerning test 22 anxiety. As a result of a study of the literature, Cronbach found test anxiety to increase gradually through the school years. According to his findings, as an individual advances in school and in age, he becomes more mature, and the test anxiety which he experiences increases. Anxiety as a Constructive Force Mowrer (1950) took the viewpoint that anxiety performs a constructive and positive role in human development. He did not believe it to be the cause of personal disorganization, but rather an outcome or expression of such a state. He pointed out that Kierkegaard also considered that anxiety can be con structive . He stated that not only is anxiety construc tive; it is also a saving and educative experience (Mowrer, 1950, p. 545). According to both Marmor and Mowrer, the practice of regarding all anxiety as an abnormal experience to be annihilated if possible is a mistake. Marmor (1962) viewed anxiety as an aspect of normal human behavior with a psychological reaction comparable to its physiological analogue, the sensation of pain. "Both are signals to the organism that something is threatening its integrity, and both are essential 23 alerting mechanisms which enable the organism to make the proper adaptive responses." Just as an individual who lacks the capacity to feel pain would be seriously handicapped, so also would an individual who lacks the capacity to feel anxiety. On the other hand, if either tendency becomes excessive, the result can be destruc tive. May (1950, pp. 206-208) wrote of normal anxiety and neurotic anxiety. He referred to Freud, who took the position that objective anxiety is inherent in the child and is an expression of the self-preservation instinct with an obvious biological utility. This biological utility was observed by Janis (1958, pp. 395-412), who noted that his major surgical patients who experienced a moderate amount of pre operative anxiety were significantly less likely to have postoperative emotional disturbances than those who experienced extremely high or extremely low pre- i operative anxiety. Patients who denied any concern, worry, or anticipatory anxiety; who were constantly cheerful and optimistic; who slept well and showed no observable evidences of tension, were more likely than the others to display postoperative reactions of intense resentment and irritability. A moderate amount of anxiety seemed to help prepare the | individual for the coming event. | The Canadian physician Selye (1956), who made an extensive study of stress, likewise held that stress is not always something to be abolished. In his book, The Stress of Life, he implies that living a "full life" requires that a person learn to "enjoy" stress. In reviewing the previous literature on anxiety, Ruebush (1963) concluded that, according to psycho analytic theory, the major functions of anxiety in the normal individual are to signal the presence of psychic danger and to signal the withdrawal of an inhibitory response. Different situational stimuli trigger the danger signal for different individuals. Children differ not only in the number of stimuli that elicit the anxiety but also in the number and strength of unconscious processes, and in the number, strength, and flexibility of their defenses. A child who rarely experiences anxiety may have defense systems which are highly effective in protecting him from experiencing his unconscious drives. Although these defense systems are effective in this limited sense, they may or may not be effective in the role they play in the over-all personality functioning of the child. An inflexible defense system may interfere with other behavioral processes; whereas, a flexible 25 system, made up of a variety of defenses, may allow an individual to match the defense with the danger in a way that will maximize protection from danger and will minimize interference with other aspects of intellectual and personality functioning. These latter defenses may be thought of as being learned by the organism. Anxiety; A Learned Concept Anxiety is a construct that can be learned. The anxiety response becomes readily conditioned to stimuli that do not innately elicit the response. Martin (1961) explained that this characteristic complicates any attempt to define anxiety on the basis of stimuli that elicit it, since there will be wide interindividual differences among the stimuli that elicit anxiety. According to May (1950, p. 119) Freud considered the capacity for anxiety to be innate in the organism and a phylogenetically inherited characteristic. A clarification of this conflict in views was made by May (1950, p. 208) , who noted that learning psychologists tended to consider anxiety as a learned behavior since each particular fear or focus of anxiety is closely related to the individual's specific experience. On the other hand, neurophysiologists, 26 centering their attention on the given capacities of the organism, tended to assume that anxiety is not learned. May attempted to reconcile this conflict by suggesting that the capacity for anxiety is not learned, but that it is the quantities and forms of anxiety experienced by a given individual which are learned. Most individuals experience anxiety in situations in which his vital values are threatened. These values, however, are shaped primarily by learning. It might be inferred from Cronbach's (1960, p. 54) writing that anxiety is learned. He stated that it increases gradually through the school years. Mowrer (1950, p. 65) explained that anxiety is a learned response since a reduction in the intensity of anxiety functions as a satisfying state, and thereby brings about the learning of new stimulus-response sequences. Anxiety as a Drive Horney (1939) propounded that the concept of "drive" implies some compulsion from within the organism, but that the impulses and desires do not become drives except in such cases as they are moti vated by anxiety. She placed anxiety prior to the instinctual drives and the drives themselves as one of » * 27 the products of anxiety. May (1950, pp. 138-140) summarized a number of approaches to anxiety as a drive. He pointed out that "Freud conceived of environmental influences chiefly as a factor in molding instinctual drives. . „ Freud recognized anxiety as the central problem of neuroses, but did not see the all-pervasive role of anxiety as a dynamic factor driving toward certain goals. The drive property of anxiety was seen by Mowrer (1950, pp. 65, 66), a learning theorist who realized that fear, which he equated with anxiety, is pre-eminently a drive, a goad to action, and that the ensuing drive reduction can function as a "satisfying state of affairs" which provides emphatic reinforce ment of action, and thereby brings about the learning of new stimulus-response sequences. More specifically, he took the position that just as a reduction in hunger, thirst, sex drive, fatigue, lack of oxygen, or the reduction of any other organic need or discomfort tends to reinforce the behavior that brought about the reduction, so a reduction in the discomfort called anxiety is effective in fixating behavior that is associated with it. Coleman (1964, p. 75) pointed out that anxiety operates as a powerful driving force toward maintenance 28 on a psychological level. Spence (1956, p. 165) also considered anxiety as a drive. According to him, the general drive level is a function of two motivational variables, appeti- tional and aversive, both of which operate as emotional drives. The experimental variables determining drive combine in a multiplicative fashion. He assumed that the needs or drive states contribute singly and in combination to the organism's general drive level. This concept of drive is similar to Hull's. The essence of the position held by the group in Iowa is similar to that held by Mandler and Sarason (1952) in that both groups considered anxiety within the stimulus-response theory, and as Nicholson (1958) pointed out, both attributed drive properties to anxiety. Mandler and Sarason (1952) held that anxiety can serve as a strong response-produced stimulus with the functional characteristics of a drive. Primarily, anxiety drive elicits responses that tend to reduce the drive. This reduction can be accomplished in either of two ways. In terms of the test situation, anxiety can be reduced by competing responses which are test-irrelevant and may be manifested as feelings of inadequacy, helplessness, heightened somatic reactions, 29 anticipations of punishment, loss of esteem, and attempts to leave the test situation. On the other hand, it may be reduced by facilitating responses which are test-relevant and lead to completion of the test or task. The two groups concluded that anxiety ip. the testing situation is an important variable in test performance. They also stated that "It is questionable whether intelligence test scores adequately describe the underlying abilities of individuals who have high anxiety drive in the testing situation, particularly since the relation of the type of the test to the test performance seems to play an important determining role." The present study has addressed itself to the task of investigating the influence of anxiety on test performance. A number of instruments, both psycholog ical and physiological, were used to measure the construct of anxiety. Research on Instruments Used Task Performance The examination used to measure performance under various instruction stimuli was the Cooperative Academic Ability Test Form "A" (1963) published by the Educational Testing Service. This test was designed 30 for use with college-bound students and yields a verbal, a mathematical, and a total score. Reliability. The Educational Testing Service (1964) provides information concerning the reliability and validity of Form "A" of the Academic Ability Test. Internal consistencies of the test as computed for grade twelve by the Kuder-Richardson Formula 20 were .88 for the verbal score, .92 for the mathematical score, and .94 for the total score (total score 100 items) . Validity. The Educational Testing Service provides data to show that "Correlations between parts of the Academic Ability Test and the Scholastic Apti tude Test were found to be .83 for the verbal score and .96 for the mathematical score" (Educational Testing Service, 1964, p. 11). Measure of Proficiency The measure of proficiency was the subject's scores on the Iowa Test of Educational Development, published by Science Research Associates (1962). Although this test is accepted as a high school level achievement test, it was developed to measure broad intellectual skills, understanding, and ability to apply learned material, rather than recall of specific facts. 31 Buros (1965, p. 15) reported the composite score to be a good predictor of college grades. Psychological Measures of Anxiety Four psychological paper-and-pencil tests of anx iety were used. These were stapled together under the title "Student Survey" and designated as Parts I, II, III, and IV. (A copy of this "Student Survey" is provided in Appendix A.) Part I; S-R Inventory of Anxiousness (S-R I) Part I of the Student Survey was the S-R In ventory of Anxiousness by Norman S. Endler of York University, Toronto; J. McV. Hunt of the University of Illinois? and Alvin J. Rosenstein of the Psychological Corporation, New York City; (Endler, Hunt, and Rosen stein, 1962) . The inventory employed fourteen stimuli quantified on a five-step scale ranging from none to very much. The scale was designed to measure responses associated with many experiences by using the same stimuli for each experience. One of the experiences was the examination situation. The test was introduced by the experience the experimenter wished to investigate. Following is an example of the questions posed. 32 "You are Taking an Important Examination" 1. Heart beats faster 1 2 3 4 5 Not at all Much faster Two changes in wording were made on the advice of the principal of the high school, who believed that some of the students would not understand the meaning of certain words. Since the test was not meant to measure vocabulary proficiency, and since any expres sion of response would be invalid without comprehension of the nature of the task, the changes seemed justified. In No. 4, "Peel exhilarated and thrilled," was modified to read, "Peel excited and thrilled," and in No. 10, "Become immobilized," was changed to, "Become unable to move." Studies by Endler and collaborators with the S-R I established the following figures on the test's reliability and validity: Reliability. From Cronbach's Coefficient Alpha, the reliability was .87 with the S-R I for a final examination in an important course. Validity. The correlation between the total score of the S-R I for test anxiety plus the ten other experiences investigated was .46 with the Taylor Manifest Anxiety Scale, .66 with the Mandler and 33 Sarason Test Anxiety Quotient, and -.06 with the Palmar-Sweat Index (PSI) (post stress). As can be observed, there was little discernible relationship between the S-R I paper-and-pencil test and the PSI (Endler et al., 1962). Part II: Affect Adjective Check List (AACL) Part II of the Student Survey was the Affect Adjective Check List developed by Marvin Zuckerman (1960). The scale is composed of a list of 61 adjectives with affective connotations collected from Gough’s and Nowlis1 lists and from a thesaurus. The scale is designed to be checked either the way the individual feels at present (the "today" scale) or the way the individual usually feels under the indicated circumstances (the "general" scale). Only 21 of the adjectives are actually scored. Eleven anxiety-plus words, such as afraid, are scored 1 if marked true and ten anxiety-minus words, such as calm, are scored 1 if marked false. Reliability. The "today" scale of the AACL has an estimated reliaiblity of .85 by the Kuder- Richardson Formula 20, and .31 for retest. It seems plausible that the responses at a given time were quite stable, but that interpretation of the retest results 34 do not confirm consistent feelings from one time to another. The "general" scale, used to indicate one’s general feelings, yielded a reliability of .72 by the Kuder-Richardson Formula 20, and .68 with retest (Zuckerman, 1960). Validity. The validity coefficient was .28 between the mean of three examination days for the AACL and the MAS. This correlation was not statistic ally significant. However, for the first examination day the correlation was .40 which was significant at the *05 level (Zuckerman, 1960). The difference be tween the average of five pre-examination AACL scores for each subject and the examination day AACL scores provided a t of 1.76 which was significant at the .05 level using a one-tailed test (Zuckerman & Biase, 1962). Winter and collaborators (1963) found the correlation between Taylor's Manifest Anxiety Scale and the AACL to be .44, p 4, .05, with a one-tailed test, but only an insignificant correlation of .03 between the AACL and the PSI which utilized ferric chloride solution and filter paper impregnated with five per cent tannic acid. As can be seen, there was a lack of congruence between the paper-and-pencil test and the physiological measure of anxiety. Part XIX; Test Anxiety Scale (TAS) Part III was the Test Anxiety Scale developed by Irwin G. Sarason (1959) of the University of Washington. The rationale supporting the development of the Test Anxiety Scale (TAS) was based on the observation that individuals are not anxious every minute of the day, and often one can specify the conditions which will lead to an increase in anxiety. Sarason's theory was that a scale designed to measure the anxiety experienced in a testing situation is more related to achievement than a scale designed to measure general or manifest anxiety (I. G. Sarason, 1957, 1960). The questionnaire is composed of sixteen true or false statements describing responses related to examinations. For instance: 1. While taking an important examination, I perspire a great deal. 2. I freeze up on things like intelligence tests and final exams. Validity. Walter, Denzler, and Sarason (1964) obtained correlations of .43 and .34 between test anxiety and general anxiety for boys and girls in grade ten, and .57 and .37 for high school senior boys and girls. However, the tenth graders were from a higher socioeconomic status than were the seniors. 36 I. G. Sarason (1961a) found similar correla tions between his Test Anxiety Scale and Taylor's Manifest Anxiety Scale. These correlations were .46 for college men and .53 for college women. Part IV; Taylor's Manifest Anxiety Scale (MAS) Part IV is Bendig's (1956) short form of Taylor's Manifest Anxiety Scale (Taylor, 1953). Originally, five clinical psychologists selected items from the Minnesota Multiphasic Personality Inventory that possessed face validity for measuring manifest anxiety. The resulting sixty-five items were then re duced to fifty on the basis of internal consistency. Bendig (1956) further reduced the scale to twenty items; This scale imposed such true or false statements as: 1. I believe I am no more nervous than most other s. 2. I work under a great deal of tension. Bendig (1956) shortened the long form of Taylor's Manifest Anxiety Scale on the basis of the results of two studies on the validity of the fifty individual items of the scale. One study by Hoyt and Magoon (1954) had eight clinical psychologists rate 289 college students, whom they had counseled, for manifest anxiety, and selected eighty-eight from the extreme top and eighty-six from the extreme bottom of 37 the rating continuum. The two groups were further dichotomized to provide replicated samples. For both groups, sixteen items discriminated between the high I and low anxiety subjects at the .05 level. Seventeen items discriminated in one pair but not the other pair of the sample, and seventeen were not significantly related to clinical ratings of anxiety, j The other study reported by Bendig was con ducted by Buss (1955). He had four clinicians rate sixty-four psychiatric subjects on the Manifest Anxiety | | Scale, and compared the responses of the twenty-two j high-anxious and the twenty low-anxious subjects. i Fourteen items on the scale discriminated between the high and low anxiety subjects at the .05 level, whereas thirty-six items did not discriminate between the two j groups. Perhaps the reason for only moderate clinical validity was the rationale used for the construction of the scale. The items that clinically defined f I anxiety were eliminated from the Taylor Manifest i I Anxiety Scale. From the data reported by Hoyt and | Magoon (1954) and Buss (1955), Bendig (1956) selected twenty items to constitute the short form of the Taylor Manifest Anxiety Scale. I Bendig (1956) reported the reliabilities for | the short form to be .75 for the males, .74 for the 38 females, and .75 for the total group. Neither the differences between the means nor variances were statistically significant for any of these groups. Differences between the sexes for the two groups were also nonsignificant. Validity. I. G. Sarason (1961a) found the Taylor Manifest Anxiety Scale to correlate with his Test Anxiety Scale .46 for college men and .53 for college women. The Taylor Scale correlates .44 with the AACL (Winter et al., 1963) , and .66 with the S-R Inventory of.Anxiousness using the total score with the test anxiety plus ten other experiences (Endler et al., 1962). There was no significant correlation with the Palmar-Sweat Index (PSI) (Endler, et al., 1962; Winter et al., 1963). Concluding Remarks Paper-and-pencil measures of anxiety: a. are easily administered, b. can be presented to an entire group or to an individual, c. are objective and are therefore easily scored. In spite of these important assets of the scales, there are some serious reservations concerning their validity. 39 Aiken, Jr. (1962) pointed out that one problem with anxiety scales is the numerous definitions of anxiety offered. According to Getzels and Jackson's (1963, pp. 574-576) theory, measurements developed around one definition may be contradictory to measure ments developed around a different definition. The measures in the present study were chosen because of their rationale for the development and the method of construction. Another problem with anxiety scales is the possibility of conscious or unconscious faking. S. B. Sarason and Mandler (1952) realized the weaknesses in using anxiety questionnaires to assess anxiety, pointing out that two individuals might have the same degree of anxiety but differ in their readi ness, conscious or unconscious, to reveal it. X. G. Sarason (1960) considered convenience to be probably the major reason for the wide use of paper-and-pencil indices of anxiety and felt that "While convenience is a desirable characteristic, research is needed to investigate less convenient but perhaps more useful indices." In summing up his position, he stated, "Perhaps the most parsimonious statement that one can make concerning what is measured by existing scales of anxiety is that they measure the extent to which an individual is willing to admit to 40 experiencing anxiety in certain situations." Perhaps physiological measures are the less convenient but more useful indices. Physiological Measures of Anxiety Physiological measures of anxiety, unlike psy chological measures already discussed, are not dependent on the process of externalized introspection. According to Jost (1953), the physical changes due to anxiety in a human being take place before any verbal response. However, though physiological measures are relatively objective and free from faking (conscious or unconscious), it cannot be said that a physiological response to a given anxiety stimulus is an absolute measure of anxiety. Clearly, there are several factors besides anxiety which manifest themselves in the form of physical responses. Among these factors are: environmental temperature, muscular exercise, digestion, fever, general level of excitement, sleep, time of day, and seasons (Best & Taylor, 1961, pp. 274-276, 302, 884, 1270? Mac Bryde, 1944, p. 1367? Wenger, 1943). Luria (1932, pp. 46-76) was among the first to study physiological responses in connection with test-taking anxiety. He "applied the method of associated motor reactions, by giving the subject speech 41 speech stimuli/ and recording the speech responses connected with simultaneous motor pressures." The reactions to thirty word-stimuli were taken for each subject by means of a dynamoscope. Of the thirty words, eight were considered as critical {pertaining directly to the testing siutation), nine were doubtful, and thirteen were indifferent. One study (Luria, 1932, pp. 46-76) was concerned with the "cleansing" of "purgation" that was required at the university. Each student was called in before a special commission that reviewed the student1s academic record, his social-political inclinations, his academic activity, and then made its decision. An unfavorable judgment meant the student was expelled from the university and, as a result, all his work and future plans came to naught. Luria considered this experience more traumatic than a regular school examination. Thirty students, nineteen women and eleven men, were taken directly from the line awaiting the stressful event and tested. Part of the group was also tested again after the examination. Luria found the average reactive time to depend upon the character of the stimulus used. Approximately thirty per cent more time was required to respond to the critical stimuli than was required in the case of the indifferent stimuli. 42 The presentation of the stimuli directly connected with the traumatic event usually produced an obstruction of the associative processes and a marked disturbance of the motor reactions. He was led to believe that the affective influence of the situation was more closely connected with the situation of expecting the trauma than with the trauma itself. Another study he pursued was concerned with an ordinary school examination. The results of the ordinary school examination were similar to those of the "cleansing." Even though the examination was less traumatic, the psychological structure of both were analogous. The average reaction time for the "cleans ing" was 2.29, whereas for the examination, it was 2.2. Normally the time does not exceed 1.5-1.7. He also found "a great disturbance of the accompanying motor reactions corresponding to the marked delay of the associative process." The movements indicated that the delays were not connected with the simple lowering of the energetic tones of behavior as would occur in fatigue or in drowsiness, but they were the result of a diffuse excitation, which broke the normal associa tive process. Not all students showed the same responses. Approximately thirty per cent of the subjects presented a picture of intense reactive lability, whereas twenty- five per cent showed reactive stability,, Luria's first supposition was that some feared the examination, while others, feeling secure and well prepared, did not, and thug were stable. Controlled experiements did not show this to be true. The symptoms obtained were almost identical in the well-prepared and in the incompetent students. He felt the degree of fitness apparently did not influence the affective reactions. The problem he uncovered in the 1920's is a problem still being^in- vestigated today. He divided his students into two groups: (a) those who became quite excited before an examination and showed neurotic symptoms; (b) those who did not become excited before an examination and did not show neurotic symptoms. His belief was that a large propor tion of students who become excited before an examination were high in neurotic tendency, and a large proportion of those who did not become excited were low in neurotic tendency. Brown (1938b) tested this theory by correlating scores on a questionnaire (Brown, 1938a) designed to measure test anxiety, with those on Willoughby's Clark Revision of tl>e Thurstone Personality Schedule. He found a moderate relationship between neurotic tendency 44 and emotional reactions in students before examinations/ but he considered the correlation to be far too low to warrant regarding examination neurosis as merely a special case of general neurotic tendency. He pointed out a few reasons why he thought one student | more than another might become excited before examina- I tions: (a) grades may be stressed more in his home/ (b) he may need high grades to get into the professional ; school he wishes to enter, (c) or his older brothers may b'e better students than he. He also correlated the scores on the question naire with changes in physiological measures (Brown & I Van Gelder, 1938) and found no correlation between them. Despite the lack of relationship between the two types i of measures, he did not consider that this invalidated i either the questionnaire or the physiological measures. I He pointed out that the questionnaire was concerned i with examinations in general, whereas the physiological l measures were taken before a specific examination. In order to determine the relationship between ! scores on the questionnaire and achievement, Brown and . i I Van Gelder used partial correlation with intelligence I | held constant. The obtained correlation was -'.19 + .05. He considered the correlation to be low but | highly suggestive of lower examination grades in general for those who become excited before an examina tion. This is in direct contrast to cannon's (1932/ p. 311) emergency theory. Brown pointed up the fact that, according to Cannon, students who are emotionally excited before an examination should perform better than those who are calm before an examination, other : things being equal. Lacey, Bateman, and VanLehn (1953) recorded multiple physiological measures on eighty-five male icollege students under four conditions of stress in an investigation of response specificity, Though this ,particular investigation did not involve an actual ■ examination, two of the four stress stimuli imposed on the subjects, mental arithmetic and letter association, :did require mental activity closely akin to that ;required=by the usual achievement or aptitude-measuring examination. Palmar conductance, heartbeat rate, and heartbeat rate variability were measured while the subject underwent hyperventilation, the cold pressor |test, and the previously mentioned mental activities. |The study supported the principle of relative specifi- ] !city of response-level scores to the extent that, for given autonomic functions, there existed quantitative variation among individuals in the degree to which a response pattern was stereotyped. 46 | In 1961 Wenger, Clemens, Coleman, Cullen, and i | Engel, using autonomic variables, retested autonomic response specificity on male college students. Among the physiological measures used were: electrical skin I resistance, heartbeat rate, respiration rate, finger ; and face temperatures, and systolic and diastolic : blood pressure. These measures were taken while the subjects underwent hyperventilation, the cold pressor i test, letter association tasks, and mental arithmetic tasks. This study supported the conclusion arrived at in the previously mentioned investigation conducted by Lacey and his colleagues (1953). Ax (1953) and Schachter (1957) have shown that the sympathetic nervous system is extensively involved during anxiety. According to Brown and Van Gelder ; (1938), it is "common knowledge" that examinations j "increase the excitment of students," and that the degree of increase varies directly with the severity and importance of the examination. If, then, examina tions do increase anxiety, the responses of the sympathetic nervous system during test performance I should serve to indicate levels of anxiety. In the present investigation, five types of phys- I iological measures were recorded. A brief review of the literature for each measure follows. 47 Respiration In 1929, Cannon (p. 211) found that animals experiencing pain and emotional excitement showed deep and rapid respiration. Seven years later, Freud (1936, p. 91) considered the respiratory organs, along with the heart, to be the most common and the most definite physiological indicators of anxiety. Though neither Cannon nor Freud assessed anxiety induced by an examination-taking situation, | Brown and Van Gelder (1938) found similar results in i | studies of anxiety produced by examinations of varying | degrees of difficulty. They found a statistically significant increase in the subjects1 respiratory rates before taking the difficult examinations. Clemens (1957) investigated the responses of the autonomic nervous system under the influence of epinephrine injections. If during anxiety, there is an increased flow of epinephrine, it seemed probable that, with other factors held constant, the responses during anxiety would be in the same direction as those under the influence of injected epinephrine. Clemens discovered that the subjects' responses to the injection were an initial decrease in respiration rate followed by an increase in respiration rate during later time intervals. 48 Rogers (1961, p. 346), whose findings agree with | those of Clemens, claimed that sympathetic stimulation dilates the bronchi, Fulton (1955, pp. 238, 239) held i j that when the organism is called upon to cope with a ! sudden emergency, the sympathetic nervous system causes j | a secretion of great quantities of epinephrine. If this i greater quantity of epinephrine results in a dilation | ! of the bronchi, surely respiration, or at. least the amount of oxygen intake would be influenced. j Heartbeat Rate Wenger, Clemens, Coleman, Cullen, and Engel i (1961) observed a slightly higher heartbeat rate (66 ! i beats per minute) when subjects performed mental arith metic tasks than when making letter associations (64 | beats per minute). Lewinsohn (1956) found his subjects' heartbeat rates increased from a mean base level of 88.10 i beats per minute to a mean stress level of 93.5 beats per | | minute while taking a modified form of the Digit Symbol i test under failure stress conditions. increased heartbeat rate was also observed by Brown and Van Gelder (1938) in subjects just before i | taking a senior comprehensive examination in psychology I } at the University of Chicago (a mean increase of 22.77 j beats per minute above normal). After the examination. 49 the subjects' mean heartbeat rate showed only a 5<,43 beats per minute increase above normal. The same measures were taken on a group of second year medical students prior to their taking a relatively less important quarterly pharmacology examination. In this case the mean pulse rate rose to only 8.13 beats per minute above normal. Chambers (1962) recorded the pulse rates of a group of medical students taking a sectional examina tion in pharmacology. The mean increase of 6.4 beats per minute was similar to that of 8.13 found by Brown and Van Gelder in a similar situation. Smith and Wenger (1965) recorded heartbeat rates for eleven doctoral students just before a preliminary oral Ph.D. examination. Their findings confirmed those of Brown and Van Gelder (1938) and Chambers (1962). They recorded a mean heart period of 116.65 mmin/lON (85.7 heartbeats per minute) for the examination reading in contrast to a mean heart period of 145.90 mmin/lON (68.5 heartbeats) for the control reading. Harleston, Smith, and Arey (1965) investigated anxiety measured by heartbeat rate. They found no significant difference between low, medium, and high anxious individuals while solving anagrams. 50 Galvanic Skin Response (GSR) and Other Epidermal Measures One of the most popular measures of autonomic activity associated with affective and emotional states as evaluated by Stevens (1951, pp. 474, 475) is the galvanic skin response (GSR). He states that, "The GSR, perhaps more than any other indicator of bodily change (with the possible exceptions of blood pressure and heartbeat rate), is a sensitive index of cortical and higher-level mental functions." According to Stevens and others, the nerve supply of the sweat glands is exclusively sympathetic, but the neurohumoral agent at the effector is acetylcholine rather than the usual adrenergic substance. The resistance of the skin involved in the GSR is believed to be due to a polarization-capacity effect that varies as a result of the sweat gland activity. Silverman and Powell (1944, p. 300) explained that "Emotional, intellectual and sensory stimuli will cause a type of sweating involving palms, soles and axillae." They state further that "normally emotional sweating is commonly seen in states of anticipation and has sometimes been referred to as anticipatory sweating. A student before an examina tion . . . will show sweating, particularly of the palms." 51 Darrow (1936) considered palmar galvanic skin reflex and blood pressure to be preparatory and facilita- tive reactions and ones that are especially valuable as indicators at the more moderate levels of "adaptive mobilization," In sleep, he pointed out, the resistance tends to be high and the conductance (the reciprocal of skin resistance) low, Stevens (1951) also explains that". . . the resistance level frequently increases steadily as a subject relaxes, and rises to still higher levels during sleep." He considers this to be a disadvantage of the resistance measures, for, if two responses of the same magnitude are superimposed on different resistance levels, they will not be recorded as equal. For this reason some type of resistance change should be employed. Test anxiety and skin conductance were investi gated by Kissel and Littig (1962) in an introductory psychology class at the University of Buffalo. They measured palmar skin conductance while the subject worked on a modified form of Feather1 s perceptual reasoning task. High test anxiety was more related to high skin conductance (reciprocal of resistance) under conditions of failure than was low test anxiety (p < .01) . Clemens (1957) gave subcutaneous injections of 52 USP epinephrine (o,3 cc in 1/1000 saline) to 45 male subjects and observed an increase in skin conductance. In a study conducted by Smith and Wenger (1965) palmar skin conductance was recorded for doctoral stu dents just prior to their preliminary oral examination. The mean palmar conductance just prior to the oral examination was higher, but not significantly higher, than were the readings for the comparison test. Berry and Martin (1957) attempted to determine a relationship between GSR, the kind of instructions presented to the students, and Sarason’s test anxiety scale. The subjects were presented with one of three types of instructionss apprehension arousal, neutral, or reassurance followed by an identical treatment in a conditioning situation. The findings supported the hypothesis that instructions affected the extent of GSR conditioning. There was a significant difference in the reactions of the sexes caused by instruction interaction. For the males, reassuring instructions resulted in lessened conditionability, whereas the opposite was true for the females. The Sarason test anxiety scale was not significantly related to GSR conditioning. The following investigation used GSR and test anxiety and found a relationship. Winter and other collaborators (1963) 53 investigated the relationship between the Palmar-Sweat Index (PSI) and the Affect Adjective Check List (AACL) under six experimental conditions, two of which were classroom examinations„ Two weeks after the last ex perimental session, the Taylor's Manifest Anxiety Scale (MAS) was administered. The combined AACL scores for all six conditions related significantly with their scores on the MAS (r = .44 p ^.OS). The PSI was not significantly related to either the AACL or the MAS. This low correlation of a physiological measure with pencil-and paper-test of anxiety is a usual finding. Blood Pressure Darrow (1936) considered blood pressure, along with the GSR, to be probably the best indicator of the facilitative, preparatory, or emergency functions that are mediated predominantly by the sympathetic nervous system. According to Best and Taylor (1961, pp. 274- 275), excitment, fear, and worry affect markedly the arterial blood pressure, especially the systolic. They report that the systolic blood pressure for boys about age 17 reaches 120 mm Hg. The systolic blood pressure for girls of the same age is approximately 4 or 5 mm lower. The average diastolic blood pressure of an 54 adult young male at mental and physical rest and in a sitting position is 80, and the average pulse pressure is 40. However, the normal range of blood pressures may be from 90 to 120 mm for systolic and from 60 to 80 mm for diastolic blood pressure (Best & Taylor, 1961, p. 274). A sex difference in blood pressure was observed by Milliken (1964, pp. 309-311). He investigated blood pressure change as it related to increased problem difficulty in mathematical and verbal areas. The male subjects who had high test scores on both the mathemat ical and verbal sections had the smallest increase in blood pressure. The female subjects who scored high on both sections had the greatest increase in blood pressure. Milliken found that both sexes increased in anxiety under stress to the extent that they exhibited a mathematical deficit. However, the males who scored high on the mathematical section but low on the verbal section reacted with greater physiological changes during -the mathemati cal testing than during the verbal testing. The female subjects of this group were only slightly more anxious, as measured by blood pressure, during the mathematical testing than during the verbal sections. Rather than studying blood pressure change with difficult and easy sections of an examination as Milliken did, Brown and Van Gelder (1938, pp. 1-9) 55 studied blood pressure changes as related to difficult and easy examinations. The change in the mean systolic blood pressure before a final comprehensive examination was 15.11 as compared to 3.60 before a quarterly examina tion. The diastolic blood pressure change was in the same direction as the systolic, but the change was less. Before the comprehensive, the diastolic change was 4.00 compared with -0.50 before the quarterly examination. Milliken and Spilka (1962) also found higher blood pressure during performance on a difficult mental task than during performance on an easy task. A comparison was made by Smith and Wenger (1965) of the blood pressure readings recorded on students before taking an oral doctoral examination. The mean systolic blood pressure before the examination was 126.05 compared with 116.23 for the comparison reading. The mean diastolic pressure was 80.09 for the examination reading and 76.14 for the comparison reading. The difference between the two readings was significant at the .01 level for both the systolic and the diastolic blood pressures. Wenger and collaborators (1961) observed blood pressure for thirty-six male university students. They found a higher systolic blood pressure and pulse pressure when the subject was performing letter association tasks 56 than when performing tasks requiring mental arithmetic. The reverse was true of diastolic blood pressure. The systolic blood pressure was 133 for letter association and 111 for mental arithmetic. The pulse pressure was 70 for the letter association and 32 for mental arithmetic. The diastolic blood pressure was 63 for letter association and 72 for mental arithmetic. Prom the studies just cited it would appear that the systolic blood pressure, rather than the diastolic blood pressure, would be more likely to rise during mental stress. Temperature Wynn (1919) discovered an elevation of oral temperature in draftees at the time of their physical examination. The average temperature of the 324 men observed was 99.3° P. In a second study, in which he recorded the temperatures of nursing applicants taking a registration examination, Wynn found a .60° P. elevation in temperature before the examination for two-thirds of the group, and a .5° F. depression in temperature for three-fifths of the group after the test. The temperature of doctoral students was recorded by Smith and Wenger (1965) prior to their oral examina tion and a comparison test one month earlier or later. 57 The mean comparison temperature was 98.65° F. whereas the reading before the oral examination was 99.16° F. Medical students had an even greater difference in temperature. In a study conducted by Chambers (1962), medical students were found to experience a mean oral temperature increase of 1.26° F. while taking a pharmacology examination, This increase was measured as the difference between the mean examination readings and the mean control readings taken at approximately the same time of day during a week when the subjects were not taking examinations. Lucio and Wenger (1961, pp. 35-61) recorded twelve measures on student teachers and found finger temperature to show a significant positive relation ship with teaching performance. An investigation by Wenger and collaborators (1961) resulted in higher face temperatures than finger temperatures, and both were higher for letter associa tion tasks than for mental arithmetic. Their readings for finger temperature were 29.7 and 31.2 for mental arithmetic and letter association respectively, while the face temperatures were 32.2 and 32.6 respectively. As can be noted, these differences were only slight. Ax (1953) recorded physiological measures during 58 fear and anxiety for 43 subjects. He found increases in face temperature to be greater during fear than during anger, and decreases in finger temperature to be less during fear than during anger, Schachter (1957) showed the drop in face temperature during anger to be less in hypertensives than in normotensives. Since the emotional experi ence of an examination may be anxiety, fear, or even anger, or, indeed, a combination of any two or all three of these, the above studies indicate that a subject's face and finger temperatures would depend not only upon the degree of emotion experienced but also upon the type of emotion experienced to the greatest degree. > Clemens (1957) found an increase in forehead temperature, and decreases in finger and hand temperatures of subjects just previously injected sub- cutaneously with USP epinephrine (0.3 cc in 1/1000 saline). Stevens (1951, p. 477) suggested skin temperature as an indicator of emotion. He says emotional stress is reported to produce a fall in skin temperature. "Conflict is associated with vasoconstriction and a fall in temperature, whereas uninhibited action and emotional security are said to result in vasodilatation and a rising skin temperature." 59 Best and Taylor (1961, p. 884) present some interesting figures on body temperature. The oral temperature for a large sampling of normal subjects ranged from 96.6° F. to 100.0° F. with a mean reading of 98.4° F. Intraindividual and interindividual | variations in temperature occur according to the time l of day. A person's temperature may be two or three degrees Fahrenheit higher in the late afternoon or evening, when body temperature is customarily at its | | maximum, than during the hours between three and five in the morning when it is usually at its minimum. Cannon (1932, p. 177) gives slightly different I figures for this 24 hour temperature variation cycle. He reported a variation span of 1.8° F„, setting the maximum about 4:00 P.M. and the minimum about 4:00 A.M. Because of this fluctuation of body temperature, it is important that research plans take into account the time of day the subjects' temperatures are recorded. Summary Theories of Anxiety Anxiety has been studied by Freud, Rank, Hornery, May, Mowrer, James, Goldstein, and many others. It has been approached from many points of view: e.g., the 60 analytical, physiological, psychological, and philo sophical, to mention a few. It is one of the more perplexing problems of our day. Its consequences touch individuals in every walk of i|.ife. There are almost as many theories of anxiety as there are individuals who have studied it. One view, held by many, is that it is the anticipation of danger (Freud, 1936, pp. 149-150). In general, the term fear refers to reactions to known, tangible and objective dangers, while the term anxiety refers to reactions to unknown, intangible and subjective ones (Marmor, 1962). According to Horney (1939, pp. 194, ! i | 195), the same situation may elicit fear or anxiety. I If the person is afraid, but does something about the situation, the feeling is called fear. But if he is | afraid and helpless to do anything about it, the | | experienced emotion is called anxiety. i Although anxiety may have a destructive power, it can also have a constructive force (Janis, 1958, pp. 395-412; Marmor, 1962; May 1950, pp. 206-208; Mowrer, 1950, p. 545). Anxiety has been considered by many to be a learned characteristic (Martin, 1961; May, 1950, p. 208; Mowrer, 1950, p. 65) with a drive property (Coleman, 1964, p. 75; Mandler & Sarason, 1952; May, 1950, pp. 138-140; Mowrer, 1950, pp. 65, 66; 61 Spence, 1956, p. 165) . As evidenced by the diverse and contradictory theories and ramifications of leading psychologists, anxiety and its consequences are far from being completely understood. Measure of Task Performance The Cooperative Academic Ability Test Form f "A" published by the Educational Testing Service (1964) was used as the measure of performance. The Kuder- Richardson Formula 20 reliabilities were .88 for the verbal, .92 for the mathematical, and .94 for the total score (the total 100 items) for the twelfth grade. Correlations between the Academic Ability Test and the Scholastic Aptitude Test were found to be r = .83 for the verbal score, and r = .86 for the mathematical score. j Measure of Proficiency The measure for proficiency was the Iowa Test of Educational Development (ITED) published by the Science Research Associates (1962). It was designed to measure broad intellectual skills, understanding, and ability to apply learned materials, rather than recall specific facts. Validity coefficients with college grades range between .50 and .60. 62 Psychological Measures of Anxiety Four psychological measures of anxiety were used. S-R Inventory of Anxiousness (S-RI). The S-RI, developed by Endler, Hunt, and Rosenstein (1962), employs fourteen responses quantified on a five-step scale ranging from none to very much. The scale is designed to measure responses associated with many experiences, one of which is the examination situation. i [ The Coefficient Alpha reliability was .87 with | a final examination used as the stimulus. j j The S-RI correlated r = .46 with the Taylor I j | Manifest Anxiety Scale, and r = -.06 with the Palmer- Sweat Index (PSI). Affect Adjective Check List (AACL). The AACL, developed by Zuckerman (1960) , is a list of 61 adjectives with affective connotations to be checked | either the way the individual feels at present (the "today" scale) or the way the individual usually feels under the indicated circumstances (the "general" scale) The reliability for the "today" scale was .85 by the Kuder-Richardson Formula 20 and .30 for retest reliability. The AACL correlated r = .40 (p < .05) with the 63 MAS for the first examination day and r = .28 (non significant) for a mean of three examination days. Test Anxiety Scale (TAS). The TAS, developed by I. G. Sarason (1961b) was designed to measure the anxiety associated with test-taking experiences. It is composed of sixteen true or false statements. The TAS correlated r = .46 (p ^ .05) with the MAS for male college students and r = .53 (p ^ .01) for female college students. Taylor's Manifest Anxiety Scale (MAS). The MAS was developed by Janet Taylor (Spence) (1953) and shortened by Bendig (1956). It was designed to measure the general anxiety experienced. The reliability for the short form was .75 by the Kuder-Richardson Formula 21. The MAS correlated r = .46 (p ^ .05) with the S-R I (Endler et al., 1962), r = .40 (p ^ .05) with the AACL (Zuckerman, 1960), and r = .45 (p ^ .05) for males and r - .53 (p < .01) for females with the TAS (I. G. Sarason, 1961b). Physiological Measures of Anxiety Ten physiological measures of anxiety were used. Respiration. According to Cannon (1929, p. 211), 64 animals experiencing pain and emotional excitement showed deep and rapid respiration,, In humans, increased respiration rate before an examination was observed by Brown and Van Gelder (1938)„ Clemens (1957) found injected epinephrine to produce an initial decrease in respiration rate followed by an increase in respira tion rate during later time intervals. Heartbeat Rate. An increase in heartbeat rate associated with mental activity was observed by Wenger and collaborators (1961), Lweinsohn (1956), Brown and Van Gelder (1938), Chambers (1962), and Smith and | Wenger (1965). Galvanic Skin Response and Other Epidermal Measures. Stevens (1951, pp. 474, 475) considers GSR to be one of the most popular measures of autonomic activity associated with affective and emotional states. High test anxiety scores were more related to skin conductance (the reciprocal of resistance) under conditions of failure than were low test anxiety scores (Kissel & Littig, 1962). Smith and Wenger (1965) found the mean palmar skin conductance for 11 doctoral students to be higher just prior to an oral Ph.D. examination, but not significantly higher, than the readings a month before or a month after the examination. Winter and collaborators (1963) investigated the PSI for 19 undergraduate students and found no signifi cant relationship between the PSI and the MAS or the AACL. Blood Pressure. Darrow (1936) considered blood pressure, along with the GSR, to be probably the best j I indicators of facilitative, preparatory, or emergency j functions that are mediated predominantly by the sympa thetic nervous system. Excitement, fear and worry i I affect markedly the arterial blood pressure, especially the systolic (Best & Taylor, 1961, pp. 274-275). They j report the average adult male to have a systolic blood ! i pressure of 120 mm Hg and diastolic blood pressure to be j 60 to 80 mm Hg. The average young adult female's blood pressure is approximately 4 or 5 mm lower than that of the male's. The average pulse pressure is 40. Brown and Van Gelder (1938, pp. 1-9) found a greater change in blood pressure prior to an important examination than prior to a less important examination. A higher blood pressure was also evidenced during performance on a difficult mental task than during performance on an easy task (Milliken & Spilka, 1962). i The systolic and diastolic blood pressures were | significantly higher before a Ph.D. oral examination j 66 than they were for the comparison readings. Wenger and collaborators (1961) found higher systolic blood pressure and higher pulse pressure when subjects were performing letter association tasks than when performing mental arithmetic. The reverse was true for diastolic blood pressure. The direction of systolic blood pressure seems to be better predicted than that of the diastolic blood pressure or pulse pressure. j Temperature. Stevens (1951, p. 477) suggests I | skin temperature as an indicator of emotion. Wenger | and collaborators (1961) found the face temperature to i | be higher than the finger temperature for letter I I association tasks and for mental arithmetic tasks. Clemens (1957) found an increase in forehead temperature and a decrease in finger and hand tempera ture for subjects who had just previously been injected subcutaneously with U.S.P. epinephrine (0.3 cc in 1/1000 saline). Wynn (1919) and Chambers (1962) observ ed a higher oral temperature associated with an examination. Stevens (1951, p. 477) reported emotional stress to produce a fall in skin temperature. CHAPTER III ANXIETY AND ITS RELATIONSHIP TO OTHER VARIABLES The plan of the present chapter is to present a brief review of the literature concerning the relationship of anxiety to performance* Anxiety and Performance on an Ability Test An important aspect of the study of anxiety was to determine its relationship to performance on an ability test. Manifest anxiety scales and the test anxiety scales have both been investigated. During the last few years the literature has been prolific with studies concerned with paper-and-pencil measures of anxiety and their relation to intellectual performance, Calvin, McGuigan, Tyrrell, Sybil and Soyars (1956) investigated the relationship between the Taylor Manifest Anxiety Scale (MAS) and scores on the higher form of the Otis Intelligence Test. With a group of 54 female undergraduate students, they found no significant relationship between the MAS and the Otis (r = -.02)* 67 68 Dana (1957) also found a nonsignificant correla tion. It was between the MAS and the Wechsler-Bellevue Form I Intelligence Test. This finding was true when 100 normals and 100 neurotics, similar in intelligence and education, were tested. Faber and Spence (1955, p. 10), who made an extensive study of anxiety, state that they have "been unable, over a period of years, to find any relation between the Ar-scale scores of college students and con ventional measures of intellectual ability, such as entrance-examination scores and grade-point averages." These studies made use of the Taylor MAS. Erickson (1963, p. 43), as a result of a review of the literature, found the mean score of the Childrens' Manifest Anxiety Scale (CMAS) to vary for different geographical locations. His means ranged from a low of 12.87 to a high of 23.29. If there are differences of this magnitude with the CMAS, perhaps similar differences prevail with the Taylor Manifest Anxiety Scale. This factor should be considered in making comparisons between studies in various geographical areas. I. G. Sarason (1960) pointed out that for college students, it appears, tests of the American Council on Education (ACE) type are unrelated, or only slightly 69 related to, measures of general anxiety such as the MAS. I. G. Sarason (1957) was surprised to find the most reliable studies in the literature to show anxiety to have no demonstrable relationship to academic performance. The scales that had been used were those designed to measure anxiety in general. It seemed to him that these scales ignored an important observation. People are not anxious all the time, and often, one can specify the conditions that will lead to increased anxiety. Therefore, he felt, a measure designed to i assess the anxiety under the specific conditions that i | aroused the anxiety would be more meaningful. In his i investigation he computed the correlation of the Scholastic Aptitude Test (SAT) scores with test anxiety and with general anxiety. In addition he determined the relationship between the Mathematical Aptitude Test (MAT) scores and the above two anxiety scales. Test anxiety was shown to be significantly related to performance on both tests with the correlations of r = -.14 (p < .05) for the SAT and r = .20 (p ^ .01) for the MAT. The more anxious the subject, the lower his performance tended to be. General anxiety was not significantly related to performance on either test. 70 In another study, I. G. Sarason (1961b) found significant negative correlations between test anxiety and six college entrance scores in a study with college students. However, general anxiety did not relate significantly with any of the six scores. Sarason interpreted the results as supporting the hypothesis that an anxiety scale designed to measure anxiety in a specific situation, like a test situation, is more predictive of the anxiety elicited in that situation than a scale developed to measure general anxiety. In a study with four grade levels, S. B. Sarason, Davidson, Lighthall, Waite, and Ruebush (1960, pp. 125- 135) showed the Test Anxiety Scale for Children (TASC) to be more related to intelligence test scores than the General Anxiety Scale for Children (GASC). I. G. Sarason (1959) found a negative correlation between the test anxiety scale and the 1948 Linguistic scores of the American Council on Education test. However, this relationship was only significant for the women r = -.36 (p .01). The correlation for the men, even though in the same direction, did not reach the level of significance. A positive correlation of r = .21 (p <.05) was found between scores on the Henmon-Nelson Test of Mental Ability (college level) and scores on the 71 anxiety questionnaire in a study by S. B. Sarason and Mandler (1952). Grooms and Endler (1960) attempted to do a partial replication of S. B. Sarason and Mandler's (1952) study. They trichotomized a group of 9.1 male college students at the Pennsylvania State University as to high, medium, or low test-anxious groups according to their scores on the Mandler and Sarason Test Anxiety Questionnaire. The results showed a significant negative correlation between test anxiety and the apti- ! tude scores as compared with the positive correlation (r = .21 p < .05). S. B. Sarason, Davidson, Lighthall, Waite, and Ruebush (1960, pp. 136-147) also found a signifi- | cant positive relationship. His correlation was between j the TASC and Primary Mental Abilities (PMA) . The re- ! i suits showed an overall correlation of r = .24 (p^.001),j I with test anxiety being more closely related to the test-; like tasks (word grouping test of reasoning, number) than! to the non-test-like tasks (perception, spatial) . Silverstein (1961) reanalyzed the data of S. B. Sarason and Associates (1960, pp. 136-147) in order to determine how factors of test-likeness, reading | | requirement, and cultural familiarity of the PMA related ; to test anxiety. He showed that in combination, the 72 three factors were highly related to the effects of anxiety, while in isolation only the test-likeness and cultural familiarity factors related to the effects of test anxiety. Reading requirement appeared to be related only through its relation to cultural familiar ity. The results of the investigation suggest that the influence of anxiety differs depending upon the characteristics of the task of the test. Anxiety and Achievement According to Ruebush (1963, p. 498, 499), many studies have found a negative relationship between anxiety and achievement test scores. However, there are some investigators who have reported either a positive relationship or none at all. In an investigation of the grade point averages (GPA) of 305 students at Yale University, I. G. Sarason (1957) found a significant negative correlation between test anxiety and GPA for the first two years. The GPA's for the same group in their junior and senior years did not correlate significantly with the subject's anxiety scores. At the same time, the subject's general anxiety (GAS) showed a positive relationship with GPA, significant at the .01 level for the first two years 73 and at the .05 level for the fourth year. The correlation for the third year was not statistically significant. Suinn (1965) failed to replicate the signifi cant negative correlation between test anxiety and grades that was found by I. G. Sarason (1957). Suinn's subjects were 55 students at a highly selective private college in Washington State and 70 students in a California state college. All students in both samples were freshman psychology students. Suinn attempted to explain a possible cause for the failure to replicate by pointing out that his subjects were freshman and a great majority of Sarason's subjects at the time of taking the questionnaire were sophomores and juniors. Perhaps, he suggested, test anxiety has a more detri mental influence upon academic performance for some grade levels than for others. Even Sarason's study bears this fact out. When the scores on the anxiety scale were correlated with GPA's of the students in their junior and senior years, Sarason's correlations failed to reach significance also. I. G. Sarason (1957) found a positive correlation between general anxiety (GA) and GAP. Suinn did replicate these findings of a positive correlation for his first sample, private college students, but not for his second 74 sample, state college students. Experimental Studies with Psychological | Measures of Anxiety Studies involving paper-and-pencil tests of anxiety are numerous in the literature. Only a sampling of studies will be mentioned. The measures discussed will not be restricted to the instruments used in this investigation, because valuable information has been reported involving other measures as well. Investigators have bought an interpretation of the correlation between anxiety and intelligence. I. | G. Sarason and Palola (1960) hypothesized that the per- j ! formance of low and high-anxious subjects would vary according to type of instruction and task difficulty. They manipulated simultaneously the variables of anxiety, difficulty of task, and instructions (motivational and neutral). The tasks used in the study were a digit- symbol test and an arithmetic test. The analysis of variance showed significant triple interactions involving the three variables in every case. In general, high difficulty of task and highly motivating instructions combined to lower the performance of high-anxious subjects. The low-anxious group performed better than the high-anxious group with the difficult task and high 75 anxiety instructions. The high-anxious subjects performed better than the low-anxious subjects with the difficult task and neutral instructions and also on the easy task with both high anxiety and neutral instructions. However, the difference between the means for the easy form did not reach the level of signifi- i | cance. As can be seen from these results, the high-: j anxious subjects performed better than the low-anxious subjects under every condition except the difficult task and high-anxious instructions. The fact that they performed better with the low-anxiety instructions on difficult material would probably indicate that their performance was not dependent upon extrinsic motivation. i i On the other hand, the low-anxious subjects performed well only when external stimulation was applied. For many individuals an IQ test is an anxiety-producing stimulus. The test situation, involv ing both anxiety and difficult material, would probably result in impaired performance of the high-anxious individuals. Since test anxiety is usually more negatively correlated with IQ scores than with achieve ment or course grades, these individuals would probably show performance superior to that anticipated by their IQ scores. Consequently, they would be labeled by psychologists and educators as overachievers, when in 76 reality their true potential might be above many of their higher scoring peers. Accordingly, individuals of this type may be disqualified for gifted classes or scholarships. Allison (1964) manipulated five experimental conditions to determine the effect of anxiety on intelligence test performance for sixth grade boys and girls. The results revealed no significant difference in performance for anxiety levels, experimental condi tions, sex or any of the interactions. Allison con cluded that the high and low stress conditions gave no evidence of affecting group intelligence test perform ance of elementary school children. At the University of Colorado, Smith (1964) investigated the results of induced stress in college students just prior to taking a regular course exami nation. Irrespective of sex differences, the perform ance of high-anxious subjects under the stress conditions was significantly lower on the academic examination than that of their low-anxious classmates. Paul and Eriksen (1964) also investigated the effects of test anxiety on "real-life" examinations with college students under anxious and non-anxious conditions. The results showed significant negative correlations between the TAQ and SCAT scores as well as 77 between the TAQ and scores on the regular course exami nation. However, there was no significant correlation between the TAQ and performance on the parallel form administered under anxiety-reducing conditions. For the sample as a whole, there was no significant re lationship between the TAQ and the differential performance on the anxious and non-anxious examination. Mandler and S. B. Sarason (1952) attempted to investigate the role of drive states in a testing situation by using success, failure, and neutral reports with students in an introductory psychology course. The success or failure report elicited im proved performance for the low-anxious group but de pressed performance for the high anxious group. From the results of the investigation, optimal conditions for the high anxiety group were neutral instructions; optimal condition for the low anxiety group was the failure report. X. G. Sarason and Minard (1962) attempted to determine the effect of test anxiety on the performance of college students on an individually administered short form of the Wechsler Adult Intelligence Test. Half of the subjects received achievement-orienting instructions, and half received neutral instructions. Under the achievement-orienting instructions on the 78 comprehension section the low-test-anxious subjects performed better than the high-test-anxipus (p .02) , whereas, with neutral instructions, there was no significant difference* Sex Difference In the previously mentioned investigation (Smith, 1964) involving stress and nonstress in high- and low-anxious students taking a regular course examination in psychology, the high-anxious male subjects' performance was significantly jeopardised by the induced stress; whereas, the performance of the females was unaffected by high anxiety or stress. Even though S. B, Sarason and his co-workers (1960, pp. 136-146, 250) found a consistent and marked sex difference in the scores on the Test Anxiety Scale for Children; they found no tendency for the scores of the boys to be more closely related to the Primary Mental Ability test scores than were those of the girls. Waite, S. B. Sarason, Lighthall, and David son (1958) studied children from grades two through five and matched groups of low scorers and high scorers on the Test Anxiety Scale for Children with respect to grade f sex, and intelligence. T?hese two 79 groups were presented with a learning task. The low- anxious group performed better than the high-anxious group. The differences between the anxiety groups re- I Imained more constant with the boys than with the girls. |In general, the girls appeared to perform better than the boys. | Information that S. B. Sarason and collaborators (I960, pp. 250-251), obtained from parent interviews revealed that high-anxious boys had significantly more i !illnesses than did the low-anxious boys. He also dis covered that girls had fewer illnesses than the boys | and that there was no difference between the high-and ;low-anxious girls in this respect. i | S. B. Sarason and collaborators (1960, p. 239), I reported that girls obtain higher scores than boys on both the Test Anxiety Scale for Children (TASC) and the General Anxiety Scale for Children (GASC) , the I difference between the boys and girls being greater on |the GASC than on the TASC. | Ruebush (1963) and S. B. Sarason (1958) reported i i I that high-test-anxious girls were found to present a |more favorable impression than low-test-anxious girls, jThere was also evidence that the emotional behavior of J I girls may be influenced more than that of boys by variations in teacher behavior (Davidson & Sarason, S. B., 1961) . As a result of a study of the literature, Rue- bush (1963, pp. 484, 486, 500) reported some interest ing findings. He found evidence for the relationship between anxiety and dependency to be greater for boys than for girls. Anxiety in boys seemed to be related to behavioral indications of dependency toward teachers, social inadequacy, insecurity in play, and immature game preference. The emotional maladjustment observed in anxious boys was not true of the anxious girls. The high-test-anxious girls presented a more favorable impression than the low-test-anxious girls. The high-test-anxious girls, more than the boys, demonstrated a greater need-achievement in the class room and were more forceful and outgoing in their verbalizations. There is some reason to believe that sex differences in defensiveness may play a role in the differences between the test-anxious girls and boys. The admission of anxiety is more compatible with the feminine role than with the masculine role. Thus high-test-anxious girls should be more numerous, psychologically less deviant, more heterogeneous, and behaviorally more inconsistent, as a group, than their test-anxious male counterparts. There is some evidence that the girl who is highly defensive about admitting 81 anxiety is psychologically and behaviorally more deviant among the girls than is the high-test-anxious girl. Summary Anxiety and Performance on an Ability Test In general, measures of manifest or general anxiety do not relate significantly with scores showing academic ability (Calvin et al., 1956; Dana, 1957; Faber and Spence, 1955, p. 10; I. G. Sarason, 1960) . In general, low to moderate, and usually negative, correlations have been observed between test anxiety and ability test performance (Grooms & Endler, 1960; Ruebush, 1963; I. G. Sarason, 1959; S. B. Sarason & Mandler, 1952; S. B. Sarason et al., 1960). Anxiety and Achievement Many studies have found a negative correlation between anxiety and achievement with the correlation usually being lower than that obtained between anxiety and measures of ability. However, some investigators have reported either a positive relationship or none at all (Grooms & Endler, 1960; Ruebush, 1963, p. 498, 499; I. G. Sarason, 1957; I. G. Sarason 1961b; Suinn, 1965; Walter et al., 1964). 82 Experimental Studies with Psycho logical Measures of Anxiety A number of investigators have attempted to find an explanation for the correlation between anxiety and intelligence by using various experimental conditions. The performance of high-anxious individuals under stress has been found to be inferior to that of their low-anxious counterparts (Smith, 1964-1965; I. G. Sarason & Minard, 1962) . Allison's (1964) investigation resulted in no significant difference in performance between levels of anxiety, whereas I. G. Sarason and Palola (1960) found the high-anxious individuals to have superior perform ance in every case except with high-anxious instructions and a difficult task. These results support the theory that anxiety is a complex construct. Sex Differences Most investigators have shown a greater degree of admission of test anxiety by girls than by boys. As a result of a study of high- and low-test-anxious subjects, Ruebush (1963) and S. B. Sarason (1958) re ported that high-test-anxious girls present a more favorable impression than do low-test-anxious girls. From a study of the literature, Ruebush (1963, 83 pp. 484, 486, 500) drew some tentative conclusions. He stated that anxiety in boys was related to be havioral indications of dependency toward teachers, social inadequacy, insecurity in play, and immature game preference. In general, high-test-anxious girls tended to present a more favorable impression than the low- test-anxious girls. There was some evidence that the girl who is highly defensive about admitting anxiety is psychologically and behaviorally more deviant among girls than is the test-anxious girl. CHAPTER IV METHODOLOGY The purpose of this chapter is to describe the sample, the procedures used, including the instruments, along with the design and statistical procedures employed. The Sample The sample for this investigation consisted of one hundred seniors drawn from government classes of West High school in the Torrance Unified School District near Los Angeles. This school closely parallels national parameters on intelligence and achievement tests. This contributes to the generalizability of the findings. The mean composite score of the first eight tests of the Iowa Test of Educational Development taken in grade eleven was slightly above the national median. The California Test of Mental Maturity (CTMM) scores were not available for these experimental subjects, but the mean IQ for students in the eleventh grade, in the same school during the year of the study, was 103. 84 85 Approximately 60 per cent of the class aspired to go to college. However, only about 35 per cent actually had transcripts sent to colleges. High school students were used as subjects in order to avoid the limitations on external validity imposed by using highly select subjects such as volunteer college students. Seniors were chosen for the study because they were in the process of making certain important decisions often revolving around college attendance versus work. Since many of these decisions are largely determined by results of their test performance, it was presumed that an important motivational ingredient would be present in their test-taking behavior. Place and Conditions for Testing The conference room of the assistant principal served as the experimental setting. Each examinee was seated in a comfortable wooden chair equipped with a desk top providing adequate space for testing and attachments for physiological measurements (see Figure 2). The thermostat was placed at 65° F. and opportunities for distraction were minimized. Instruments 86 Measurement of Proficiency The high and low proficiency categories for purposes of this investigation were determined by the students' scores on the Iowa Test of Educational Development, taken during grade eleven. These scores were obtained from the students' files. Subjects were classified into high and low proficiency groups on the basis of composite scores on the Iowa Test of Educational Development. Students whose composite score was below the class mean (56-57 on the composite of tests 1-8) were classified as low proficiency, and those above the mean were classified as high pro ficiency. Measurement of Task Performance The task performance was the Cooperative Academic Ability Test Form "A," published by the Educa tional Testing Service, Princeton, New Jersey. This test, designed for use with college-bound students, yields verbal, mathematical, and total scores. In order for the testing time to conform to the length of the class period, the test was modified by an elimi nation of every third question, thus reducing the time required by one third. This shortening of the 87 examination reduced the reliability, as computed by the Spearman-Brown formula, from .88 for the total scores to .84; from .92 for the verbal scores to .87; and from .94 for the mathematical scores to .90. The remaining questions were placed in order, at random, for their respective sections, verbal and mathematical, by use of a table of random numbers (Kendall & Smith, 1938). This arrangement provided opportunity for individuals unable to complete the test in the time provided to attempt a proportion of the more difficult questions usually placed at the end of a test. A list of the questions in the order pre sented is provided in Appendix A. A random half of the subjects was given the test with the verbal section preceding the mathematical section. The other half was given the test with the mathematics section preceding the verbal section. The former arrangement was designated as form A, and the latter form B. This procedure was adopted in order to prevent the results from being contaminated by fatigue, which might lower the scores on the section taken second. Psychological Measures of Anxiety The four psychological measures of anxiety used in the investigation were: (a) S-R Inventory of 83 Anxiousness (S-RI), (b) Affect Adjective Check List (AACL) , (c) I* G. Sarason's Test Anxiety Scale (TAS) , and (d) Bendig's Short form of Taylor's Manifest Anxiety Scale (MAS). These four instruments (see Appendix A) were stapled together under the title Student Survey. The four tests were labeled only Parts I, II, III and IV. Physiological Measures of Anxiety Measures of the following physiological variables were obtained. Respiration. Respiration measured in respiratory cycles per minute was recorded through a thorax pneumograph leading into the Statham physio logical transducer Model P23B. The output of the Statham was led into a DC preamplifier (Model 5P1) and recorded on one channel of the Grass Model 5 Poly graph. Respiration rate and depth were both figured even though the equipment was not designed to yield an accurate quantitative measure of the latter. Consequently, the measurements of respiration depth can serve only to reflect a general trend. Heart rate. Heart rate was determined by an electrocardiogram (EKG) preamplifier (Model 5P4) of a 89 Grass polygraph Model 5, equipped with standard limb leads. An electrode coated with a thin layer of sodium chloride electrode paste was placed on the palm surface ■of the wrist of the nonpreferred hand and on the left leg. Leads XI and III were used for the left and right- handed individuals respectively. Lead 1 was not used because its use required electrodes to be placed on both arms. For the test taking task in this investi gation freedom of the preferred hand was desired. Galvanic Skin Response (GSR). The GSR was recorded with a low-level DC preamplifier (Model 5Pl) of the Grass Model 5 Polygraph. Sodium chloride elec trode paste was applied to the skin-electrode junction. Two silver electrodes, 1/2 inch by 5/8 inch, bent to the contour of the finger, were coated with the ■electrode paste and securely taped to the palmar surface of the first and thi rd fingers of the non preferred hand. Basal GSR resistance levels were recorded in micromhos. Blood Pressure. Data on the systolic and diastolic blood pressure and pulse pressure were gathered in this study. Blood pressure was measured with a standard cuff and sphygmomanometer on the non preferred arm. 90 Oral and Skin Temperatures. For indicating temperature, a Yellow Springs thermister electronic telethermometer with three probes was used. The three probes were for oral, face, and finger temperatures. Of those providing skin temperatures, one was placed on the palmar surface-of the middle finger of the nonpreferred hand, and the other on the cheek. These probes were taped on the indicated surfaces after the areas had been carefully cleansed with alcohol. procedures Several weeks before physiological measures were taken, the investigator administered the four psychological tests of anxiety to all prospective subjects. Selection of the Sample There were 280 students in the senior govern- ment classes available for the experiment. Approxi mately 20 were eliminated from this group because either their scores on the Iowa Tests of Educational Development or complete scores on the psychological tests were not available. From the -remaining 260 prospective -subjects, a group of 130 was chosen at random, and balanced according to sex and proficiency 91 level as determined by the ITED. In compliance with the school's requirements, letters requesting parental consent were sent to the homes of the 130 prospective subjects. (See Appendix A.) in the event that these letters were not returned, the investigator contacted the parents by telephone in order to preserve the character of the sample. Approximately one month later a second lot of approximately 20 letters was sent out in order to supply enough subjects to replace those for whom parental consent was not secured. Parental consent was finally secured for 131 subjects. Records were taken for all but one of these. The data for 30 of the subjects were then discarded either be cause of technical difficulties undergone during the testing or in the case that their performance was measured only in order to perfect the instrumentation procedures. The data from the remaining 100 subjects were used in the analysis. To the extent that this procedure influenced the character of the sample, bias was introduced. Assignment to Treatment Groups The 100 subjects, counterbalanced for sex and 1 proficiency level were in turn subdivided into five groups, also counterbalanced for sex and proficiency levels.2 Since research has shown physiological measures to vary with the time of the day, digestion of food, and seasons of the year (Best & Taylor, 1961, pp. 274- 276, 302, 884, 1270, and Wenger & Ellington, 1943), special precaution was taken to have individuals assign ed at random to treatments, day of testing, and time of day. As computed by analysis of variance, there was no significant relationship between the treatment groups with respect to day, time of test, and date the letter of approval was received. (See Tables 23-25.) In order to facilitate the selection of subjects, the names of the students were listed alphabetically according to class period, with a plus or minus symbol indicating proficiency level and M or F indicating sex. A schedule was then compiled designating the sex and proficiency level of the subjects required for each period of the day and the type of treatment and form of iThe group of 100 subjects was composed of 25 low-proficiency boys, 25 high-proficiency boys, 25 low- proficiency girls, and 25 high-proficiency girls. 2Each of the five groups was composed of 5 low- proficiency boys, 5 high-proficiency boys, 5 low- proficiency girls, and 5 high-proficiency girls. 93 examination (form A or form B) to be administered., An available student possessing the required characteristics as to proficiency level and sex was then administered the treatment planned for that period. Each day the schedule was shifted to allow a different group to be taken at the various times of the day. Treatments stimuli in the form of test instruction, and two served as controls. Figure 1 illustrates the design. by one of the classroom teachers and recorded on tape prior to the major testing. This recording was then played as the anxiety stimuli for the treatment situa tion, thereby assuring a standard presentation of the instructions to the subjects. Physiological measures were taken for each subject while he performed on the Academic Ability Test. (ARI). The following instructions for this treatment group were intended to reduce feelings of anxiety or concern because of the test. We are conducting a research project con cerning examinations. The test you are about Of the five treatment groups, three received The various treat it instructions were read Treatment One— Anxiety-reducing Instructions 94 °1 XARI °2 ^1 XNI °1 XAPI °2 °1 XCPM °2 XCTP °21 °T XARI XNI XAPI ^PM XCTP O rp °1 °2 Anxiety-reducing instructions Neutral instructions Anxiety-producing instructions Control for physiological measures Control for test performance Test performance Physiological variable data ■^Physiological measures taken after com pletion of the test. Pig. 1. Treatment groups design (Campbell-Stanley notation) 95 to take is very long and difficult so do not be concerned if you are not sure of all the answers or cannot finish within the given time, . No one will find it easy. We wish to make it clear that this is purely for research pur poses and will have nothing to do with any course offered at West High. Attempt the questions in the order they come— do not skip around. We wish to thank you for your cooperation in this research. Please follow the General Directions on the front of the test booklet and on page three. Treatment Two— Neutral Instructions (NI). The neutral instructions, phrased neither to reduce nor to produce anxiety, were the standard instructions that came with the test. These instructions were as follows The test you are about to take is an Academic Ability Test. Attempt the questions in the order they come— do not skip around. Please follow the General Directions on the front of the test booklet and on page three. Treatment Three— Anxiety-producing Instruc tions (API). The anxiety-producing instructions were designed to elicit a high degree of anxiety by placing emphasis on the importance of the examination. The anxiety-producing instructions were as follows: The test you are about to take is a college ability intelligence test. This test has been found to predict such things as course grades, success or failure in college success or failure in various types of occupations, and success or failure in later life. Of course, your own intelligence will primarily determine whether you do well or poorly on this test. Therefore, it is an extremely important examination and you should try to do your very best. 96 At a later date the scores will be posted so your score can be compared with others who have taken the test. Your teachers and counselors will be given your results so be sure to try your very best on all questions. Attempt the questions in the order they come— do not skip around. Remember it is important for you to concentrate at all times. Please follow the General Directions on the front of the test booklet and on page three. Treatment Four— Control for Physiological Measure (CPM). Students in the control group for physiological measures had their physiological responses recorded by the same methods as the first three groups, but participants read a given section in their government books in lieu of taking the test. This group served as a control for physiological meas ures in that it provided a record of the physiological responses independent of test performance in order to detect changes due to peculiarities in the experimental environment, instrumentation, fatigue, et cetera. Treatment Five— Control for Test Performance (CTP). Students in the control group for test per formance had the same neutral instructions as Treat ment Two, and took the Academic Ability Test, but with no concomitant physiological measures. After the completion of the test, pulse rate, systolic blood pressure, diastolic blood pressure, and oral 97 temperature were taken. Group five served as the con trol for test performance in that it provided a measure of the examinee's performance on the ability test, in dependent of the presence of the instrumentation for physiological measures. The equipment for securing physiological meas ures was allowed to warm up approximately one-half hour before use. All bodily surfaces that were to have electrodes attached were thoroughly cleansed with 70 per cent alcohol. Before use with each subject, the electrodes were cleansed with alcohol and, when neces sary, with steel wool in order to provide good contact. The oral temperature probe was allowed to stand in a solution of aqueous zephiran chloride 1:750 and in a solution of 70 per cent alcohol after use with each subject. Process of Testing After the student was directed to the appointed chair for testing in the treatment room, the electrodes and attachments for the physiological measures were placed in the appropriate locations on the examinee as pictured in Figure 2. Shown in the foreground of Figure 3 is the lOne part zephiran chloride to 750 parts dis tilled water. VO 00 Fig. 2. A Student Performing Fig. 3. The Grass Polygraph 100 Grass polygraph that recorded the respiration, EKG, and GSR. To the left of this in the same picture, is the thermister electronic telethermometer. While the electrodes were being attached the subject completed the background information required on the top of the answer sheet, read the directions for the test, and asked questions if he wished to do so. Before the initiation of the experimental period, subjects were allowed to become accustomed to the recording apparatus, and pretest readings were recorded for all physiological measures. The anxiety stimuli in the form of instructions were given by a tape recorder. This method was used in order to standardize the presentation. Following the introduction of the anxiety stimuli, the student was administered the Affect Adjective Check List with the directions to check it according to his present feelings. This questionnaire had also been presented to each sub ject two weeks earlier as a part of the student survey. The scale was given on the two occasions in order to determine the relationship between the introspective evaluation of his usual feelings during an examination and his evaluation just prior to the testing event. j I Following completion of the AACL, the examinee immediately j began work on the Academic Ability Test. Concomitant 101 physiological measures were recorded throughout the testing. The completion of each question was monitored by an event marker on the recording paper of the poly graph. This monitoring indicated the time required for answering the question. A portion of this graph paper, presented in Figure 4, indicates respiration, the monitoring of the question, the EKG, and the GSR. Paper speed was 2.5 mm per second. While the student performed on the Academic Ability Test, continual measurements were recorded from prereading to postreading for respiration rate, respiration depth, EKG, and GSR. The subject's blood pressure was taken on three occasions: once immediately before beginning the test, once immediately after the completion of the first half of the test, and a third time immediately after the completion of the second half of the test. Cheek, finger, and oral temperatures were taken on four occasions: once immediately after the completion of the AACL, and again on each of the three above occasions at approximately the same times as the blood pressure readings. Analysis of Data As illustrated in Figure 5, the principal { I I F T I M TITrrrrr ! 1 [11 i i n i l t t j s i i p ! : * : m m I r i * ■ ■ 1 ; i ! = : i T ~ T _ r , ; ; , j ; , i i ; l i iii L j } - i : i ; i: i M p ! ' ! ■ N i l : ' j . :i . . i : f ! T f ! i ; ' t wmn' w - " ' i i ! : ' iitt J . i - . h i i l i ' : ! ■ i m m N - f t i ; 1 ; r i i ' / r i i ; : ! J' : ! - 1 '■ ■ O i ■ ■ ■ ‘ ■ - H 0 1 t e i . i ' - h t ! : ! 1 ! 103 Treatments 1. Anxiety- reducing 2. Neutral 3.Anxiety- producing 4.Control— for either physio logical measures or ability test 2.Girls 1. Boys Sex 2. High 1. Low Proficiency Level Pig. 5. The statistical model within which the data was analyzed. 104 I i statistical model employed for.the experiment was a balanced 4 x 2 x 2 (type of instructions x level of proficiency X sex) fixed effects model of analysis of variance and covariance. A 5 per cent level of signifi cance was employed throughout. Analyses of variance on the Academic Ability Test were run for the total score of the test, on the verbal and mathematical sub-tests, and on the ten physiological measures, namely: a. respiration rate b. respiration depth c. heartbeat rate a. galvanic skin response e. systolic blood pressure f. diastolic blood pressure g* pulse pressi^re h. oral temperature i. face temperature j • finger temperature An analysis of covariance was computed on the Academic Ability Test for the total, verbal, and mathe matical scores with the initial anxiety level for each of the physiological measures separately held constant. Correlations were computed between the following measures for each treatment group: A. Physiological measures were correlated with: j 1. Other physiological measures of anxiety 2. Psychological measures of anxiety 3. Total, verbal, and mathematical scores of the Academic Ability Test. j B. Psychological measures were correlated with: Total, verbal, and mathematical scores of the Academic Ability Test. | C. The Affect Adjective Check List given prior j to the major testing was correlated: j | With the same scale given after the j ] anxiety instructions for the ability test. D. Psychological measures were correlated with: Psychological measures for the 100 examinees. i ! Summary Problem The principal problems of the study concerned the following issues: (a) the extent to which anxiety influences ability test performance; and (b) the extent to which various physiological and psychological meas ures of anxiety are related. 106 i ; Procedures All high school seniors enrolled in any of the nine sections of the United States Government classes at West High School in the Torrance Unified School District were administered the following four self- report anxiety measures! S-R Inventory of Anxious ness, Affect Adjective Check List, Test.Anxiety Scale, and the Manifest Anxiety Scale. One hundred of the students, counterbalanced for sex and high and low proficiency levels, were in turn subdivided, at random, into five groups, also counterbalanced for sex and proficiency levels. The five groups were composed of three treat ment groups and two controls. The treatment groups received anxiety-reducing, neutral, or anxiety- ; producing instructions presented by a tape recorder before being administered the Academic Ability Test. While each student took the test, respiration rate and ; depth, heartbeat rate, GSR, systolic and diastolic | blood pressure, pulse pressure, and oral, face, and finger temperatures were taken. One control group had physiological measures recorded but read in their government books in lieu of taking the ability test. The other control group took the ability test with neutral instructions but no 107 physiological measures were taken until the completion of the test, and then only pulse rate, oral temperature, ; and systolic and diastolic blood pressures were taken. Analysis of the Data The basic statistical model was a three-way | analysis of variance (treatments x proficiency levels x i | sex). The dependent variables included the verbal, f ] i mathematical, and total ability test scores, and the physiological and psychological measures of anxiety. Analysis of covariance with the same factorial design ! was also carried out with the ability test scores, | using the initial level of anxiety (pretest readings) ! j held constant. Correlational analyses were performed on all anxiety measures and the ability test scores. CHAPTER V RESULTS This chapter presents the results of the investigation in terms of the hypotheses and provides a discussion of the findings. Hypothesis One Anxiety will influence test performance in the following ways s a. Anxiety-reducing test instructions will result in highest levels of performance on an ability test. b. Neutral test instructions will result in medium levels of performance on an ability test. c. Anxiety-producing instructions will result ih lowest levels of performance on an ability test. Results and Discussion As observed in Table 1, the results of the analyses of variance indicated no significant difference in petformance betwfeen the treatment groups. The P value for the total scores was .66, the verbal .67, and the mathematical scores 1.20. 108 TABLE 1 Results of Analysis of Variance for the Total, Verbal, and Mathematical Scores for the Academic Ability Test Source of Variation Total Scores df Mean Square Verbal Scores F Mean Square Mathematical Scores F Mean Square F A-Instruction 3 27.80 .66 9.00 .67 18.52 1.20 B-Proficiency 1 3250.59 77.56** 825.76 61.08** 796.69 51.44** C-Sex 1 .26 .01 90.75 6.71*a 138.39 8.94**b A X B 3 29.10 .69 25.44 1.88 3.74 .24 A X C 3 83.65 2.00 11.84 .88 41.70 2.69 B X C 1 1.10 .03 5.80 .43 1.71 .11 A X B X C 3 79.50 1.90 9.41 .70 37.01 2.39 Within 64 41.91 13.52 15.49 Total 79 *Significant at .05 level aGirls > boys ^Boys >- girls **Significant at .01 level 110 The means which were examined by the analyses of variance are presented in Table 26, and graphic j presentations of the three types of scores (total, -verbal, and mathematical) for both sexes are presented in: Figures 6-9 in Appendix B. Lack of a significant difference in performance i among the groups with different types of anxiety instructions indicated either that anxiety had no signi ficant effect on performance or that the various instructions did not influence anxiety. The latter . explanation is likely, in view of the analysis of.the physiological variables, which revealed no significant differences among the three instruction groups other i than respiration rate. Hypothesis Two The degree of anxiety elicited by the instructions will influence test performance in the I following ways when the initial level of anxiety is held ! j constant: ! a. Anxiety-reducing test instructions will result in the highest level of performance on an ability test. b. Neutral test instructions will result in medium levels of performance on an ability test. 1X1 c. Anxiety-producing instructions will result in lowest levels of performance on an ability test. Results and Discussion Before the analysis of covariance was computed, several of the physiological variables were plotted, by the computer, with test scores to confirm linearity,, Since the recordings for the galvanic skin response (GSR) and the respiration depth were not absolute measures, change scores (the difference between the prereading and the mean test reading) were used. In view of the fact that the members of the groups were selected at random, it was assumed that there would be no significant differences between the treatment groups for the pretest readings. This assumption was confirmed by analyses of variance of the pretest readings. (See Tables 27-32 in Appendix B.) No significant differences were evident between the treatment groups for the pretest readings for any of the physiological measures tested. The highest F value was 1.78 for diastolic blood pressure and the lowest was .04 for systolic blood pressure. Both of these F values fall short of 2.72 required for the .05 level of significance. Results and Discussion A summary of the results, from Tables 1 and 112 j 33-74 in Appendix B, of the analysis of variance for the total, verbal/ and mathematical scores without the i physiological measures held constant and the analysis of covariance with each of the physiological measures held constant is presented in Table 2. In nearly every case the F values for the ability test with the physiological measures held constant were larger than when the physiological measures were not held constant, but in no case did the F value attain the .05 level of significance. Therefore, hypothesis two was rejected. No significant differences in performance evolved among the treatment groups with the pretest readings for the physiological measures held constant. The lack of a significant difference between the treatment groups ; was also evident when the control group was included. Since the control group for test performance had no physiological measures taken during the test, the measures taken at the completion of the test were used as the covariate (Tables 63-74). The means for the pretest readings, which were analyzed by the analyses ; of variance, are presented in Tables 75-84 in Appendix B. As can be observed, the discrepancy between means is only slight from one group to another. Analysis of variance for the ability test without the pretest readings for the physiological 113 TABLE 2 Summary of Results of Analysis of Variance for the Academic Ability Test and Summary of Results of Analysis of Covariance of the Academic Ability Test with the Pretest Anxiety Level Held Constant F for F for ' F for Measures total score verbal score math, score Analysis of Variance for the Academic Ability Test Academic Ability Test .66 .67 1.20 — Analysis of Covariance for the Academic Ability Test with the Following Physiological Measures ________________________ Held Constant_____________________ Physiological Measures Held Constant Respiration Rate 1.21 1.00 1.29 Respiration Deptha 1.16 1.11 1.33 Heartbeat Rate 1.16 1.14 1.66 GSRa .92 1.02 1.16 Systolic B. P. .97 1.00 1.15 Diastolic B. P. .53 .24 1.16 Pulse Pressure .71 .61 1.14 Oral Temperature .98 .76 1.31 Face Temperature 1.17 1.10 1.36 Finger Temperature .84 .76 1.22 aDifferences between the pretest reading and test reading used as the covariate. 114 measures held constant revealed no significant instruc tion x sex interaction for the total, verbal, or mathematical scores and no significant sex difference for the total scores (see Table 1), However, there were significant sex differences for the verbal and mathemati cal scores with the girls superior on the verbal part and the boys superior on the mathematical part. Analyses of covariance for the ability test, total and part scores, with the initial readings for the physiological measures held constant, were computed to test for differences in performance,, Table 3 presents j a summary of results of the analyses of covariance provided in Tables 33-74, With the analysis of covari ance the girls' verbal scores were still significantly superior to those of the boys, but only with finger temperature held constant were the boys' scores still significantly superior to the girls' scores on the mathematical section. When the control group was included the boys were again significantly superior to the girls on the mathematical section (see Tables 63- 74 in Appendix B). In order to study these findings, the means for the ability test were adjusted for the total and part scores with the initial anxiety level for each of five physiological measures held constant. As presented in 115 TABLE 3 Summary of Results of.Analysis of Covariance for Sex and Instruction X Sex Interaction for Total, Verbal, and Mathematical Scores with the Pretest Level for Each of the Physiological Measures Held Constant Instruction X Sex Sex _________Interaction Signifi- Signifi- Measures F cance F cance Resp. Rate Total 0.05 N S 4.14 <4.05 Verbal 5.14 4.05 G>B 2.31 N S Math. 3.47 N S 3.44 4.05 Resp. Deptha Total 0.20 N S 3.99 <4.05 Verbal 5.66 4.05 G>B 2.34 N S Math. 3.71 N S 3.58 4.05 Heartbeat Rate Total 0.30 N S 4.14 4.05 Verbal 5.44 4.05 G>B 2.41 N S Math. 3.15 N S 3.81 <.05 GSRa Total 0.32 N S 4.08 4.05 Verbal 5.54 4.05 G>B 2.47 N S Math. 2.84 N S 3.70 <.05 Systolic B.P. Total 0.62 N S 4.31 <.05 Verbal 6.76 4.05 G>B 2,64 N S Math. 2.15 N S 3.87 < .05 Diastolic B.P. Total 0.47 N S 4.25 < .05 Verbal 7.65 <.01 G>B 2.41 N S Math. 3.18 N S 3.76 4.05 Pulse Pressure Total 1.38 N S 5.87 <.01 Verbal 10.22 <.01 G>B 4.24 4.05 Math. 1.73 N S 4.62 <.05 Oral Temp. Total 0.43 N S 3 .49 < .05 Verbal 7.45 4.01 G>B 1.68 N S Math. 3 .10 N S 3.44 < .05 TABLE 3— (Continued) 116 Sex Instruction X Sex Interaction Measures F Signifi cance F Signifi cance Face Temp. Total 0.27 N S 3 .90 <.05 Verbal 4.67 <.05 G>B 2.47 N S Math. 2.80 N S 3.29 <.05 Finger Temp. Total 0.00 N S 3 .56 < .05 Verbal 4.17 <.05 G>B 2.18 N S Math. 4.91 <.05 B>G 3.14 N S differences between the pretest reading and the test reading used as the covariate. 117 Tables 85-89 and Figures 10-24 in Appendix B, the adjusted means were essentially identical with the I unadjusted means due to the high comparability of the ! | mean readings and pretest readings, and the low i | correlation between the covariates and the dependent variable. The explanation for the significant inter- j actions with the ability test scores when the pretest | readings were used as the covariates, and the lack of | significant interactions when the pretest readings were | not held constant were due to the greater power of the j analysis of covariance over that of the analysis of ! variance to detect a difference when a real difference exists. i Hypothesis Three i j The degree of anxiety elicited by the instruc- ! tions will influence physiological responses during test performance in the following ways: a. Anxiety-reducing test instructions will result in the lowest physiological responses. | b. Neutral test instructions will result in | medium responses. c. Anxiety-producing instructions will result in the highest physiological responses. 118 Results and Discussion The only physiological measures that provided a significant difference between the treatment instructions was respiration rate, significant at the .01 level (see Tables 90-99). However, the order of responding was the reverse to that hypothesized. | The respiration rate was the highest for the | anxiety-reducing instructions. Under the influence ! ; of neutral instructions, a medium respiratory rate was obtained as hypothesized. The lowest levels of responding were for the groups with anxiety-producing instructions and for the control group. The investigator has not found a satisfactory • j explanation for the relatively rapid respiration rate under the influence of anxiety-reducing instructions and the relatively slow rate following the anxiety- producing instructions. The equipment used for measur ing respiration was sensitive to bodily movement as well as respiration. Hypothesis Four There will be no difference in physiological responding between high and low proficiency levels. ! Results and Discussion Hypothesis four was confirmed. Physiological 119 responses of the students with high and low proficiency levels were not significantly different, (See Table 4 and Tables 90-99 in Appendix B,) The F values for physiological .responses and proficiency levels were very low. All the values were less than 1.0/ except systolic blood pressure, diastolic blood pressure, and oral temperature. Tjhe findings under the conditions of this study indicate no ! significant relationship between physiological responses and proficiency levels. | Hypothesis Five j Boys will have higher physiological measures | | than girls for face temperature, finger temperature, i j systolic blood pressure, diastolic blood pressure, and j j pulse pressure. • Girls will have higher physiological i | measures than boys for oral temperature, respiration i rate, respiration depth, heartbeat rate, and GSR. Results and Discussion The results of the analyses of variance for physiological measures are presented in Table 5 and in Tables 90-99 in Appendix B. The means are provided in 1 j Tables 100-111 and the comparison between the sexes in I j Figures 25-34. The hypothesis that the boys would have higher face and finger temperatures than the girls was TABLE 4 Summary of Results of Analysis of Variance for Proficiency Level with the Mean Physiological Measures for the Three Instruction Groups and the Control Groups Measures F Respiration Rate .02 Respiration Depth3 .05 Heartbeat Rate .36 GSRa .13 Systolic Blood Pressure 1.63 Diastolic Blood Pressure 1.12 Pulse Pressure .02 Oral Temperature 1.53 Face Temperature .04 Finger Temperature .11 aMean minus the pretest reading 121 TABLE 5 Summary of Results for Analysis of Variance For Sex Differences for Physiological Measures Measures df Mean Square F Sex with the higher means if Significant . Face temp.,a degrees F 1 13 .41 6.12* Boys Finger temp.,a degrees F 1 246.41 4.96* Boys Systolic B.P.,*3 mm Hg 1 1109.97 12.43** Boys Pulse pressure,*3 mm Hg 1 1314.80 12.10** Boys Resp. Rate, per min. 1 40.01 4.81* Girls Oral Temp., degrees F 1 2.70 18.22** Girls GSR,a'c microhms 1 438.36 1.43 ne ither Resp. Depth,c resp. cycles per min. 1 5.26 0.09 neither Heartbeat rate per min. 1 98.90 0.02 neither Diastolic B.P. mm Hg 1 1.51 0.03 neither Total df. 79 TABLE 5— (Continued) 122 *Significant at .05 level **Significant at .01 level aHigher scores probably indicate less anxiety (Stevens, 1951), p. 477. ■ V - Boys' systolic blood pressure normally is 4 or 5 mm above that of girls. This may influence pulse pressure also. cMean minus the pretest reading. | 123 j jconfirmed. The differences between boys and girls for j | both face and finger temperatures were significant be- I yond the .05 level. The means for the face and finger ! temperatures are presented in Tables 100 and 101 in iAppendix B. The sex differences in response as indicated i i by face and finger temperatures are graphically demon- ! strated by Figures 33 and 34. Physiological measures are difficult to inter pret. As previously shown, there is a suggestion of I lower skin temperature accompanying increased anxiety. The girls' lower face and finger temperatures may indi cate greater anxiety. The hypothesis that the boys would have higher | measures for systolic blood pressure and pulse pressure i ;was confirmed. These differences were significant | !beyond the .01 level. (See Table 5.) The means for the I systolic blood pressure and pulse pressure are presented | in Tables 102 and 103. It is obvious from Figures 29 j and 31 that the boys obtained higher systolic blood pressure and pulse pressure than the girls. Normally |the systolic blood pressure is five mm higher for the | I boys than for the girls. This difference would |probably also influence the pulse pressure. The normal sex difference may be sufficient to account for the significant sex difference in systolic blood pressure and 124 pulse pressure. The hypothesis that diastolic blood pressure f would be higher for boys than for girls was rejected. * According to Table 5 no significant sex difference for diastolic blood pressure was evident, Prom Table 104 and Figure 30, one can see the sex difference is I only slight for diastolic blood pressure. i j The girls ' oral temperature was significantly i higher than that of the boys beyond the .01 level. (See Table 5.) As shown in Table 105, the mean oral | temperature over all the three instruction groups was 99.81 degrees Fahrenheit for the boys and 100.11 for i i the girls. When the temperatures for the control group were also included, the means were 99.75 for the boys and | 100.12 for the girls. Figure 32 portrays the sex I differences for oral temperature. All the groups except ! j the anxiety-producing group clearly reveal a sex differ- ,ence for oral temperature. [ The hypothesis that the girls1 measures for I | respiration rate would be higher than those for the | boys was confirmed. These differences were significant i 'beyond the .01 level. (See Table 5 and the graphic 'presentation in Figure 25 in Appendix B)„ As presented ;in Table 106, the mean respiration rate for the three |instruction groups was 17.73 respiratory cycles per 125 minute for the boys and 19.74 for the girls. When the scores for the control groups were included, the ! means were 17.55 for the boys and 19.05 for the girls. The hypothesis that the girls would have higher measures than the boys for respiration depth, heartbeat rate, and GSR was rejected. No significant sex differences in the responses were obtained. (See Table 5 and Tables 107-111 and Figures 26, 27, 28 in Appendix B) . Table 5 provides a summary of the results of the analyses of variance for sex differences and the direction of the differences for physiological measures. Although interpretations should be made with caution because of the possible effect of body movement on some of the responses, the findings for this study indicate no significant differences between the sexes for respiration depth, heartbeat rate, GSR, and diastolic blood pressure recorded during the test performance. The equipment used for measuring the respiration gave a more valid measure of rate than of depth. In addition, respiration rate, respiration depth, and GSR are somewhat more vulnerable to imprecision than the other measures, due to sensitivity to slight movements |unavoidably made by the subject during the time span of the examination. Very little change in response was 126 expected for diastolic blood pressure. An interesting observation is provided in Table 6, which presents a comparison of the initial anxiety level and the mean test anxiety level for males and ! females separately. Those with high pretest readings i | tend to have high test readings. 1 | To summarize, according to Table 5, it appears i |that the boys' responses were higher for more measures |than were those of the girls. This difference in response does not necessarily indicate greater anxiety j |experienced by the boys than by the girls. According I to Stevens (1951, p. 477), emotional stress produces i I a fall in skin temperature due to vasoconstriction. In view of Stevens' report, the lower face and finger temper i j |atures observed for the girls would indicate that they experienced more emotional stress. Best and Taylor (1961, pp. 274-275) report that I :the systolic blood pressure for boys around age 17 is ! !approximately 120 mm Hg. For girls of the same age it i jis approximately 4 or 5 mm lower. The sex difference |in systolic blood pressure cited by Best and Taylor, i |rather than anxiety, may account for the significantly i higher systolic blood pressure experienced by the boys i than by the girls. Since the pulse pressure is the dif- Iference between the systolic and the diastolic blood TABLE 6 Comparison of Means for Pretest Anxiety and Test Anxiety as Measured by Physiological Measures for the Three Instruction Groups and the Control Group Mean Pretest Readings Mean Test Readinqs Boys Girls Higher Boys Girls Higher Resp. Rate 18.8 20.0 girls 17.6 19.0 girls* Resp. Depth 25.6 23.2 boys 25.4 22.3 boys Heartbeat Rate 73.7 84.5 girls 71.7 83.7 girls GSR 86.2 80.4 boys 79.6 78.9 boys Syst. B.P. 120.8 115.2 boys* 117.5 110.5 boys** Dias. B.P. 67.1 67.3 girls 67.2 . 66.8 boys Pulse Pressure 54.4 47.8 boys* 51.9 44.0 boys** Oral Temp. 99.0 99.4 girls** 99.6 100.0 girls* Face Temp. 94.8 93.4 boys** 96.0 95.2 boys* Finger Temp. 85.0 82.0 boys* 87.4 83.9 boys* *Significant at .05 level **Significant at .01 level n = 40 127 128 pressure, the normal sex difference in the systolic blood pressure may also influence the pulse pressure. Thus the lower £ulse pressure evidenced for the girls may or may not indicate greater anxiety. However, the girls' significantly higher respiration rate probably does indicate greater anxiety. During the examination, the girls experienced a higher oral temperature than the boys. Higher oral temperature has been found to be associated with examina- ! tions. It seems likely that since the subjects were tested at random in regard to days and time of the day, | the higher oral temperature experienced by the girls i | over that of the boys indicated greater anxiety. Inter- | ! preted in view of the previous literature cited, an analysis of the patterning of autonomic variables may give evidence of a real difference in physiological sensitivity of the girls over that of the boys to test anxiety stimuli as evidenced by respiration rate, oral temperature, face temperature and finger temperature. Hypothesis Six There will be significant relationships within | each treatment group among the following physiological measures of anxiety: respiration rate, respiration depth, heartbeat rate, GSR, systolic blood pressure. 129 diastolic blood pressure, pulse pressure, oral temperature, face temperature, and finger temperature» Hypothesis six, that a significant relationship exists between the physiological measures of anxiety, was true for ten relationships, only one of which held up under all three degrees of anxiety elicited by the instructions. (See Tables 7-9.) Systolic blood pressure related to pulse pressure positively and significantly at the .01 level for all three instruction groups. The correlations between the two measures were .64, .88 and .90 for the groups which received anxiety-reducing, neutral, and anxiety-producing instructions, respectively. In view of the fact that pulse pressure is computed as the difference between the systolic and the diastolic blood pressure, the high relationship between the systolic blood pressure and pulse pressure is not surprising, and in a sense it may be misleading if they are thought of as being distinct measures. The second relationship found to be significant was systolic blood pressure with diastolic blood pressure for the group receiving anxiety-reducing instructions (r = .64, p < .01) . Four of the significant correlations for the |neutral instruction group involved pulse pressure. Pulse pressure related r = -.48 (p .05) with respiration TABLE 7 Correlational Matrix of Physiological Measures of Anxiety for the Group with Anxiety-Reducing Instructions 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 1. Resp. Rate- per min. 1.00 0.08 0.36 -0.07 -0.12 0.15 -0.35 0.10 0.24 -0.27 2. Resp. Deptha Respiratory cycle 1.00 -0.13 0,32 -0.07 0.18 -0.23 0.00 0.12 0.18 3. EKG - per min. 1,00 -0.09 -0.12 00 o o -0.21 0.26 -0.30 0.24 4. GSRa- micromhos .1.00 -0.31 -0.12 -0.29 0.31 0.30 0.25 5. Systolic B.P.- mm Hg 1.00 ** 0,64 •** 0.64 -0.27 0.12 -0.35 6. Diastolic B.P.- mm Hg 1.00 -0.16 i o 0 M O 0.13 0.01 7. Pulse Pressure - mm Hg 1.00 -0.27 0.06 -0.36 8. Oral Temp, degrees F, 1.00 -0.30 -0.08 9. Face Temp, degrees F. 1.00 0.14 o a Finger Temp, degrees F. 1.00 **Significant at .01 level Two-tailed test (n = 20) aMeans minus the pretest reading 130 TABLE 8 Correlational Matrix of Physiological Measures of Anxiety for the Group with Neutral Instructions 1. 2. 3. 4. 5. 6. 7. 8. 9. * o 1 —1 1. Resp. Rate - per min. 1.00 0.04 0.15 0.3,6 o 0 0 1 0.42 -0.48 0.41 0.02 0.13 2. Resp. Depth3 Respiratory cycle 1.00 -0.36 -0.29 -0.24 -0.24 -0.03 0.04 0.08 -0.18 3. EKG - per min. 1.00 -0.25 0.14 0.40 -0.10 0.34 0.00 -0.05 4. GSRa- micromhos 1.00 -0.18 0.23 -0.25 0.28 0.15 0.32 5. Systolic B.P.— mm Hg 1.00 -0.36 ** 0.88 -0.36 -0.08 0.16 6. Diastolic B.P.- mm Hg 1.00 ** -0.73 ★ 0.53 0.34 0.16 7. Pulse Pressure- nun Hg 1.00 * -0.54 -0.23 0.02 8. Oral Temp, degrees P. 1.00 0.27 -0.00 9. Face Temp, degrees F. 1.00 0.36 10. Finger Temp, degrees F. 1.00 ♦Significant at .05 level **Significant at .01 level Two-tailed test (n=20) Mean minus the pretest reading 131 TABLE 9 Correlational Matrix of Physiological Measures of Anxiety for the Group with Anxiety-Reducing Instructions -. 1. 2. 3. 4. 5. 6. 7, 8. 9. 10. 1. Resp. Rate- per min. 1.00 0.19 -0.12 -0.00 0.11 -0.07 0.12 -0.03 -0.15 0.02 2. Resp. Depth3 Respiratory cycle 1.00 0.04 -0.16 0.15 0.37 -0.07 -0.32 -0.08 0.18 3. EKG - per min. 1,00 -0.21 -0.40 -0.10 -0,32 0.13 -0.03 0.21 4. GSRa- micromhos 1.00 -0.15 0.41 -0.33 0,11 -0.10 * -0.48 5. Systolic B.P.- mm Hg 1.00 0.16 ** 0.90 0.16 -0.19 -0.25 6. Diastolic B.P.- mm Hg 1.00 -0.28 0.12 -0.22 -0.10 7. Pulse Pressure- mm Hg 1.00 0,15 -0.13 -0.24 s. Oral Temp, degrees F. 1.00 -0.18 0.08 9. Face Temp, degrees F. 1.00 ** 0.56 10. Finger Temp, deqrees F. 1.00 I *Significant at .05 level **Significant at .01 level Two-tailed test (n = 20) aMean minus the pretest reading 132 133 rate, r = .88 (p .01) with systolic blood pressure, r = -.73 (p <. .01) with diastolic blood pressure and r = -.54 (p .05) with oral temperature. The fifth significant correlation with this group was r = -.53 (P <T .05) between diastolic blood pressure and oral temperature. Two of the three relationships which correlated significantly for the anxiety-producing instructions | | were concerned with finger temperatures. Finger ! j temperature correlated r = -.48 (p ^.05) with GSR and j ] r = .56 (p <^„01) with face temperature, The third ! significant correlation with this group was r = .90 j (p .01) between pulse pressure and systolic blood pressure. | The other relationships did not attain statistical I i | significance. The results showed the group receiving ! neutral instruction to have the largest number of i significant correlations. This group had five correla tions as compared with two for the group receiving anxiety-reducing instructions and three for the group receiving anxiety-producing instructions. However, by using the chi square test (Siegel, 1956, pp. 175-179), 5 the differences in the number of significant correlations ! for the various instruction groups failed to reach the ! .05 level. The chi square of 1.51 (p <* .50) indicated 134 the probability that the differences in the number of significant correlations for the instruction groups were due to chance. The low correlations between physiological measures have been found by previous investigators. After a study of the literature on physiological i j measures, Martin (1961) concluded that "Research thus far j | gives little ground for optimism that these variables will correlate very highly, if at all." ; The low correlations in this investigation i | suported his general hypothesis "... that there is I | marked uniqueness in physiological expression of emotion." j To summarize, out of the 135 possible correla- ! tions, 10 reached the .05 level of significance. When i i | the spurious relationships were ignored, the overriding | | conclusion reached was that the physiological measures ! are essentially unrelated, and do not provide the basis for the identification of a preferred measure of anxiety. The uniqueness of the individual examinee's response pattern was expected, but not to the extent observed. | Hypothesis Seven There will be a positive relationship between i the following psychological measures of anxiety: S-R 135 Inventory of Anxiousness (S-R I), Affect Adjective Check List (AACL), I. G. Sarason's Test Anxiety Scale (TAS), j and Bendig's short form of Taylor's Manifest Anxiety | Scale (MAS). i t Results and Discussion Table 112a-112d in Appendix B provides the means and standard deviations for the Student Survey, composed of the four psychological tests administered approximately two weeks prior to the major testing. i | As can be observed, the means vary only slightly i from one treatment group to another. This was expected | since the subjects were assigned at random to the treat- ! ; ment groups and since the administration of the student survey was independent of the treatments, All the psychological measures of anxiety were | | positively related to each other, as observed in Table i i | 10. All the correlations among the measures taken prior to the time of the administration of the Ability Test were significant at the .01 level. These correlations ranged from r = .36 (p < .01) between the MAS and the AACL to r = .65 (p .01) between the TAS and the AACL. As presented in Table 11 the girls evidenced greater | anxiety on all the scales except the MAS. No signifi- j cant difference emerged between the sexes for the MAS i and with the AACL presented after the anxiety TABLE 10 Correlations Among Psychological Measures for the Total Group 1. 2. 3. 4. 1. S-R I 1.00 0.40** 0.59** 0.41** 2. AACL 1.00 0.65** 0.36** 3. TAS 1.00 0.64** 4. MAS 1.00 **Significant at .01 level One-tailed test (n = 100) 136 137 TABLE 11 Summary of Analysis of Variance for the Psychological Test According to Sex F Mean Test Sex Highest S-R I 11.85 Girls** AACL 9.30 Girls** TAS 13.92 Girls** MAS 3.74 Neither AACL after Anxiety Instruction 0.52 Neither **Significant at .01 level 138 instructions. (Also see Tables 113-117 in Appendix B.) ! Hypothesis seven was confirmed. In every case a positive correlation was evident between the psycho logical measures of anxiety administered approximately two weeks prior to the major testing. The correlations in the present investigation compare favorably with ! those of other investigators; and considerable community j of measurement is reflected. t i j Hypothesis Eight The Affect Adjective Check List, administered prior to the ability test, will have a positive i relationship within each treatment group with the scores on the same measure obtained immediately following the instructions for the ability test. As observed from Table 12, hypothesis eight j ; was confirmed. The AACL administered as a part of the I Student Survey prior to the test was positively related to the same scale administered after the treatment instructions. The correlations between the two testing occasions with the AACL were unexpectedly low; however, i I it does seem that the less anxiety-producing the situa tion, the more related the two testings were. As can be TABLE 12 Correlations, Means and t Values for the AACL Administered Two Weeks Prior to the Major Testing and Administered Again After the Anxiety Instructions Instruction Groups Correlations between 1st & 2nd Testing of the AACL Means of Usual Test Anxiety (1st testing) AACL Mean after Instructions (2nd testing) AACL t Between the Two Testings 1. Anxiety-Reducing .28 12.35 7.90 2.Q3** 2 o Neutral ,62** 12.10 8.20 2.1Q** 3. Anxiety-Producing .13 12.70 8.15 2.52* 4. Control-for Phys, measures .63^ 12.35 6.50 4.04** 5, Control for test performance .27 11.75 9.90 1.21 Total Means 12.25 8113 ♦Significant at ,05 level ♦♦Significant at ,01 level n = 20 139 140 observed form Table 12, the two instruction groups providing significant correlations were the neutral and the control for physiological measures. Of all the groups, these two were probably under the least amount of anxiety. Tables 118 and 119 in Appendix B present the means, standard deviations and standard error of the means (see Tables 118 and 119) for the two testings of the AACL. It is apparent that with every group the AACL i indicated greater anxiety by the students during a regular examination than was experienced after the anxiety instructions in the experimental study. Higher AACL scores indicate greater anxiety. Since these j results were unexpected, the investigator interviewed individuals, asking them on which occasion they had experienced the most anxiety, i.e., during a regular important examination or during the experimental examination. Most of them replied that they didn't know. Some added that they thought they experienced more during a regular classroom examination. Others said they experienced more anxiety when the physiological measures were being taken. Despite the uncertainty of the perception of their emotions, their responses on the \ AACL were consistently higher for the second testing. As presented in Table 12, with all the treatment groups 141 except with the control for test performance, the means for the two testings were significantly different. These differences were significant at the point .01 level for the groups with anxiety-reducing (t = 2.83 p. < .01) and neutral (t = 2.78 p. < .01) instructions and the control for physiological measures (t = 4,04 p. .01) . The difference between the means for the group with the anxiety-producing instructions (t = 2.52) was significant beyond the .05 level. Hypothesis Eight was confirmed. A positive correlation was evidenced between the two testings of the AACL. ' J | Hypothesis Nine i There will be significant relationships between the physiological and psychological measures of anxiety for each of the three treatment groups. Results and Discussion Because the test anxiety scales are usually administered at a time other than an examination time by most investigators, the psychological measures in this study were presented to the subjects approximately two \ weeks prior to the administration of the Academic i Ability test and the recording of the physiological I measures. The instructions were to respond to the scales 142 according to their usual feelings during an important examination. Therefore, the psychological measures, with the exception of the AACL, which was again presented after the anxiety instructions, were not influenced by the anxiety treatments in any way. The result of this part of the study will be discussed according to treatment groups only to determine the relationship between scores on the psychological scales and physiological measures taken under varying conditions of anxiety. measures of anxiety that were significantly correlated with physiological measures under the anxiety-reducing instructions were the S-R I and the TAS. Significant negative correlations were obtained between the TAS and pulse pressure (r = .57 p .01)• Significant correlations were also computed between the S-R I and pulse pressure (r = -f49 p ^ .05) and between the S-R I and systolic blood pressure (r = -.61 p ^.01). With neutral instructions, as presented in Table 14, the significant correlations were concerned only with the AACL administered after the treatment instructions. Under these conditions, the AACL was significantly and positively correlated with respiration rate (r =,.51 No correlations in the group with anxiety-producing As presented in Table 13, the only psychological and with oral temperature r = .48 p .05) . TABLE 13 i Correlations between Physiological and Psychological Measures of Anxiety for the Group with Anxiety-Reducing Instructions S-R I AACL TAS MAS AACL After Instruc tion 1. 2. Resp. Rate - per min. Resp. Deptha-typical 0,14 0.15 0.08 i —i CN • O 1 -0.11 resp. cycle -0.04 0.08 0.25 -0.12 0.02 3. ERG - per rain. 0.16 0.20 0.16 0.31 0.31 4. GSRa- micromhos 0.21 -0.04 0.24 0.27 0.15 5. Systolic B. P. - mm Hg -0.61^ -0.39 -0.41 -0.34 -0.10 6. Diastolic B. P. - mm Hg -0.34 -0.26 0.05 -0,05 -0.20 7. Pulse Pressure - mm Hg -0.49# -0.23 -0.57^ -0.34 0.03 8. Oral Temp, degrees F 0.37 -0.10 0.29 0.16 0.21 9. Face Temp, degrees F -0.18 -0.19 -0.05 0.00 -0.34 10. Finger Temp, degrees F 0.11 0.14 0.28 0.43 0.19 ♦Significant at .05 level Two-tailed test (n = 20) ♦♦Significant at .01 level aMean minus the pretest reading 143 TABLE 14 Correlations between Physiological and Psychological Measures of Anxiety for the Group with Neutral Instructions S-R I AACL TAS MAS AACL After Instruc tion 1. 2. Resp. Rate - per min. Resp. Deptha- typical 0.12 0.20 0.09 0.07 O e Ul | _ w ) resp. cycle -0.04 -0.22 -0.35 -0.29 -0.11 3. EKG - per min. -0.03 0.30 0.06 0.09 0.18 4. GSR3 - — micromhos 0.03 0.13 0.14 0.34 0.20 5. Systolic B. P. - mm Hg -0.06 -0.11 -0.25 0.01 -0.37 6. Diastolic B. P. - mm Hg 0.01 0.07 -0.04 0.11 0.26 7. Pulse Frassure - mm Hg -0.05 — 0 . 10 -0.11 0.04 -0.37 8. Oral Temp, degrees P 0.15 0.25 0.09 0.16 0.48 9. Face Temp, degrees F -0.28 -0.42 -0.30 -0.13 0.01 10. Finger Temp, degrees F -0.38 -0.26 -0.23 -0.03 0.10 *Significant at .05 level Two-tailed test (n = 20) aMean minus the pretest reading 144 145 instructions reached the .05 level of significance. (See Table 15.) ! Out of 150 correlations, seven or eight signifi- i I | cant correlations would be expected by chance alone. Since only five of the correlations between physiological and psychological measures reached the .05 level of | significance, the information supplied by them must be I I ! considered essentially nonsignificant and due to chance. Similar low correlations between physiological and psychological measures of anxiety have been noted by | j other investigators. j The obtained results of this investigation under the conditions of the study suggest that the psychological responses to anxiety questionnaire items i and physiological responses are different characteristics. It would appear that the way an individual thinks he feels, evidenced by responses on the anxiety scale, is ! J unrelated to the physical responses of the emergency syndrome. From a theoretical standpoint, perhaps our scales of anxiety are not measuring how the individual responds internally, but rather how he thinks he feels or how he wishes to appear to feel. It could be possible that how an individual thinks he feels would have a greater influence on test performance than the way his i body responds physiologically, if he is concentrating TABLE 15 Correlations between Physiological and Psychological Measures of Anxiety for the Group with Anxiety-Producing Instructions S-R I AACL TAS MAS AACL After Instruc tion 1. 2. Resp. Rate - per min. Resp. Deptha- typical -0.24 0.28 0.29 0.09 0.31 resp. cycle 0.20 0.01 -0.30 -0.27 -0.00 3. EKG - per min. 0.30 0.26 0.15 -0.06 -0.34 4. GSR^-micromhos 0.22 0.12 0.30 0.11 0.32 5. Systolic B. P. - mm Hg -0.19 -0.18 -0.30 -0.05 0.26 6. Diastolic B. P. - mm Hg 0.12 -0.08 -0.26 -0.13 0.16 7. Pulse Pressure - mm Hg -0.24 -0.13 -0.17 0.01 0.19 8. Oral Temp, degrees F. -0.02 -0.03 -0.07 -0.26 0.02 9. Face Temp, degrees F. 0.18 -0.02 0.13 0.24 -0.05 10. Finger Temp, degrees F. -0.18 -0.05 -0.20 -0.07 —0.10 Two-tailed test (n = 20) a Mean minus the prereading 146 147 on the test and is relatively unaware of these responses. Hypothesis Ten There will be significant relationships within each treatment group between psychological measures of anxiety and scores on the Acadeirtic Ability Test. Results and Discussion (See Tables 16-18) The only anxiety scales that were significantly correlated with test performance were the TAS and the i AACL administered prior to the testing situation and given again after the treatment instructions. The TAS was significantly and negatively correlated with performance for the total scores and mathematical scores with the anxiety-reducing and neutral instruc tions. But under the anxiety-producing instructions it was significantly and positively related to the total i test score. All these correlations varied only slightly, giving rise to r = -.48 (p .05) with the mathematical scores under the influence of the anxiety- reducing and neutral instructions and with the total scores for the neutral instructions. The relationship between the TAS and the total scores with the anxiety- reducing instructions was a negative correlation of -.46 (p .05), whereas the reverse relationship, that of a positive correlation of .46 (p .05), was TABLE 16 Correlations between Psychological Measures of Anxiety and Scores on the Academic Ability Test with Anxiety-Reducing Instructions Academic Ability Test Total Verbal Math. Score Score Score 1. S-R I -0.26 -0.05 o 0 0 1 2. AACL -0.06 0.02 o 1 — { « 0 1 3. TAS -0.46* -0.26 -0.48* 4. MAS -0.36 -0.31 00 I— 1 « o 1 5. AACL after Anxiety Instructions 0.23 0.20 0.12 *Significant at .05 level Two-tailed test (n = 20) h - * o o TABLE 17 Correlations between Psychological Measures of Anxiety and Scores on the Academic Ability Test with Neutral Instructions Ability Test Score Total Verbal Math. Score Score Score 1. S-R I 0.24 0.35 0.09 2. AACL -0.32 -0.27 -0.32 3. TAS -0.48* -0.41 -0.48* 4. MAS 0.05 0.03 0.06 5. AACL after the Test Instructions -0,46* -0.49* -0.37 *Significant at .05 level Two-tailed test (n = 20) 149 TABLE 18 Correlations between Psychological Measures of Anxiety and Scores on the Academic Ability Test with Anxiety-Producing Instructions Ability Test Total Verbal Math. Score Score Score 1. S-R I 0.09 0.14 0.00 2. AACL 0.53* 0.58** 0.34 3. TAS 0.46* 0.41 0.36 4. MAS 0.27 0.08 0.38 5. AACL given just before the Test Instructions 0.12 0.03 0.14 *Significant at .05 level **Significant at .01 level Two-tailed test (n = 20) / \ 150 151 obtained for the anxiety-producing instructions. A positive relationship was obtained with the anxiety-producing instructions for the AACL administered prior to the testing situation, r = .53 (p .05) for the total scores and r = .58 (p <^.01) for the verbal scores. These were the only significant correlations with the AACL administered prior to the testing situation? however, when given after the treatment instructions, negative correlations were evident as, r = -.46 (p <^.05) for the total scores and r = -.49 (p <.05) for the verbal scores. In summary, of the significant correlations between performance and psychological measures, the TAS and the AACL were the only ones significantly related to test performance. It was interesting to note that the tendency was for students with high test performance under the anxiety-reducing and neutral instructions to be the low-anxious individuals, whereas under the anxiety- producing instructions the best test performance was evidenced by the high-anxious students. In general, low to moderate negative relation ships have been found between anxiety, as measured by different psychological questionnaires, and scores on an ability test by various investigators (Cowen, 1957? 152 I. G. Sarason, 1961b/ 1963). Similar results were obtained in the present investigation. The present findings clearly indicate that the TAS was the single psychological measure of anxiety most consistently related to test performance under varying anxiety stimuli. The next most related scale to test performance was the AACL. The MAS and the S-R I did not relate significantly with test performance. The results indicate that a scale designed to measure the anxiety experienced during an examination is more related to test performance than is a more general measure of anxiety such as the MAS. Hypothesis Eleven Within each treatment group there will be significant relationships between physiological measures of anxiety and performance on an ability test. Results and Discussion As presented in Tables 19-21, out of a possible 30 correlations between physiological measures and test performance for each treatment group, there were 8, 0, and 2 significant correlations for the anxiety-reducing neutral, and anxiety-producing instruction groups, respectively. 153 Of the eight significant correlations between physiological measures and test performance under the anxiety reducing instructions/ respiration rate, heart beat rate, diastolic blood pressure, and pulse pressure were related to test performance. Heartbeat rate related to test performance with significant negative correlations of r = -^.50 (p < .05) , r = -.44 (p <^".05) , and r = -*52 (p <.05) for the total, verbal, and mathematical scores, respectively. Pulse pressure correlated at r = .50 (p <T .05) and r = .71 (p <.01) with the total and mathematical scores respectively. The correlations between diastolic blood pressure and i I j the total and verbal scores were r = -.55 (p .05) and r = -.61 (p .01) respectively. One other signi ficant correlation was that of r = -.50 (p -C.05) between respiration rate and mathematical scores, j No significant correlations were obtained between physiological responses and test performance with neutral instructions. The highest correlation was that of .34 between respiration depth and the mathemati cal scores. Only one physiological measure, that of oral temperature, related significantly with responses on the Academic Ability Test under the influence of the anxiety' producing instructions. Correlations were r = -.46 TABLE 19 Correlations between Physiological Measures of Anxiety and Scores on the Academic Ability Test with Anxiety-Reducing Instructions Academic Ability Test Total Score Verbal Score Math. Score 1. Respiration Rate - per min. -0.33 -0.15 -0.50* 2. Resp. Deptha - typical resp. cycle -0.01 0.06 -0.11 3. EKG - per min. -0.50* -0.44* -0.52* 4. GSRa- micromhos 0.11 0.13 0.05 5. Systolic Blood Pressure - mm Hg -0.02 -0.29 0.31 6. Diastolic Blood Pressure - mm Hg -0.55* -0.61** -0.30 7. Pulse Pressure - mm Hg 0.50* 0.22 0.71** 8. Oral Temperature degrees F. -0.11 0.17 -0.33 9. Face Temperature degrees F. 0.01 -0.03 0.13 10. Finger Temperature degrees F. -0.35 -0.30 -0.23 *Significant at .05 level **Significant at .01 level Two-tailed test (n = 20) aMean minus the pretest reading 154 TABLE 20 Correlations between Physiological Measures of Anxiety and Scores on the Academic Ability Test with Neutral Instructions Academic Ability Test Total Verbal Math. Score Score Score 1. Respiration Rate - per min. -0.00 0.13 -0.13 2. Resp. Deptha - typical resp. cycle 0.31 0.24 0.34 3. EKG - per min. -0.12 -0.06 -0.17 4. GSI^ - micromhos -0.18 -0.09 -0.24 5. Systolic Blood Pressure - mm Hg 0.13 0.14 0.11 6. Diastolic Blood Pressure - mm Hg 0.01 -0.06 0.07 7. Pulse Pressure - mm Hg 0.15 0.17 0.11 8. Oral Temperature degrees F. -0.23 -0.19 -0.24 9. Face Temperature degrees F. 0.19 0.16 0.19 10. Finger Temperature degrees F. 0.06 -0.02 0.13 Two-tailed test (n = 20) Mean minus the pretest reading 155 TABLE 21 Correlations between Physiological Measures of Anxiety and Scores on the Academic Ability Test with Anxiety-Producing Instructions Academic Ability Test Total Score Verbal Score Math. Score 1. Respiration Rate - per min. 0.38 0.42 0.23 2. Resp. Depth3 - typical resp. cycle 0.02 0.06 -0.08 3. EKG per min. -0.23 -0.18 -0.28 4. GSR3 - micromhos 0.06 0.20 -0.14 5. Systolic Blood Pressure - mm Hg -0.10 -0.09 -0.04 6. Diastolic Blood Pressure - mm Hg -0.14 -0.02 -0.31 7. Pulse Pressure - mm Hg -0.04 -0.08 0.09 8. Oral Temperature degrees F. -0.46* -0.42 -0.48* 9. Face Temperature degrees F. -0.08 -0.18 0.03 10. Finger Temperature degrees F. -0.28 -0.38 -0.18 *Significant at .05 level Two-tailed test (n = 20) aMean minus the pretest reading 156 157 (p < .05) between oral temperature and the total scores, and r = -.48 (p <„05) with the mathematical scores. To summarize, with the anxiety-reducing instruc tions, eight correlations out of a possible thirty were significant. For the other treatment groups, only two correlations were significant and those Were with anxiety-producing instructions. The implications of the results of this study are that there is, in general, no relationship between physiological measures of anxiety and intellectual performance under varying degrees of anxiety. Under the anxiety-reducing, instructions, better performance was suggested by slower respiration rate, slower heartbeat rate, lower diastolic blood pressure, and greater pu^se pressure. Summary The findings in the present investigation suggest the following statements of observation. 1. Under the conditions of the present investiga tion, there was no significant difference in performance under various types of instruction stimuli. 2. No significant difference in performance emerged under the various types of instruction stimuli with the initial level of anxiety held constant? but 158 there were significant instructions by sex interactions with the anxiety-producing instructions when the initial I level of anxiety was held constant„ 3. Respiration rate was the only physiological | measure that differed significantly according to ! treatment groups. It is possible that the measure was i subject to some inaccuracy, since the equipment used was sensitive to bodily movement as well as to respiration. Even though an attempt was made to reduce bodily movement as far as possible, some did inevitably I occur. The physiological measures were essentially j independent, a finding that confirms the complexity j j of the construct. 4. The boys' measures for systolic blood pressure, pulse pressure, face temperature, and finger temperature were significantly higher than those for the i | girls. Respiration rate and oral temperature were I ! significantly higher for the girls than for the boys. There was no.sex difference in responses for changes in respiration depth, heartbeat rate, GSR, and diastolic blood pressure. However, higher scores do not necessarily indicate a greater degree of anxiety experienced. It is quite likely that the girls evidenced greater levels of anxiety than the boys by oral temperature, face tempera- I ture, finger temperature and possibly-by respiration 159 rate (if possible inaccuracies are ruled out). 5. Unlike the physiological measures, all four of the psychological measures of anxiety were positively and moderately interrelated. The correlations were significant at the .01 level, 6. The AACL administered prior to the major testing as part of the Student Survey was significantly correlated with the scores of the same scale given after the anxiety stimulus only for the neutral anxiety group (r = .62 p .01) and the control for physiological measures (r = ,63 p .01) 7. The physiological and psychological measures of anxiety were generally not correlated to a signifi cant degree. 8. Correlations between the psychological measures of anxiety and the Academic Ability Test performance were largely influenced by the rationale inherent in their construction. The TAS and the AACL were moderately related with performance on the ability test. The MAS, designed to measure general anxiety, and the S-R I, largely concerned with situational anxiety, were essentially unrelated to ability test performance. 9. The significant correlations between physiological measures of anxiety and performance on the Academic Ability Test were largely concerned with oral X60 temperature, heartbeat rate, pulse pressure, diastolic blood pressure, and respiration rate. The number of significant correlations is not greatly in excess of the number of correlations which would be significant by chance. CHAPTER VI SUMMARY, RECOMMENDATIONS AND CONCLUSIONS i The purpose of this chapter is to summarize the investigation and the results of the study and to present recommendations that have developed. | Summary | Problem The principal problems 6f the study concerned the following issues? (a) the extent to which experimentally induced anxiety influences ability test performance? and (b) the extent to which the various | physiological and psychological measures of anxiety are related. j I Procedure All high school seniors enrolled in any of the nine sections of the United States Government classes at West High School in the Torrance Unified School District were administered the following four self- i report anxiety measures: (a) Endler, Hunt, and ; Rosenstein's S-R Inventory of Anxiousness, (b) Zuckerman' : Affect Adjective Check List, (c) I. G. Sarason's Test 161 162 Anxiety Scale, and (d) Bendig's Short Form of Taylor's Manifest Anxiety Scale. In harmony with the requirement of the principal, i ) request for parental permission to involve the students j in the study was sought until an excess of 100 subjects were available. The students were then assigned at | random to one of five treatment groups, stratified by sex and proficiency level (by ITED scores). Thus there | were five subjects in each of the twenty cells of the 5X2X2 (treatments by proficiency levels by sex) design. In three of the five groups, each subject received one of the following types of anxiety stimuli: i (a) anxiety-reducing instructions, (b) neutral instruc tions, and (c) anxiety-producing instructions. Two j separate control groups were used: one which took the j j same tests but without concomitant physiological instru- j | mentation, and another which had the instrumentation ! but was engaged in non-test behavior (reading), i I I The three treatment groups were administered the i j' ] Academic Ability Test concurrently with the recording of | the following physiological measures: respiration rate, i respiration depth, heartbeat rate, galvanic skin response, systolic blood pressure, diastolic blood pressure, pulse | pressure, oral temperature, face temperature, and finger ; temperature. The other two groups were controls. The 163 control for test performance took the ability test without the physiological measures being recorded at the time of testing. These, however, had pulse rate, oral temperature, systolic blood pressure, and diastolic blood pressure recorded at the completion of the test. All groups took the AACL (part II of the Student Survey) after the anxiety stimulus (instructions) with directions to score it according to their present feelings. Analysis of the Data The basic statistical model for the analysis of the data was a three-way analysis of variance (treat ments by proficiency levels by sex). The dependent variables included the verbal, mathematical, and total ability test scores, as well as the physiological measures. The organismic variables were the psychological measures and the pretest physio logical responses. Analyses of covariance with the same factorial design was performed with the ability test scores using the initial level of anxiety (pretest reading) held constant. Correlational analyses were also performed on all anxiety measures. Findings An analysis of the data provides the following summary statements» 164 1. There was no significant difference among treatment groups with respect to performance on the verbal, mathematical, or total ability test scores, with or without the initial anxiety level held constant. The lack of significant difference in performance among treatment groups with varying degrees of anxiety instructions indicated either that performance was not affected by anxiety or that the various instructions did not influence anxiety, A lack of significant differences between treatment groups was also observed by Allison (1964), who manipulated five experimental conditions. The results revealed no significant differences in perfor mance for anxiety levels, experimental conditions, sex or any of the interactions. The present study did obtain significant sex vs, instruction (anxiety stimuli) interactions when the initial anxiety level was held constant. Similar results were observed by Smith (1964-65) who found no significant difference in performance between stressed and nonstressed subjects irrespective of sex. When the data were analyzed by sex, the high anxious males, under stress, suffered a significant impairment in performance, whereas the performance of the female subjects was not disturbed by anxiety and/or stress. In the present study, even though the difference was not significant, the performance of the females was facilitated by the anxiety-producing instructions, whereas the reverse effect was true for the boys. One explanation Smith presented for the differences in I performance of males and females under stress was that j highly anxious male subjects under stress are motivated either to reduce the anxiety or to eliminate the stressful conditons rather than attend to the task. On the contrary, according to him, the highly anxious female subjects are able to consistently perform under stress on academic tasks because to them anxiety is ; viewed as acceptable and therefore not disturbing, | Another factor that might be considered in the present study concerns the sample. It is to be remember- I ed that parental permission was required by the school for subjects taking part. Perhaps a larger number of females than males who feared their performance would be impaired refused to take part in the study. The limitations imposed on the sample by required permission may also serve to explain the lack of significant differ ence in performance between the treatment groups. i 2. Among the physiological measures, respiration i rate alone was significantly different among the 166 treatment groups. This fact suggests the strong possibility that in actuality the instructions, although phrased as strongly as ethically possible, did not influence differential anxiety in the examinees. 3. There were no significant differences between the proficiency levels for the physiological responses. 4. The physiological measures of face tempera- | ture, finger temperature, systolic blood pressure, and : pulse pressure were significantly higher for the males than for the females. None of these elevated responses need indicate greater anxiety on the part of the males. ; Stevens (1951, p. 477) reported that emotional stress j produced a fall in skin temperature. Therefore the I higher scores for the males than for the females for | face and finger temperature would indicate less anxiety. ! t The systolic blood pressure is 4 or 5 mm Hg higher for t I the females than for the males (Best & Taylor, 1961, pp. 274-275). This physiological difference between the sexes could account for the higher systolic blood pressure and pulse pressure for the boys. Respiration rate and oral temperature were s I significantly higher for the females than for the males. i j An elevation of oral temperature associated with an examination has been observed by Chambers (1962), Smith | 167 and Wenger (1965), and Wynn (1919)„ No significant dif- i ferences were obtained between the responses of the males ! and females in diastolic blood pressure, GSR, respiration : depth, and heartbeat rate. Smith and Wenger (1965) did not use respiration depth as a measure; however, they did find a significant difference in responses in diastolic i i blood pressure and heartbeat rate but not in GSR, 5, Systolic blood pressure consistently showed a significant positive relationship to pulse pressure under all three anxiety stimuli (instructions). The neutral instruction group yielded more significant correlations than the other two treatment groups. However, when chi square by (Siegel, 1956, pp. 174-179), was computed for ! significant differences in number of significant correla- J | tions between the treatment groups, no significant differ- j ence among groups was found. Pulse pressure was signifi- j cantly related to: a. systolic blood pressure i b. diastolic blood pressure c. respiration rate | d. oral temperature ; Other significant correlations were: a. respiration depth with heartbeat rate b. respiration depth with systolic blood pressure c. diastolic blood pressure with oral temperature 168 d. face temperature with finger temperature Out of a total of 90 correlations the number of ; significant correlations was not much different from what ! was expected by chance. This is especially true when the relationships between the related measures, such as the three types of blood pressures, are eliminated. There- i fore, the important generalization evolving from the ! j I interrelationship is that they were essentially independ- i ent; i.e., they were generally not correlated either in absolute magnitude or in change. | Low correlations among physiological measures have been observed by other investigators. Ax (1953) was sur prised to find the general lack of correlation among the j variables. He concluded that "there is a marked unique- I ness in physiological expression of emotion." Lewinsohn | (1956) and Lacey and his associates (J. I. Lacey, Kagan, i I Beatrice C. Lacey, & Moss, 1963, p. 162) also found startlingly low correlations among autonomic responses. Lacey and collaborators did not attribute the low cor- ! relations solely to attenuation of errors of measurement I j or to other sorts of unreliability, but they established | the fact that there are reliable individual differences in patterns of autonomic response to a given stressor and even to different stressors. 6. All psychological measures of anxiety were 169 positively and moderately related to each other. Correlations among the four psychological measures ranged from .36 between the AACL and the MAS to .65 between the TAS and the AACL. All correlations were significant beyond the .01 level. Psychological measures of anxiety have generally been found to intercorrelate moderately. The correlations in the present study were comparable to those obtained by other investigators. The correlation of .64 between the MAS and the TAS in the present study is compared to .46 for males and .53 for females computed by I. G. Sarason (1961a); r = .41 between the MAS and the S-R I as opposed to .46 and .34 found by Endler and associates (1962); and r = .36 between the MAS and the AACL against .40 for the first examination day observed by Zuckerman (1960). The correlations indicated a fair amount of common variance among the scales. 7. The AACL administered prior to the major testing as part of the Student Survey was only moderately related to the AACL administered just following the anxiety stimulus. No significant relationships were obtained between the two testings with the AACL for the anxiety-reducing (r = .28) and the anxiety-producing (r = .13) instruction groups and for the control for test 170 performance (r = .27) » There were significant correla tions between the two testings for the neutral instruction group (r = ,62 p <[ . 01) and the control for physiological measures (r = .63 p <.01), With every group the AACL indicated greater anxiety in the students during a regular examination than was experienced after the anxiety instructions in the experimental study. It may be that they really did not perceive this examination to be as anxiety-provoking as a regular important classroom examination. However, Windle (1954) found individuals to rate themselves better on a personality test the second time than on the first. In this study, since the higher scores represent greater anxiety, these students did, on the I average, rate themselves better on the retest, thus providing lower scores on the second testing. By use of the t test this difference was significant for all groups except the control for test performance. This group was the only group that did not have the imposed physiological measures. They did have the examination, but their feelings were evidently about the same as dur- | ing a usual important examination. 8. The physiological and psychological measures ! were essentially uncorrelated. Out of the matrix of | 150 correlations only five attained the .05 significance 171 level. A survey of the psychological, literature indicated that the findings were compatible with those of previous investigators. Winter and associates (.1963) investigated the relationship between the Palmar Sweat Index (PSI) and two psychological measures, the MAS and the AACL. No significant correlations emerged. Endler and associates (1962) investigated the relationship between the same physiological measure, the PSI, and the IPAT, MAS, and S-R I. Only with the IPAT did he obtain a significant correlation with the PSI. Lewinsohn (1956) determined the relationship between the MAS and four physiological measuress Finger tremor, heart rate, salivary output, and GSR. His results showed that only between finger tremor and the MAS was a significant correlation ob served. No adequate explanation has presently been advanced for the general lack of interrelationship between psychological and physiological measures. Con sideration should be given to the possible defensiveness involved with the psychological scores and the fact that conditions other than anxiety can influence physio logical measures. The results of the present and previous studies suggest that the two types of responses are measurements of different characteristics. 172 9. The only psychological measures significantly related to test performance were the TAS and the AACL, and these were not related under all conditions. An 'Unexpected finding was the absence of significant |correlations between the TAS and the verbal scores; i I ihowever, the correlations were short of significance j !under the neutral and anxiety-producing instructions. i ;A second interesting observation was the direction of the relationships. Under the anxiety-reducing and neutral instructions, low anxiety scores were associated with high test performance, whereas under the anxiety- producing instructions, high anxiety scores were associat ed with high test performance. In general, small to ;moderate negative relationships have been found between | test anxiety and scores on an ability test by various [ | investigators (Cowen, 1957; Sarason, I. G., 1960, 1961b, i |1963). Similar results were obtained in the present j investigation for anxiety-reducing and neutral instruc- ] Itions. The positive relationship between anxiety and ! !performance under the anxiety-producing instructions was I : due to the facilitative effect the anxiety producing I I | instructions had on the performance of the females. | The MAS and the S-R I were not related to tested performance under any of the conditions studied. Previous reports have described comparable 173 results. Studies have shown the TAS to be more related to intellectual performance than the MAS (Sarason, I. G., ; I i 1961b? Sarason, I.G. & Palola, 1960? Walter, Denzler & I. j j | G. Sarason, 1964). Eaber and Spence (1955), over a ! i j period of years, were unable to find any relationship | between the MAS and conventional measures of intellectual j ’ j | ability, such as entrance examination scores and grade- j I j j point average. ' I | i \ The results of the present and previous investi gations clearly indicate that a scale designed to measure I the anxiety experienced during an examination is more ; related to test performance than a scale designed to ; measure more global anxiety. j .10. All but one of the six significant correlations between the physiological measures and | scores on the Academic Ability Test for the verbal and ] 1 s mathematical scores were with the groups that received | the anxiety-reducing instructions. Significant | correlations were obtained between scores on the Academic Ability Test and the following measures: oral temperature, j heartbeat rate, pulse pressure, diastolic blood pressure, ! | and respiration rate. All these correlations were nega- j | tive except pulse pressure. Considering the chance | : factor, the number of significant correlations was not impressive. 174 There is a paucity of research concerned with physiological measures and performance on mental tasks. The studies that have been conducted have generally investigated the changes in physiological responses under conditions of stress rather than the relationship between the responses and performance scores. However, Brown and Van Gelder (1938) and Milliken (1964) found the physiological measures to increase from easy to difficult material. 11. Since no significant differences were obtained between treatment groups for performance on the ability test, all groups were combined in order that a multiple correlation could be computed. This was done in an attempt to predict ability test scores from a I combination of physiological measures. Both absolute I and change readings were used. (See Table 22 in Appendix B). The multiple correlation coefficient (R * .49 p .01) appeared to be significant; however, after Guilford's (1965, p. 401) formula for correction for shrinkage was used, the correlation (R = .08) failed | to reach the level of statistical significance. This I finding suggests that either a. anxiety is not a hindrance in test perform ance within the limits of this study, or j b. test anxiety is not measured by these 175 physiological responses. These findings are independent of the question of whether t or not anxiety was experimentally manipulated. I 5 [ I Recommendations j Theory i Anxiety has been studied by many psychologists and as yet it is far from being understood. Anxiety as a Drive. Spence (1956, p. 165), | Taylor (1951) and others have studied the drive properties i I of anxiety. According to English and English (1958, p. 35), one definition of anxiety is, "an unpleasant I { emotional state in which a present and continuing strong ! desire or drive seems likely to miss its goal." In terms j j of an examination the drive may be towards a sought for ! | grade, be it a high "A" or a "just passing" "D". Horney ] | (1939) pointed out that our desires do not become drives unless they are motivated by anxiety. Mowrer (1950, pp. 65, 66) considered fear, which he equated with anxiety to be pre-eminently a drive, a goad to action. Under the conditions of the present study, anxiety did not serve as an influencing factor. However, although the differences in performance were not significant, the anxiety-producing instructions tended to be facilitative for the girls and detrimental for the boys. 176 It is recommended that consideration be given to the fact that the .'lack of significant difference in per formance between the groups would indicate that even though it did not function as a drive toward better performance, neither did it function as an inhibitor of performance,, Anxiety as Type- Endler and collaborators (1962) considered anxiety as situational. Smith and Wenger (1965) wrote of phasic anxiety. I. G. Sarason (1963) referred to test anxiety. Taylor (1953) was interested in manifest anxiety. Are these individuals investigating the same construct? In the present in vestigation all the anxiety scales were moderately related irrespective of the unique philosophy inherent in the construction of the instrument. Despite the high relationship among the anxiety scales, the ones designed to measure anxiety in the testing situation were more related to test performance than the scales designed to measure anxiety in general. This relation ship to performance has also been observed by other investigators (I. G. Sarason, 1959). Considering the large number of correlations studied, the number of significant relationships was not impressive between psychological measures and physiological measures nor among the physiological measures. However, it does shed some light on the topic. It is recommended that investigators take into account the fact that the j psychological measures and the physiological measures are measuring different factors, be they different types of anxiety, aspects of anxiety or degrees of anxiety. This is a topic that needs further investi- j gation. i Anxiety as a Normal Aspect of Life. As j | described by Marmor (1962), anxiety serves as a signal of present or future danger and within limits is not something abnormal to be annihilated if possible. In fact Janis (1958, pp. 395-412) found that moderate amounts of preoperative anxiety served to better prepare his surgical patients for the event. ; Likewise, the students in the present study, j i accepted the stressful conditions as part of their normal program and were not excessively disturbed by them. It is recommended that students accept moderate j anxiety and learn to adjust to the anxiety producing challenges of life. Anxiety as a Response or Behavior. Anxiety as I a response or behavior has been a topic of investigation | by many psychologists. It has been studied in terms of responses on paper-and-pencil scales and responses of physiological measures. The theory that both responses 178 are indicators of anxiety is interesting in view of the fact that the two types of measures are not significantly related, nor do they relate to test performance in the same way. One physiological measure is not related to another, and when all physiological measures are pooled the aggregate of measures do not relate significantly with test performance. Lacey, Kagan, Lacey, and Moss (1963) observed that, "The amount of reaction shown in one physiological function cannot be predicted with accuracy from the amount of reaction shown in another function." They also observed the low correlation among physiological measures. Prom a theoretical view point it is recommended that consideration be given to the fact that these responses are either measuring different factors or different aspects of the same factor. The cognitive awareness of anxiety is un related to the physiological syndrome of anxiety. Practical The results of the investigation suggest cer tain practical recommendations. Instructions. The concern psychologists have experienced over instructions for examinations in order to elicit the exact degree of anxiety for maximum performance does not seem to be as crucial as previously thought. It would seem that students are not 179 excessively influenced by type of instructions. It is recommended that the same healthy atmosphere recommended I for a classroom is suggested for the examination day. Appeal should be made to the student's intellect rather than to his emotions. Explanation as to the nature, : importances, and the correct procedure for the test should be presented. ! Student Remarks. Psychologists and educators j have repeatedly heard students voice remarks, such as: "I could have done better, but I became anxious or I became so excited I just couldn't think." It is true that the results of the present investigation were con cerned with the average or typical performance rather than extreme or rare cases. It is not recommended that i j any student's defense system be torn down; however, he could be told sometime prior to the examination that research has shown that anxiety has little or no effect on test performance. This might allow the student to perform with ease and to concentrate on the examination rather than think about his emotions. Anxiety in General Versus Test Anxiety. Stu- | dents and individuals who experience test anxiety will not necessarily experience anxiety in general. In j fact Ruebush (1963) found girls with high test anxiety i to be less deviant among girls than those with low test 180 anxiety. Students with high general or manifest anxiety may not necessarily suffer impairment in intellectual performance. It is recommended that a moderate amount of anxiety be regarded with favor if the individual is able to enjoy it and grow under it. Psychologists should be pleased that anxiety has so little effect on performance. This fact increases the robustness and validity of examinations. Methodology Anxiety Scales. It is recommended that careful consideration be given to the scales used and the construction of such scales to measure anxiety. What is measured seems to be largely determined by the way the investigator defines anxiety. The relationship between psychological measures of anxiety and scores on an ability test is largely influenced by the rationale inherent in the construction. Physiological Measures of Anxiety. Any measure of the emotions is of necessity obtained by an indirect method. The most frequently used procedure is the acceptance of verbal or written introspective reports of anxiety. According to Rush (1967, p. 423) and S. B. Sarason and Mandler (1952) self reports of anxiety cannot be relied on as precise descriptions of what actually is felt. 181 Physiological measures not normally vulnerable to conscious or unconscious faking or self deceit were thought to be more valid measures of anxiety. It is recommended that psychologists not consider the measurements of anxiety to be simple. Consideration should be given to the fact that psychological and physiological responses do not measure the same factors and/ to complicate the problem, the physiological measures are not highly interrelated. It is recommended that instrumentation be refined so that, as far as possible, variables, other than emotion be eliminated when recording physiological measures. Investigators should be careful, just as was done in this study, to take subjects for testing at random for various periods of the day and time of the year. This should minimize the effect on physiological responding due to time of the day and seasons. Testing Conditions. It is recommended that future research in this area avoid the pitfall of creating a testing situation that is too artifical to allow test anxiety to emerge. Conclusions Under the conditions of the present study, it is concluded that anxiety is manifested by extremely 182 intricate and perplexingly elaborate interwoven vari ables. Measures which were used singly as indicators of anxiety were unrelated to other measures used for the same purpose. For example, paper-and-pencil scales designed to measure physiological responses may be un related to the bodily expressions of anxiety. Likewise, two measures shown to be moderately related to each other, may be significantly disparate in terms of test performance. The effect of anxiety on performance, under the same conditions tend to be facilitative for one sex and detrimental for the other. Under conditions of stress, one physiological response may be accelerated and another physiological response for the same indivi dual be undisturbed, and yet according to the response of the same physiological variable, one person may suffer stimulation, while another may experience no change in response. One subject may express cognitive presence of anxiety and yet his internal environment provide no indication of the emotion, while another person with paper and pencil may express the presence of relatively no anxiety and at the same time have accelerated physiological responses. The most parsi monious statement that can be made is that anxiety in its complexity manifests great inter and intra individual differences. Not only are an individual's 183 emotions a mystery to others, but they may be to a degree unknown or unadmitted to himself„ However, it is concluded that one factor re mains predominant: regardless of the seemingly elusive character of anxiety, the intellect is capable of maintaining remarkable homeastasis. Irrespective of the effects anxiety may manifest upon the internal organism, the individual is able to engage, relatively unaffected, in the most distinctive of human behavior, that of thinking. I i References 184 References Aiken, Jr., L. R. Paper and pencil anxiety. Psychol. Reps., 1962, 10, 107-112. Allison, D„ E. An experimental study of the effect of anxiety and stress on the group intelligence test performance in elementary school children. Unpublished doctoral dissertation, Univer. of Southern California, 1964. American Psychiatric Association, Committee on Public Information. A psychiatric glossary. (1st ed.) Washington: American Psychiatric Association, 1957. Anastasi, Anne. Psychological testing. (2nd ed.) New York: Macmillan, 1961. Ax, A. F. The physiological differentiation between fear and anger in humans. Psychom. Med., 1953, 15, 433-442. Bendig, A. W. The development of a short form of the Manifest Anxiety Scale. J. consult. Psychol., 1956, 20, 384. Berry, J. L., & Martin, B. GSR reactivity as a function of anxiety, instructions, and sex. J. abnorm. soc. Psychol.. 1957, 54, 9-12. 185 186 Best, C. H., & Taylor, N. B. The physiological basis of medical practice„ Baltimore: Williams & Wilkins, 1961. Brown, C. H. Emotional reactions before examinations: II Results of a questionnaire. J. Psychol., 1938, 5, 11-26. (a) Brown, C„ H. Emotional reactions before examinations: III Interrelations. J. Psychol., 1938, 5, 27- 31. (b) 1 Brown, C. H., & Van Gelder, D. Emotional reactions before examinations; I Physiological changes. J. Psychol., .1938, 5, 1-9. Buros, 0. K. (Ed.) The sixth mental measurements yearbook. Highland Park, New Jersey: The Gryphon Press. 1965. Buss, A. H. A follow-up item analysis of the Taylor Anxiety Scale. J. clin. Psychol., 1955, 11, 409-410. Calvin, A. D., McGuigan, P. J., Tyrrell, Sybil, & Soyars, M. Manifest anxiety and the Palmar perspiration index. J. consult. Psychol., 1956, 20, 356. Cannon, W. B. Bodily changes in pain, hunger, fear, and rage. New York: Appleton, 1929. i 187 i Cannon, W. B. The wisdom of the body. New York: Norton, 1932. Chambers, Alma C. Possible relationship between stress and test performance. A paper prepared for i evaluation of pupil progress, Ed. Ps. 522, for Dr. Kenneth D. Hopkins, Univer. of Southern California, 1962. Clemens, T. L. Autonomic nervous system responses related to the Punkenstein Test. Psychosom. Med., 1957, 19, 267-274. ' Coleman, J. C. Abnormal psychology and modern life. j (3rd ed.) Chicago: Scott, Poresman, 1964. I Cowen, Judith E. Test anxiety in high school students and its relationship to performance on group j ; tests. Unpublished doctoral dissertation, School of Education, Harvard Univer., 1957. i j Cronbach, Lee J. Essentials of psychological testing. i i New York: Harper & Brothers, Publishers, | 1960. Dana, R. H. Manifest anxiety, intelligence, and ; psychopathology. J. consult. Psychol., 1957, | 21, 38-40. | Darrow, C. W. The Galvanic skin reflex (sweating) and blood-pressure as preparatory and facilitative functions. Psychol. Bull. , j 1936, 33, 73-94. Davidson, K. S., & Sarason, S. B. Test anxiety and classroom observations. Child Develpm., 1961, 32, 199-210. Dickel, H. A., & Dixon, H. H. Inherent dangers in use of tranquilizing drugs in anxiety states. J. A. M. A . , 1957, 422-426. Educational Testing Service. Cooperative Academic Ability Test, Form A. Princeton, N. J.: Educational Testing Service, 1963. Educational Testing Service. Cooperative Academic Ability Tests handbook. Princeton, N. J.: Educational Testing Service, 1964. Endler, N. S., Hunt, J. McV., & Rosenstein, A. J. An S-R Inventory of Anxiousness. Psychol. Monoqr., 1962, 76, No. 14 (Whole No. 536). English, H. B., & English, Ava C. A comprehensive dictionary of psychological and psycho analytical terms. New York: Longmans, Green & Co., 1958. Erickson, D. K. A comparison of measured student anxiety and recorded teacher comments. Unpublished master's thesis, Univer. of Southern California, 1963. 189 Faber, I. E., & Spence, K. W. Main and interactive effects of several variables on reaction time. Technical report 3, Studies of influence of motivation on performance in learning. Con tract N9 onr-93802, Project NR154-107, Office of Naval Research, State Univer. of Iowa, March 1, 1955, 1-24. Freud, S. Inhibitions, symptoms, and anxiety. London: Hogarth Pre s s, 1936. Fulton, J. F. A textbook of physiology. (17th ed.) Philadelphia & London: W. B. Saunders, 1955. Getzels, J. W., & Jackson, P. W. The teacher's personality and characteristics. 506-582. In Gage, N. L. (Ed.) Handbook of research on teaching. Chicago: Rand McNally & Co., 1963, 574-576. Goldstein, K. Human nature in the light of psycho pathology. Cambridge, Massachusetts: Harvard University Press, 1940. Grooms, R. G., & Endler, N. S. The effect of anxiety on academic achievement. J. educ. Psychol., 1960, 51, 299-304. i Hopkins Horney, Hoyt, D Janis, Jost, H Kissel, Lacey, 190 , K. D. Validity concomitants of various scoring procedures which attenuate the effects of response sets and change. Unpublished doctoral dissertation, Univer. of Southern California, 1961. Karen. New ways in psychoanalysis. New York: Norton, 1939. . P., & Magoon, T. M. A validation study of the Taylor Manifest Anxiety Scale. U. clin. Psychol., 1954, 10, 357, 361. [. L. Psychological stress. Psychoanalytic and behavioral studies of surgical patients. New York: Wiley, 1958. The use of polygraphic technigues in psychophysiological research and clinical psychology. 3758 West Belmont Ave., Chicago: Associated Research, Inc., 1953. S., & Littig, L. W. Test anxiety and skin conductance. J. abnorm. soc. Psychol.. 1962, 65, 276, 278. r. I., Bateman, Dorothy E., & Van Lehn, Ruth. Autonomic response specificity. Psychom. Med.. 1953, 15. 191 Lacey, J. I., Kagan, J., Lacey, Beatrice C., &Moss, H. A. The visceral level: situational determinants and behavioral correlates of autonomic response patterns. In P. H. Knapp (Ed.), Expression of the emotions in man. New York: International Universities Press, 1963 . Lacey, J. I., & Lacey, Beatrice C. The law of initial value in the longitudinal study of autonomic constitution: Reproducibility of autonomic responses and response patterns over a four year interval. Ann. N. Y. Acad. Sci., 1962, 98, 1257-1290,*-1322-1326. Lewinsohn, P. M. Some individual differences in physiological reactivity to stress. J. comp. physiol. Psychol.. 1956, 49, 271-277. Lucio, W. H., & Wenger, M. A. Prediction of teacher performance and emotional stability; a psycho- physiological pilot study of female student teachers. Final report of contract No. SAE 8311. United States Office of Education, Department of Health, Education, and Welfare through the University of California at Los Angeles, September, 1961. 192 Luria, A. R. The nature of human conflicts. (trans lated by W. H. Gantt). New York: Liverright, 1932. Mac Bryde, C. M. (Ed.) The analysis and interpreta tion of symptoms. Philadelphia: Lippincott, 1944. Mandler, G., & Sarason, S. B. A study of anxiety and learning. J. abnorm. soc. Psychol.. 1952, 47, 166-1.73. Marmor, J. Anxiety and worry as aspects of normal j | behavior. Calif. Med., 1962, 97, 212-215. ! | Martin, B. The assessment of anxiety by physiological | behavioral measures. Psychol. Bull., 1961, 58, 234-255. May, R. The meaning of anxiety. New York: Ronald Press, 1950. Milliken, R. L. Mathematical-verbal ability dif- ! | ferentials of situational anxiety as measured by blood pressure change. J. exp. Educ., 1964, 32, 309-311. Milliken, R. L., & Spilka, B. Mathematical-verbal ability differentials and somatic expressions of situational anxiety. J. exp. Educ., 1962, 31, 313-326. 193 Mowrer, 0. . H. Learning theory and personality dynamics. New Yorks Ronald Press, 1950. Paul, G. L., & Eriksen, C. W. Effects of test anxiety on "real-life" examinations. J. Pers., 1964 i j 32, 480-494. i ] Pavlov, I. P. Lectures on conditioned reflexes twenty-five years of objective study of the higher nervous activity (behaviour! of I animals vol. 1. New Yorks International Publishers, 1928. jRogers, T. A. Elementary human physiology. New Yorks Wiley, 1961. Ruebush, B„ K. Anxiety. In H. W. Stevenson (Ed.), Child psychology. The sixty-second yearbook \ i of the National Society for the Study of Education. Chicagos Univer. Chicago Press, 1963. Rush, F. L. Psychology and life. (7th ed.) Palo Altos Scott, Foresman & Co., 1967. Sarason, I. G. Test anxiety, general anxiety and intellectual performance. J. consult. Psychol., 1957, 21, 485-490. I j Sarason, I. G. Intellectual and personality corre- ! I lates of test anxiety. J. abnorm. soc. | Psychol., 1959, 59, 272-275. 194 Sarason, I. G. Empirical findings and theoretical problems in the use of anxiety scales, Psvchol. Bull.. 1960, 57, 403-415. Sarason, I. G. Characteristics of three measures of anxiety. J. clin. Psvchol.. 1961, 17, 196- 197. (a) Sarason, I. G. Test anxiety and the intellectual per formance of college students. J. educ. Psvchol., 1961, 62, 201-206. (b) Sarason, I. G. Critique and notes test anxiety and intellectual performance. J. abnorm. soc. Psvchol., 1963, 66, 73-75. Sarason, I. G., & Minard, J. Test anxiety, experi mental instructions, and the Wechsler Adult Intelligence scale. J. educ. Psvchol., 1962, 53, 299-302. Sarason, I, G., & Palola, E. G. The relationship of test and general anxiety, difficulty of task, and experimental instructions to performance. J. exp. Psvchol., 1960, 59, 185-191. Sarason, S. B. Classroom observations of high- and low-anxious children. Child Develpm., 1958, 29, 287-295. 195 Sarason, S„ B„, Davidson, K. S., Lighthall, F. F., Waite, R0 R., & Ruebush, B„ K. Anxiety in elementary school children. New York: Wiley, 1960» Sarason, S. B., & Mandler, G. Some correlates of test anxiety. J. abnorm. soc. Psychol., 1952, 47, 810-817. Schacter, J. Pain, fear, and anger in hypertensives. Psychosom. Med», 1957, 19, 17-29. Science Research Associates. Iowa Test of Educational Development. Chicago: Science Research Associates, Inc., 1962. Selye, H. The stress of life. New York: McGraw-Hill Book Company, Inc., 1956. Siegel, S. Nonparametric statistics for the behavioral sciences. New York: McGraw-Hill, 1956. Silverman, J. J., & Powell, V. E. Studies on palmar sweating. I A technique for the study of palmar sweating. Amer. J. med. Sci., 1944, 208, 297-305. Silverstein, A. B. Test anxiety and the primary mental abilities. Psychol. Reps., 1961, 8, 415-417. Smith, D. B. D., & Wenger, M. A. Changes in autonomic balance during phasic anxiety. Psychophysiol., 1965, 1, 267-271. 196 Smith, G. W. A study of the effects of manifest anxiety and induced psychological stress on college performance. An abstract in Dissertation Abstracts, abstracts of disserta tions and monographs in microfilm. Ann Arbor, Michigan: University Microfilm, 1964-1965, 25, 5753. Spence, K. W. Behavior theory and conditioning. New Haven: Yale Univer. Press, 1956. Stevens, S. S. (Ed.) Handbook of experimental psychology. New York: John Wiley & Sons, Inc., 1951. Suinn, R. M. Anxiety and intellectual performance: a partial failure to replicate. J. consult. Psychol., 1965, 29, 81-82. Taylor, Janet A. The relationship of anxiety to the conditioned eyelid response. J. exp. Psychol., 1951, 41, 81-92. Taylor, Janet A. A personality scale of manifest anxiety. J. abnorm. soc. Psychol., 1953, 48, 285-290. Thorndike, R. L., and Hagen, Elizabeth. Measurement and evaluation in psychology and education. New York: John Wiley & Sons, Inc., 1961. Waite, Walter, Wenger, Wenger, Wenger, Windle , Winter, 197 R. R., Sarason, S. B., Lighthall, F. F., & Davidson, K. S. A study of anxiety and learning in children. J. abnorm. soc. Psychol., 1958, 57, 267-270. D., Denzler, L. S., & Sarason, I. G. Anxiety and the intellectual performance of high school students. Child Develpm., 1964, 35, 917-926. M. A. Pattern analyses of autonomic variables during rest. Psvchosom. Med., 1957, 19, 240-244. M. A., Clemens, T. L., Coleman, D. R., Cullen, T. D., & Engel, B. T. Autonomic response specificity. Psvchosom. Med., 1961, 23, 185- 193. M. A., & Ellington, Margaret. The measurement of autonomic balance in children: method and normative data. Psvchosom. Med., 1943, 5, 241-253. C. Test-retest effects on personality question naires. Educ. psvchol. Measmt. 1954, 14, 617- 633. W. D., Ferreira, A. J., & Ransom, R. Two measures of anxiety: a validation. J. consult. Psvchol., 1963, 27, 520-524. ! 198 i ; j Wolf, S., & Wolff, H„ Go Human gastric function. New York: Oxford Univer. Press, 1943. Wynn, F. G. The psychic factor as an element in temperature disturbance. J. Amer. med. Ass., 1919, 73 , 31-33. Zuckerman, M. The development of an affect adjective check list for the measurement of anxiety. J. consult. Psvchol., 1960, 24, 457-462. Zuckerman, M., & Biase, D. V. Replication and further data on the validity of the Affect Adjective Check List measure of anxiety. J. consult. Psvchol., 1962, 26, 291. APPENDICES APPENDIX A LETTER FOR APPROVAL TO THE PARENTS COOPERATIVE ACADEMIC ABILITY TEST THE STUDENT SURVEY PARTS I, IT, III, IV TREATMENT INSTRUCTIONS 200 TORRANCE UNIFIED SCHOOL DISTRICT West High School 20401 Victor Torrance, California November 12, 1964 Dear Parents: jWest High School is participating in a research program which |is sponsored by the University of Southern California; Loma iLinda University and the Department of Health, Education, and ! Welfare. The purpose of the study is to assist in the devel opment of more valid, reliable achievement and intelligence (tests by determining the possible effect of certain factors ion test scores. While the student is taking an educational test, certain physiological measures, temperature, pulse rate, blood pres- i sure, perspiration, and respiration will be recorded to de termine whether they are related to success on the test. The entire amount of student time involved will be one class !period. |Your student has been chosen by random sampling to partici- I pate in the project. If you approve of his/her participa-. | tion please indicate by placing your signature below. Dr. Kenneth D. Hopkins Parent Signature Alma C. Chambers r. Robert jPord Approved: 201 202 COOPERATIVE ACADEMIC ABILITY TEST Due to the limited time available for performance on the test the middle question of each three was elimi nated. This procedure reduced the length of the test by one tbird. The remaining questions were placed in order at random by use of a random table of numbers (Kendall & Smith, 1938). This arrangement provided opportunity for individuals unable to complete the test in the time pro vided to attempt a proportion of the more difficult questions usually placed at the end of a test. The following list provides the numbers of the questions and the arrangement used for the verbal section and the same for the mathematical section. Questions 43 28 27 21 40 49 13 45 12 15 36 1 33 18 3 7 6 22 34 48 31 10 42 16 46 9 37 30 19 25 24 4 39 203 INSTRUCTIONS FOR THE STUDENT SURVEY j I We are interested in knpwing how students feel while taking an important examination. The best way we know of to find this out is to ask you how you feel on such an occasion. This is a general survey and not especially related to how you feel towjard any course in particular. Although we are not especially interested in the response of each person individually, we still would like j your true attitude. One of the main reasons for conduct ing this survey is the fact that very little is known about people's feelings toward the taking of various kinds of tests. We assume that people differ in the degree to which they are affected by the fact that they I are going to take a test, or by the fact that they have I taken a test. What we are particularly interested in i ! here is how people differ in their opinions and reactions i to the various kinds of testing situations. The value ; of this survey will depend in a large part on how frank you are in stating your opinions, feelings, and atti tudes. Read each statement carefully and-make your immediate response. Do not ponder over these state ments. Your response to this survey will be held strictly confidential. Great care will be faken in the use of the information received. Your response will not 204 in any way affect your grade in this course or any other course. STUDENT SURVEY Please do not write or mark on this booklet in any way. Your answers to the statements in this in ventory are to be recorded only on the separate ANSWER j SHEET. Remember to give YOUR OWN opinions. Please do not leave any question unanswered. Make your marks heavy and black. Erase completely any answer you wish jto change. Please notice the choices on the ANSWER SHEET read from left to right. This inventory represents a means of studying people's reactions during an important examination. Certain common types of personal reactions and feelings to examinations are listed. Indicate choices on the ANSWER SHEET, which represent the degree to which you would show certain reactions and feelings during an important examination. Here is an example: You are taking an important examination. Heart beats faster 1 2 3-- 4--- 5--— much not at all faster 205 If your heart beats much faster in this situation, you would darken alternative 5 on the ANSWER SHEET? if your heart beats somewhat faster, you would darken either alternative 2, 3, or 4, depending on how much faster? if in this situation your heart does not beat faster at all, you would darken alternative 1 on the ANSWER SHEET. If you have no questions, you may begin. 206 PART I "You are Taking an Important Examination" PLEASE DO NOT MARK THIS BOOKLET Mark on the ANSWER.SHEET one of the five alternative degrees of reactions or attitude for each of the following 14 items. 1. Heart beats faster 1 2 Not at all 2 . Get an "uneasy feeling" 1 2 None 3. Emotions disturb action 4. Feel excited and thrilled 5. Want to avoid situation 6. Perspire 7. Need to urinate frequently 1 2 Not at all 1 2 Very much 1 2 Not at all 1 2 Not at all 1 2 Not at all 4 5 Much faster 4 5 Very strongly 4 5 Very disturbed 4 5 Not at all 4 5 Very much 4 5 Perspire much 4 5 Very frequently 207. 8. 9. 10. 11. 12. 13. 14. Enjoy the challenge Mouth gets dry Become unable to move Get full feeling in stomach Seek experiences like this Have loose bowels Experience nausea 1 2 3 Enjoy much 1 2 3 Not at all 1 2 3 Not at all 1 2 3 None 1 2 3 Very much 1 2 3 None 1 2 3 Not at all 4 5 Not at all 4 5 Very dry 4 5 Completely 4 5 Very full 4 5 Not at all ; 4 5 Very much 4 5 Much nausea , 208 PART II For Parts II, III, and IV, there are only two choices, true or false, to each question* Below you will find words which describe differ- . ent kinds of feelings. Please mark "true" by darkening alternative "1" on the ANSWER SHEET for the words that describe how you feel while taking an important exami nation. Mark "false" by darkening alternative "2" on the ANSWER SHEET for the words that do not describe your feelings while taking an important examination. Some of the words sound alike but we want your response to all the words indicating whether they do or do not describe your feeling while taking an important examination by darkening either alternative 1 or 2 on the ANSWER SHEET. Remember; True = 1, False = 2. Afraid If this word is true or usually true, as applied to you,, darken the first alternative. (True) (False) 1 2. 3 4 5 If the word is false or not usually true, as applied to you, darken the second alternative. 209 15. afraid 39. helpless 63. shaky 16. agitated 40. hopeless 64. solemn 17. angry 41. insecure 65. steady 18. bitter 42. jealous 66. tender 19. calm 43. joyful 67. tense 20. charming 44. kindly 68. terrified 21. cheerful 45. light-hearted 69. threatened 22. complaining 46. lonely 70. thoughtful 23. contented 47. loving 71. unconcerned 24. contrary 48. mad 72. uneasy 25. cool 49. mean 73. upset 26. cross 50. merry 74. warm 27. desperate 51. miserable 75. worrying 28. easy-going 52. nervous 29. fearful 53. overconcerned 30. fearless 54. overwhelmed 31. fretful 55. panicky 32. friendly 56. peaceful 33. frightened 57. pleasant 34. furious 58. rattled 35. gay 59. sad 36. gloomy 60. secure 37. grim 61. sentimental 38. happy 62. serious | 2.10 PART III Continue to answer the statements in Part III by j giving your first reaction. Mark "I" for true and "2" for "false" as you did in Part II. 76. While taking an important examination, I perspire j : I a great deal. • 77. I get to feel very panicky when I have to take a surprise exam. 78. During tests, I find myself thinking of the i consequences of failing. 79. After important tests, I am frequently so tense that my stomach gets upset. 80. While taking an important exam, I find myself thinking of how much brighter the other students are than I am. ! 81. I freeze up on things like intelligence tests and ; final exams. 82. If I were to take an intelligence test, I would j worry a great deal before taking it. 83. During course examinations, I find myself thinking j ! of things unrelated to the actual course material. i ; 84. During a course examination, I frequently get so nervous, that I forget facts I really know. 85. If I knew I was going to take an intelligence test, I would feel confident and relaxed before hand. 211 86. I usually get depressed after taking a test. 87. I have an uneasy, upset feeling before taking a ' final examination. 88. When taking a test, my emotional feelings do not interfere with my performance. ! 89. Getting a good grade on one test doesn't seem to ■ increase my confidence on the second. ] j 90. After taking a test, I always feel I could have done better than I actually did. 91. I sometimes feel my heart beating very fast during important tests. PART IV 212 The following questions are concerned with your feelings toward general experiences in life, not just on examinations. Mark "1" for "true'' and "’ 2" for "false" as before. 92. I believe I am no more nervous than most others. 93. I work under a great deal of tension. 94. I cannot keep my mind on one thing. 95. I am more sensitive than most other people. 96. I frequently find myself worrying about something. 97. I am usually calm and not easily upset. 98. I feel anxiety about something or someone almost all the time. 99. I am happy most of the time. 100. I have periods of such great restlessness that I cannot sit long in a chair. 101. I have sometimes felt that difficulties were piling up so high that I could not overcome them. 102. I certainly feel useless at times. 103. I find it hard to keep my mind on a task or job. 104. I am not unusually self-conscious. 105. I am inclined to take things hard. 106. I am a high-strung person. 107. Life is a strain for me much of the time. 213 108. At times X think I am no good at all. 109. I am certainly lacking in self-confidence. 110. I sometimes feel that I am about to go to pieces. 111. I shrink from facing a crisis or difficulty. 214 TREATMENT INSTRUCTIONS Anxiety-Reducing Instructions We are conducting a research project concerning examinations. The test you are about to take is very long and difficult so do not be concerned if you are not sure of all the answers or cannot finish within the given time. No one will find it easy. We wish to make it clear that this is purely for research purposes and will have nothing to do with any course offered at West High. Attempt the questions in the order they come— do not skip around. We wish to thank you for your cooperation in this research. Please follow the General Directions on the front of the test booklet and on page three. Neutral Instructions The test you are about to take is an Academic Ability Test. Attempt the questions in the order they come— do not skip around. Please follow the General Directions on the front of the test booklet and on page three. 215 Anxiety-Producing Instructions The test you are about to take is a college ;ability intelligence test. This test has been found ;to predict such things as course grades, success or !failure in college, success or failure in various types I of occupations, and success or failure in later life, iOf course, your own intelligence will primarily determine I whether you do well or poorly on this test. Therefore, j it is an extremely important examination and you should i I try to do your very best. i At a later date the scores will be posted so ;your score can be compared with others who have taken :the test. ; Your teachers and counselors will be given your results so be sure to try your very best on all questions, iAttempt the questions in the order they come— do not i !skip around. Remember it is important for you to concentrate at all times. ! Please follow the General Directions on the |front of the test booklet and on page three. APPENDIX B TABLES OF THE ANALYSIS OF THE DATA 216 217 TABLE 22 Multiple Correlation for Physiological Measures Used as Predictors of Ability Test Scores Variables Mean Standard Deviation Regression Coefficient Resp. Rate 18.7 3.2 0.28 EKG 89.3 68.7 -0.16 GSR 743.6 5158.6 0.00 Change Resp. Rate -.5 2.0 -0.48 Change EKG 1.5 5.7 0.27 Change GSR -5.9 15.3 -0.03 Systolic B. P. 111.2 10.5 0.01 Diastolic B. P. 67.8 6.9 -0.24 Oral Temp. 99.0 0.4 -4.55 Face Temp. 95.8 1.4 0.31 Finger Temp. 85.9 6.8 -0.45 Change Oral Temp. 0.7 0.5 -1.12 Change Face Temp. 1.5 1.1 i —i . O 1 Change Finger Temp. 2.5 3.6 0.52 Ability Test 25.8 9.6 Coefficient of.Determination 0.24 Multiple Correlation Coefficient 0.49 Multiple Correlation Coefficient after Correction for Shrinkage 0.08 N.S. n = 60 218 TABLE 23 Results of the Analysis of Variance for the Date the Letter of Approval was Received Source of Variation df Mean Square F A-Instruction 3 306.53 0.69 B-Profic iency 1 0.62 0.00 C-Sex 1 900.59 2.04 A X B 3 314.66 0.71 A X C 3 677.34 1.53 B X C 1 26.22 0.06 A X B X C 3 633.66 1.43 Within 63 442.21 — Total 78 ! 3 219 TABLE 24 Results of the Analysis of Variance for the Period of the Day for Testing the Students Source of Variation df Mean Square F A-Instruction 3 6.09 1.25 B-Proficiency 1 0.15 0.03 C-Sex 1 0.25 0.05 A X B 3 0.78 0.16 A X C 3 3.94 0.81 B X C 1 0.25 0.05 A X B X C 3 1.28 0.26 Within 64 4.87 — Total 79 220 SCORES . MATHEMATICAL SCORE (F=l.20) 24 VERBAL SCORE (F=.67) 22 TOTAL SCORE (F=.66) 20 18 16 14 12 10 8 6 ANXIETY- NEUTRAL ANXIETY- CONTROL REDUCING PRODUCING INSTRUCTIONS Fig. 6. Means, for the total, verbal, and mathematical scores on the Academic Ability Test for the anxiety-reducing, neutral, and anxiety-producing instruction groups, and for the control group for test performance. 221 BOYS 34 GIRLS 32 30 28 26 24 22 20 18 16 | ANXIETY- NEUTRAL ANXIETY- CONTROL s REDUCING PRODUCING ! INSTRUCTIONS i j Fig. 7* Means for the total test scores on the Academic Ability Test for the anxiety-reducing, neutral, and anxiety-producing instruction groups and the control for test performance. (Treatment F = .66 NS |Sex F = .01 NS 222 BOYS 24 GIRLS 22 20 18 16 14 12 10 NEUTRAL ANXIETY- CONTROL ANXIETY- REDUCING PRODUCING j INSTRUCTIONS i Fig. 8. Means for the verbal scores on the Academic Ability j Test for the anxiety-reducing, neutral, and anxiety-producing I instruction groups and the control for test performance for boys | and girls. i [ Treatment F - .67 NS | | Sex F = 6.71 P < -05 1_ 223 BOYS 24 GIRLS 22 20 18 16 14 12 10 ANXIETY- NEUTRAL ANXIETY- CONTROL REDUCING PRODUCING INSTRUCTIONS Fig. 9- Means for the mathematical scores on the Academic Ability Test for the anxiety-reducing, neutral, and anxiety- producing instruction groups and the control for test performance. Treatment F = 1.20 NS Sex F = 8 .9 4 p < .01 224 TABLE 25 Results of the Analysis of Variance for the Day of Testing for Each Student Source of Variation df Mean Square P A-Instruction 3 10.82 1.96 B-Proficiency 1 0.74 0.14 C-Sex 1 10.96 1.98 A X B 3 1.66 0.30 A X C 3 7.37 1.33 B X C 1 0.00 0.00 A X B X C 3 15.10 2.73 Within 64 5.52 — Total 79 I ! TABLE 26 Means and Standard Deviations for the Total, Verbal and Mathematical Scores on the Academic Ability Test Means and Standard Deviations for Total Score Instruction Groups Boys Girls Mean S.D. Mean S.D. Total Mean S.D. Anxiety-Reducing 28.1 9.5 24.5 * 00 26.3 9.2 Neutral 25.6 8.6 23.4 9.8 24.5 9.2 Anxiety-Producing 23.7 10.8 29.5 10.9 26.6 10.9 Total 25.8 9.6 25.8 9.9 25.8 9.7 Control for Test Performance 26.5 8.5 24.9 8.4 25.7 8.4 Total 26.0 9.3 25.6 9.4 25.8 9.4 225 TABLE 26— (Continued) Means and Standard Deviations for Verbal Score Instruction Groups Boys Girls Mean S.D. Mean S.D. Total Mean S.D. Anxiety-Reducing 12.7 5.3 14.0 5.0 13.4 5.1 Neutral 13.0 4.7 13.2 5.1 13.1 4.9 Anxiety-Producing 12.1 5.2 16.3 6.3 14.2 5.8 Total 12.6 5.1 14.5 5.4 13.6 5.3 Control for Test Performance 11.5 2.6 13.9 5.5 12.2 4.0 Total 12.3 4.4 14.4 5.5 13.3 5.0 226 TABLE 26— (Continued) Means and Standard Deviations for Mathematical Score I Instruction Boys Girls Total ; Groups Mean S.D. Mean S.D. Mean S.D. Anxiety-Reducing 16.2 4.5 10.5 4.3 13.4 4.4 Neutral 12.6 4.8 10.2 5.0 11.4 4.9 Anxiety-Producing 12.2 6.6 13.2 5.4 12.7 6.0 Total 13.7 5.3 11.3 4.9 12.5 5.1 Control for Test Performance 15.0 6.6 11.0 3.6 13.0 5.1 Total 14.0 5.4 11.2 4.6 12.6 5.1 n = 10 227 228 TABLE 27 Results of the Analysis of Variance for the Pretest Systolic Blood Pressure for the Three Instruction Groups and the Control Group for Physiological Measures Source of Variation df Mean Square F A-Instruction 3 5.15 0.04 B-Proficiency 1 41.64 0.36 C-Sex 1 559.28 4.81 A X B 3 209.64 1.80 A X C 3 253.76 2.18 B X C 1 112.89 0.97 A X B X C 3 84.64 0.73 Within 62 116.28 — Total 77 Significant at .05 level aBoys Girls 229 TABLE 28 Results of the Analysis of Variance for the Pretest Diastolic Blood Pressure for the Three Instruction Groups and the Control for Physiological Measures Source of Variation df Mean Square F A-Instruction 3 84.97 1.78 B-Pro ficiency 1 181.75 3.80 C-Sex 1 2.00 0.04 A X B 3 25.90 0.54 A X C 3 20.87 0.44 B X C 1 3.10 0.06 A X B X C 3 25.52 0.53 Within 62 47.80 Total 77 230 TABLE 29 Results of the Analysis of Variance for the Initial Pulse Pressure for the Three Instruction Groups and the Control Group for Physiological Measures Source of Variation df Mean Square P A-Instruction 3 125.69 0.91 B-Proficiency 1 30.35 0.22 C-Sex 1 686.30 4.99*a A X B 3 212.27 1.54 A X C 3 244.01 1.77 B X C ■1 112.90 0.82 A X B X C 3 72.91 0.09 Within 64 137.63 — Total 79 *Significant at .01 level aBoys > Girls 231 TABLE 30 Results of the Analysis of Variance for Pretest Oral Temperature for the Three Instruction Groups and the Control Groups for Physiological Measures Source of Variation df Mean Square F A-Instruction 3 0.56 1.64 B-Proficiency 1 0.88 2.57 G-Sex 1 2.95 8.59**a A X B 3 0.03 0.08 A X C 3 1.14 3.31* B X C 1 0.03 0.09 A X B X C 3 0.43 1.26 Within 64 0.34 — Total 79 *Significant at .05 level **Significant at .01 level aGirls > Boys 232 TABLE 31 Results of the Analysis of Variance for the Initial Face Temperature for the Three Instruction Groups and the Control Group for Physiological Measures Source of Variation df Mean Square F A-Instruction 3 3.16 0 .86 B-Proficiency 1 0 .5 4 0 * 15 C-Sex 1 4 0 .6 6 H .1 0 * * a A X B 3 3.42 0.17 A X C 3 0.07 0.93 B X C 1 0 .2 0 0.06 A X B X C 3 2.33 0 .6 4 Within 64 3 .6 6 — Total 79 **Significant at .01 level aBoys > Girls 233 TABLE 32 Results of the Analysis of Variance for the Pretest Finger Temperature for the Three Instruction Groups and the Control Group for Physiological Measures Source of Variation df Mean Square F A-Instruction 3 18.84 0.56 B-Proficiency 1 3.52 0.10 C-Sex 1 173.60 5.13 A X B 3 49.32 1.46 A X C 3 27.63 0.82 B X C 1 49.01 1.45 A X B X C 3 11.41 0.34 Within 64 33.85 Total 79 *Significant at .05 level aBoys Girls 234 TABLE 33 Results of the Analysis of Covariance for the Total Test Score with the Initial Respiration Rate Held Constant Source of Variation df Mean Square P A-Instruction 2 50.15 1.21 B-Proficiency 1 3047.72 73.34** G-Sex 1 2.15 0.05 A X B 2 15.55 0.37 A X C 2 172.20 4.14^ B X C 1 77.24 1.86 A X B X C 2 46.54 1.12 Within 47 41.55 Total 58 ♦Significant at .05 level ♦♦Significant at .01 level TABLE 34 235 Results of the Analysis of Covariance for the Verbal Score With the Initial Respiration Rate Held Constant Source of Variation df Mean Square F A-Instruction 2 12.40 1.00 B-Proficiency 1 877.20 10.51** C-Sex 1 63.86 5.14#a A X B 2 5.52 0.44 A X C 2 28.72 2.31 B X C 1 17.78 1.43 A X B X C 2 10.11 0.81 Within 47 12.43 — Total 58 ♦Significant at .05 level ♦♦Significant at .01 level aGirls y Boys 236 TABLE 35 Results of the Analysis of Covariance for the Mathematical Score with the Initial Respiration Rate Held Constant Source of Variation df Mean Square F A-Instruction 2 21.00 1.29 B-Proficiency 1 615.91 37.97** C-Sex 1 56.29 3.47 A X B 2 8.44 0.52 A X C 2 57.63 3.55* B X C 1 15.98 0.98 A X B X C 2 8.18 0.50 Within 47 16.22 — Total 58 ♦Significant at .05 level ♦♦Significant at *01 level 237 TABLE 36 Results of Analysis of Covariance for the Total Test Scores with Differences Between Pretest Reading . and Test Reading Respiration Depth Held Constant Source of Variation df Mean Square F A-Instruction 2 48.38 1.16 B-Proficiency 1 3015.40 72.59 C-Sex 1 8.32 .20 A X B 2 16.61 0.40 A X C 2 165.82 3.99 B X C 1 68.42 1.65 A X B X C 2 43.19 1.04 Within 47 41.54 — Total 58 ^Significant at .05 level **Significant at .01 level 238 TABLE 37 Results of Analysis of Covariance for the Verbal Scores with Differences Between the Pretest Reading and Test Reading Respiration Depth Held Constant Source of Variation df Mean Square F A-Instruction 2 13.76 1.11 B-Proficiency 1 862.01 69.77** C-Sex 1 69.91 5.66*a A X B 2 5.77 .47 A X C 2 28.87 2.34 B X C 1 16.75 1.36 A X B X C 2 10.24 .83 Within 47 12.35 — Total 58 1 *Significant at .05 level **Significant at .01 level aGirls Boys 239 TABLE 38 Results of Analysis of Covariance for the Mathematical Scores with Differences Between Pretest Reading and Test Reading Respiration Depth Held Constant Source of Variation df Mean Square F A-Instruction 2 21.57 1.33 B-Proficiency 1 613.22 37.75** C-Sex 1 60.29 3.71 A X B 2 8.82 .54 A X C 2 58.12 3.58* B X C 1 15.09 .93 A X B X C 2 8.06 .05 Within 47 16.24 — Total 58 *Significant at .05 level Significant at .01 level 240 TABLE 39 |Results of the Analysis of Covariance for the Total Test Score with the Initial Heartbeat Rate Held Constant Source of Variation df Mean Square F A-Instruction 2 47.75 1.16 B-Proficiency 1 3080.94 74.59^ C-Sex 1 12.23 0.30 A X B 2 20.26 0.49 A X C 2 171.12 4.14* B X C 1 73.12 1.77 A X B X C 2 35.68 0.86 Within 47 41.30 — Total 58 ♦Significant at .05 level ♦♦Significant at .01 level TABLE 40 241 Results of the Analysis of Covariance for the Verbal Score with the Initial Heartbeat Rate Held Constant Source of Variation df Mean Square P A-Instruction 2 14.20 1.14 B-Proficiency 1 863.73 69.61** C-Sex 1 67.57 5.44*a A X B 2 4.78 0.38 A X C 2 29.96 2.41 B X C 1 18.48 1.49 A X B X C 2 11.32 0.91 Within 47 12.41 — Total 58 *Significant at .05 level **Significant at .01 level aGirls > Boys TABLE 41 242. Results of the Analysis of Covariance for the Mathematical Score with the Initial 'Heartbeat' Rate Held Constant Source of Variation df Mean Square P A-Instruction 2 25.58 1.66 B-Proficiency 1 640.45 41.56** C-Sex 1 48.52 3.15 A X B 2 13.31 0.86 A X C 2 58.67 3.81* B X C 1 15.09 0.98 A X B X C 2 3.67 0.24 Within 47 15.41 — Total 58 *Significant at .05 level **Significant at .01 level 243 TABLE 42 Results of Analysis of Covariance for the Total Test Scores with Differences Between Pretest Reading and Test Reading GSR Held Constant Source of Variation df Mean Square P A-Instruction 2 38.16 <N • B-Proficiency 1 3066.90 73.96 C-Sex 1 13.52 .32 A X B 2 14.22 .34 A X C 2 169.34 4.08 B X C 1 69.84 1.68 A X B X C 2 42.49 1.02 Within 47 41.47 — Total 58 *Significant at .05 level **Significant at .01 level 244 TABLE 43 Results of Analysis of Covariance for the Verbal Scores with Differences Between Pretest Reading and Test Reading GSR Held Constant Source of Variation df Mean Square F A-Instruction 2 12.63 1.02 B-Proficiency 1 876.59 70.46 C-Sex 1 68.88 5.54' A X B 2 5.45 .44 A X C 2 30.68 2.47 B X C 1 18.21 1.46 A X B X C 2 10.47 0.84 Within 47 12.44 — Total 58 *Significant **Significant at at .05 level .01 level aGirls .> Boys I 245 TABLE 44 i j Results of the Analysis of Covariance for the Mathematical Scores with Differences Between Pretest Reading and Test Reading GSR Held Constant Source of Variation df Mean Square F A-Instruction 2 18.35 1.16 B-Proficiency 1 621.81 39.16 C-Sex 1 45.05 2.84 A X B 2 6.63 .42 A X C 2 58.72 3.70' B X C 1 13.53 .85 A X B X C 2 10.47 .62 Within 47 15.88 — Total 58 *Significant at .05 level **Significant at .01 level 246 TAB&E 45 Results of the Analyses of Covariance fpr thp Total Test Score with the Initial Sys^plic Blood Pressure Held Constant Source of Variation df Mean Square 1 i • p A-Instruction 2 41.28 0.97 B-Proficiency 1 2856.37 67.20** C-Sex 1 26.54 0.62 A X B 2 14.41 0.34 A X C 2 183.08 4.31* B X C 1 74.77 1,76 A X B X C 2 41.31 0.97 Within 45 42.50 — Total 56 ^Significant at ,05 level ♦^Significant at .01 level 247 TABLE 46 Results of the Analysis of Covariance for the Verbal Score with Initial Systolic Blood Pressure Held Constant Source of Variation df Mean Square F A-Instruction 2 12.62 1.00 B-Proficiency 1 819.90 64.84^ C-Sex 1 85.43 6.76^a A X B 2 4,55 0.36 A X C 2 33.44 2.64 B X C 1 18.98 1.50 A X B X C 2 9.38 0.74 Within 45 12.64 — Total 56 ♦Significant at .05 level ♦♦Significant at .01 level aGirls Boys 248 TABLE 47 Results of the Analysis of Covariance for the Mathematical Score with the Initial Systolic Blood Pressure Held Constant Source of Variation df Mean Square F A-Instruction 2 19.08 1.15 B-Proficiency 1 572.99 34.52** C-Sex 1 35.66 2.15 A X B 2 8.36 0.50 A X C 2 64.23 3.87* B X C 1 16.00 0.96 A X B X C 2 9.09 0^55 Within 45 16.60 amm m Total 56 ^Significant at .05 level **Significant at .01 level 249 TABLE 48 Results of the Analysis of Covariance for the Total Test Score with the Initial Diastolic Blood Pressure Held Constant Source of Variation df Mean Square F A-Instruction 2 20.82 0.53 B-Proficiency 1 2406.16 61. 38^ C-Sex 1 18.32 0.47 A X B 2 17.11 0.44 A X C 2 166.62 4.25^ B X C 1 74.45 1.90 A X B X C 2 31.47 0.80 Within 45 39.20 Total 56 ♦Significant at .05 level ♦♦Significant at .01 level 250 TABLE 49 Results of the Analysis of Covariance for the Verbal Score with the Initial Diastolic Blood Pressure Held Constant Source of Variation df Mean Square F A-Instruction 2 2.70 0.24 B-Proficiency 1 672.02 60.32** C-Sex 1 85.27 7.65**a A X B 2 4.00 0.36 A X C 2 26.82 2.41 B X C 1 19.31 1.73 A X B X C 2 7.24 0.65 Within 45 11.14 — Total 56 **Significant at .01 level aGirls Boys 251 TABLE 50 Results of the Analysis of Covariance for the Mathematical Score with the Initial Diastolic Blood Pressure Held Constant Source of Variation df Mean Square P A-Instruction 2 18.81 1.16 B-pro fic iency 1 488.39 30.15^ G-Sex 1 51.50 3.18 A X B 2 11.25 0.69 A X C 2 60.98 3.76* B X C 1 15.25 0.94 A X B X C 2 5.09 0.31 Within 45 16.20 Total 56 ♦Significant at .05 level ♦♦Significant at .01 level 252 TABLE 51 Results of the Analysis of Covariance for the Total Score with the Initial Pulse Pressure Held Constant Source of Variation df Mean Square F A-Instruction 2 26.80 0.71 B-Proficiency 1 2743.48 72.62♦♦ C-Sex 1 52.50 1.38 A X B 2 25.04 0.66 A X C 2 221.89 5.87^ B X C 1 97.14 2.57 A X B X C 2 51.59 1.36 Within 47 37.78 — Total 58 ♦Significant at .05 level ♦♦Significant at .01 level 253 TABLE 52 Results of the Analysis of Covariance for the Verbal Score with the Initial Pulse Pressure Held Constant Source of Variation df Mean Square F A-Instruction 2 6.76 0.61 B-Proficiency 1 777.96 70.60** C-Sex 1 112.69 10.22*** A X B 2 8.30 0.75 A X C 2 46.78 4.24^ B X C 1 25.59 2.32 A X B X C 2 12.80 1.16 Within 47 11.02 — Total 58 ♦Significant at .05 level ♦♦Significant at .01 level aGirls ^ Boys TABLE 53 254 Results of the Analysis of Covariance for the Mathematical Score with the Initial Pulse Pressure Held Constant Source of Variation df Mean Square F A-Instruction 2 17.38 1.14 B-Pro fic i enCy 1 549.66 35.98 C-Sex 1 26.40 1.73 A X B 2 11.86 0.78 A X C 2 70.40 4.62 B X C 1 20.85 1.38 A X B X C 2 9.73 0.64 Within 47 15.0 Total 58 • f f Significant at .05 level **Significant at .01 level 255 TABLE 54' Results of the Analysis of Covariance for the Total Test Score with the Initial Oral Temperature Held Constant Source of Variation df Mean Square F A-Instruction 2 40.69 0.98 B-Proficiency 1 2925.44 70.83** C-Sex 1 17.88 0.43 A X B 2 17.83 0.43 A X C 2 144.15 3.49* B X C 1 74.51 1.80 A X B X C 2 50.13 1.21 Within 45 41.30 — Total 56 ^Significant at .05 level **Signifleant at .01'level TABLE 55 Results of the Analysis of Covariance for the Verbal Score with the Initial Oral Termperature Held Constant Source of Variation df Mean Square F A-Instruction 2 9.11 0.76 B-Proficiency 1 819.61 68.41 C-Sex 1 89.22 7.45 A X B 2 6.41 0.54 A X C 2 20.10 1.68 B X C 1 18.86 1.57 A X B X C 2 14.33 1.20 Within 47 11.98 — Total 58 Significant at .01 level aGirls y Boys 257 TAB IE 56 Results of. the Analysis pf Covariance for the Mathematical Test Score with the Initial Oral Temperature Maid Constant Source of Variation df Meajti Squar^ P A-Instruction 2 21.24 1.31 B-Proficiency 1 598.15 36.86** C-Sex 1 50.24 3 .10 A X B 2 9.08 0.56 A X C 2 55.84 3.44* B X C 1 15.50 0.96 A X B X C 2 8.28 0.51 Within 47 16.22 — Total 58 ^Significant at .05 l$vel **Significaht at .01 level 258 TABLE 57 Results of the Analysis of Covariance for the Total Test Score with the Initial Face Temperature Held Constant ■ i i m ii i , i i i Source of Variation df Mean Square F A-Instruction . 2 48.63 1.17 B-Proficiency 1 3061.52 73.43^ C-Sex 1 11.24 0.27 A X B 2 15.49 0.37 A X C 2 162.51 3.90^ B X C 1 75.72 1.82 A X B X C 2 43.25 1.04 Within 47 41.69 — Total 58 ♦Significant at .05 level ♦♦Significant at .01 level i i TABLE 58 Results of the Analysis of Covariance for the Verbal Score with the Initial Face Temperature Held Constant Source of Variation df Mean Square F A-Instruction 2 13.63 1.10 B-Pro ficiency 1 876.33 70.45 C-Sex 1 58.10 4.67 A X B 2 5.56 0.45 A X C 2 30.73 2.47 B X C 1 17.78 1.43 A X B X C 2 10.68 0.86 Within 47 12.44 Total 58 *Significant at .05 level **Significant at .01 level aGirls y Boys 260 ■ i TABLE 59 Results of the Analysis of Govariance for the Mathematical Score with the Initial Face Temperature Held Constant Source of Variation df Mean Square F A-Instruction 2 22.14 1.36 B-Profici ency 1 618.47 38.13** C-Sex 1 45.36 2.80 A X B 2 8.44 0.52 A X C 2 53.44 3.29* B X C 1 16.21 1.00 A X B X C 2 7.52 0.46 Within 47 16.22 M M Total 58 *Significant at .05 level **Significant at .01 level 261 TABLE 60 Results of the Analysis of Covariance for the Total Test Score with the Initial Finger Temperature Held Constant Source of Variation df Mean Square F A-Instruction 2 31.51 0.84 B-Proficiency 1 3026.66 80.40^ C-Sex 1 0.10 0.00 A X B 2 23.24 0.62 A X C 2 134.19 3,56* B X C 1 107.89 2.86 A X B X C 2 49.61 1.32 Within 47 37.65 — Total 58 ♦Significant at .05 level ♦♦Significant at .01 level 262 TABLE 61 Results of the Analysis of Covariance for the Verbal Test Score with the Initial Finger Temperature Held Constant Source of Variation df Mean Square F A-Instruction 2 8.45 0.76 B-Proficiency 1 866.59 77.58** C-Sex 1 46.56 4.l7*a A X B 2 5.64 0.50 A X C 2 24.37 2.18 B X C 1 28.10 2.52 A X B X C 2 12.49 1.12 Within 47 11.17 — Total 58 *Significant at .05 level **Significant at .01 level aGirls > Boys j i I 263 TABLE 62 Results of the Analysis of Covariance for the Mathematical Test Score with the Initial Finger Temperature Held Constant Source of Variation df Mean Square F A-Instruction 2 18.94 1.22 B-Proficiency 1 611.62 39.43** C-Sex 1 76.17 4.91*a A X B 2 13.22 0.85 A X C 2 48.66 3.14 B X C 1 21.98 1.42 A X B X C 2 7.33 0.47 Within 47 15.51 — Total 58 ^Significant at .05 level **Significant at .01 level aBoys ^ Girls TABLE 63. Results of the Analysis of Covariance for the Total Test Score for the Three Instruction Groups and the Control Group with the Mean EKG Held Constant Source of Variation df Mean Square F A-Instruction 3 28.72 0.68 B-Profic iency 1 2868.47 68.20** C-Sex 1 1.11 0.03 A X B 3 27.53 0.65 A X C 3 86.00 2.04 B X C 1 0.58 o.oi A X B X C 3 88.83 2.11 Within 62 42.06 — Total 77 ^^Significant at .01 level 265 TABLE 641. Results of the Analysis of Covariance for the Verbal Score for the Three Instruction Groups and the Control Group with the Mean EKG Held Constant Source of Variation df Mean Square P A-Instruction 3 16.30 1.25 B-Proficiency 1 692.23 53.06^ C-Sex 1 87.85 6.73^a A X B 3 23.42 1.79 A X C 3 11.18 0.86 B X C 1 4.73 0.36 A X B X C 3 11.36 0.87 Within 62 13.04 — Total 77 ♦Significant at ♦♦Significant at aGirls > Boys .05 level .01 level 266 TABLE 65 . Results of the Analysis of Covariance for the Mathematical Score for the Three Instruction Groups and the Control Group with the Mean EKG Held Constant Source of Variation df Mean Square P A-Instruction 3 15.58 0.98 B-Proficiency 1 730.07 45.79** C-Sex 1 86.26 5.41*a A X B 3 3.77 0.24 A X C 3 41.62 2.61 B X C 1 1.53 0.10 A X B X C 3 36.66 2.30 Within 62 15.94 M Total 77 ^Significant at .05 level **Signifleant at .01 level aBoys ^ Girls 267 TABLE 66 Results of the Analysis of Covariance for the Total Test Score for the Three Instruction Groups and the Control Group with the Mean Systolic Blood Pressure Held Constant Source of Variation df Mean Square P A-Instruction 3 24.00 0.57 B-Proficiency 1 3169.82 75.29** G-Sex 1 3.. 76 0.09 A X B 3 36.04 0.86 A X C 3 87.74 2.08 B X C 1 4.10 0.10 A X B X C 3 82.67 1.96 Within 63 42.10 Total 78 **Significant at .01 level 268 TABLE 67 Results of the Analysis of Covariance for the Verbal Score for the Three Instruction Groups and the Control Group with the Mean Systolic Blood Pressure Held Constant Source of Variation df Mean Square F A-Instruction 3 7.73 0.57 B-Proficiency 1 800.84 59.22** C-Sex 1 103.17 7.63**a A X B .3 29.28 2.16 A X C 3 13.47 1.00 B X C 1 9.17 0.68 A X B X C 3 10.28 0.76 Within 63 13.52 Total 78 **Significant at .01 level aGirls > Boys 269 TABLE 68 Results of the Analysis of Covariance for the Mathematical Score for the Three Instruction Groups and the Control Group with the Mean Systolic Blood Pressure Held Constant Source of Variation df Mean Square F A-Instruction 3 17.34 1.11 B-Proficiency 1 777.60 49+76** C-Sex 1 89.38 5.72** A X B 3 3.51 0.22 A X C 3 42.16 2.70 B X C 1 0.67 0.04 A X B X C 3 38.04 2.43 Within 63 15.63 Total 78 ♦Significant at ,05 level ♦♦Significant at ,01 level aBoys > Girls 270 TABLE 69 Results of the Analysis of Covariance for the Total Test Score for the Three Instruction Groups and the Control with the Mean Diastolic Blood Pressure Held Constant Source of Variation df Mean Square F A-Instruction 3 30.90 0.75 B-Proficiency 1 3230.47 78.17^ C-Sex 1 0.44 0.01 A X B 3 17.70 0.43 A X C 3 94.94 2.30 B X C 1 0.28 0.01 A X B X C 3 93.48 2.26 Within 63 41.32 Total 78 ♦♦Significant at .01 jlevel 271 TABLE 70 Results of the Analysis of Covariance for the Verba.1 Score for the Three Instruction Groups and the Control Group with the Mean Diastolic Blood Pressure Held Constant Source of Variation df Mean Square F A-Instruction 3 8.76 0.64 B-Proficiency 1 822.11 60.54^ C-Sex 1 89.68 6.60Aa A X B 3 19.32 1.42 A X C 3 11.97 0.88 B X C 1 4.94 0.36 A X B X C 3 10.84 0.80 Within 63 13.58 Total 78 ♦Significant at .05 level ♦♦Significant at .01 level aGirls ^ Boys 272 TABLE 71 Results of the Analysis of Covariance for the Mathematical Score for the Three Instruction Groups and the Control Group with the Mean Diastolic Blood Pressure Held Constant Source of Variation df Mean Square F A-Instruction 3 2 1 . 5 7 1 . 4 2 B-Proficiency 1 7 9 0 . 2 6 52.00** C-Sex 1 140.76 9 „ 2 6 * * a A X B 3 6 . 9 0 0 . 4 5 A X C 3 4 8 . 3 9 3 . 1 8 * B X C 1 2 . 7 0 0 . 1 8 A X B X C 3 4 3 . 0 8 2 . 8 3 Within 63 1 5 . 2 0 Total 78 *Significant at .05 level **Signifleant at .01 level aBoys y Girls 273 TABLE 72 Results of the Analysis of Covariance for the Total Score for the Three Instruction Groups and the Control Group with the Mean Oral Temperature Held Constant Source of Variation df Mean Square F A-Instruction 3 36.64 0.85 B-Proficiency 1 3251.83 77.22 C-Sex 1 5.09 0.12 A X B 3 29.28 0.70 A X C 3 92.36 2.19 B X C 1 0.89 0.02 A X B X C 3 87.68 2.08 Within 63 42.11 Total 78 **Significant at .01 level 274 TABLE 73 Results of the Analysis of Covariance for the Verbal Score for the Three Instruction Groups and the Control Group with the Mean Oral Temperature Held Constant Source of Variation df Mean Square F A-instruction 3 8.42 0.61 B-Proficiency 1 803.55 58.54^ C-Sex 1 77.49 5.64^a A X B 3 25.44 1.85 A X C 3 11.85 0.86 B X C 1 5.74 0.42 A X B X C 3 9.52 0.69 Within 63 13.73 Total 78 ♦Significant at .05 level ♦♦Significant at .01 level aGirls > Boys 275 TABLE.7 4 Results of the Analysis of Covariance for the Mathematical Score for the Three Instruction Groups and the Control Group with the Mean Oral Temperature Held Constant Source of Variation df Mean Square F A-Instruction 3 26.14 1.70 B-Proficiency, 1 521.59 53.57** C-Sex 1 162.64 10.61*** A X B 3 4.35 0.28 A X C ■3 48.10 3.14# B X C 1 1.98 0.13 A X B X C 3 42.98 2.80 Within 63 15.34 Total 78 ♦Significant at .05 level ♦♦Significant at .01 level aBoys > Girls TABLE 75 Means and Standard Deviations for Pretest Respiration Rate Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxiety-Reducing 17 . 9 3 o 5 22.6 3.4 20.2 3.5 Neutral 17 . 5 2 »8 20.6 3.7 19.0 3.3 Anxiety-Producing 170 5 2 c 9 19.2 2.9 18.3 2.9 Total 17 e6 3 ol to o 0 00 3.3 19.2 3.2 Control for Physiological Measures 18.4 3.6 17.6 3.9 18.0 3.8 Total 00 0 00 3.2 20.0 3.5 18.9 3.3 n = 10 276 TABLE 76 Means and Standard Deviations for Pretest Respiration Depth Instruction Groups Boys Mean S.D= Girls Mean S.D. Total Mean S.D. Anxiety-Reducing 30.9 11.4 20.8 9.2 25.9 10.3 Neutral 23.2 9.1 29.2 13.8 26.2 11.4 Anxiety-Producing 27.2 11.3 21.2 9.3 24.2 10.3 Total 27.1 10.6 23.7 10.7 25.4 10.7 Control for Physiological Measures 21.0 4.6 21.4 10.5 21.2 7.5 Total 25.6 9.1 23.2 10.7 24.4 9.9 n = 10 j 277 TABLE 77 Means and Standard Deviations for Pretest Heartbeat Rate Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxiety-Reducing 72.8 10.1 00 0 00 16.6 00 0 00 r - ~ 13.4 Neutral 74.7 12.1 85.0 11.1 79.8 11.6 Anxiety-Producing 74.6 11.6 80.3 12.9 77.4 12.2 Total 74.0 11.2 83.4 13.5 78.7 12.4 Control for Physiological Measures 72.7 11.5 87.8 12.9 80.2 12.2 Total 73.7 11.3 84.5 13.4 79.1 12.3 TABLE 78 Means and Standard Deviations for Pretest GSR Instruction Groups Boys Mean S.D, Girls Mean S.D. Total Mean S.D. Anx i e ty-R educ ing 82.2 21.1 71.5 15.3 76.8 18.2 Neutral 87.2 38.8 86.8 28.1 87.0 33.5 Anxiety-Producing 95.3 62.9 73.5 16.6 84.4 39.8 Total 88.2 41.0 77.3 20.0 82.8 30.5 Control for Physiological Measures 80.2 28.5 90.0 43.9 85.1 36.2 Total 86.2 37.8 80.4 26.0 83.3 31.9 n = 10 279 TABLE 79 Means and Standard Deviations for Pretest Systolic Blood Pressure Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxiety-Reducing 121.8 10.3 115.4 10.7 118.6 10.5 Neutral 116.2 10.5 118.2 11.9 117.2 11.2 Anxiety-Producing 126.0 12.3 110.7 7.1 118.3 9.7 Total 121.3 11.0 114.8 9.9 118.1 10.4 Control for Physiological Measures 119.0 11.6 116.6 11.4 117.8 11.5 Total 120.8 11.2 115.2 10.3 118.0 10.7 n = 10 TABLE 80 Means and Standard Deviations for Pretest Diastolic Blood Pressure Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxiety-Reducing 69.4 7.2 68.8 8.2 69.1 7.7 Neutral 67.4 9.0 70.0 5.9 68.7 7.5 Anxiety-Producing 64.0 4.9 64.9 6.8 64.4 5.8 Total 66.9 7.0 67.9 7.0 67.4 7.0 Control for Physiological Measures 67.6 5.5 65.6 6.4 66.6 5.9 Total 67.1 6.6 67.3 6.8 67.2 6.7 n = 10 281 TABLE 81 Means and Standard Deviations for Pretest Pulse Pressures Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxiety-Reducing 52.4 12.7 46.6 9.3 49.5 11.0 Neutral 48.8 14.9 47.5 13.7 48.2 14.3 Anxiety—Producing 62.0 12.8 46.2 6.5 54.1 9.6 Total 54.4 13.5 46.8 9.8 50.6 11.6 Control for Phys iological Measures 51.4 11.0 51.0 9.1 51.2 10.0 Total 54.4 22.9 47.8 9.7 50.7 11.2 n * 10 282 TABLE 82 Means and Standard Deviations for Pretest Oral Temperature Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxiety-Reducing 99.2 0.5 99.5 0.4 99.4 0.4 Neutral 98.9 0.6 99.7 0.7 99.3 0.6 Anxiety-Producing ’ 99.1 0.6 98.8 0.6 99.0 0.6 Total 99.1 0.6 99.4 0.6 99.2 0.6 Control for Physiological Measures 98.9 0.6 99.6 0.6 99.3 0.6 Total 99.0 0.6 99.4 0.6 99.2 0.6 n = 10 283 TABLE 83 Means and Standard Deviations for Pretest Pace Temperature Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxiety-Reducing 95.3 1.8 92.8 1.4 94.0 1.6 Neutral 94.9 1.7 94.3 1.5 94.6 1.6 Anxiety-Producing 94.7 1.3 93.7 2.0 94.2 1.6 Total 94.9 1.6 93.6 1.6 94.3 1.6 Control for Physiological Measures 94.4 1.6 92.8 3.0 93.6 2.3 Total 94.8 1.6 93.4 2.0 94.1 1.8 n = 10 284 TABLE 84 Means and Standard Deviations for Pretest Finger Temperature Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxiety-Reducing 84.1 7.0 83.8 5.2 83.9 6.1 Neutral 85.2 5.8 83.4 4.5 84.3 5.2 Anx iety-Produc ing 84.5 7.3 79.6 2.6 82.0 5.0 Total 84.6 6.7 82.2 4.1 83.4 5.4 Control for Physiological Measures 86.3 6.6 81.2 5.7 83.8 6.6 Total 85.0 6.7 82.0 4.5 83.5 5.6 n = 10 285 286 TABLE 85 Unadjusted Means and Adjusted Means with the Respiration Rate Held Constant for the Ability Test with the Three Types of Instruction Bovs Girls Unadj. Adj. Unadj. Adj . Groups________________Means Means Means_______ Means Total Scores A-Reducing I 28.1 28.2 24.5 24.1 Neutral I 25.6 25.8 23.4 23.2 A-Producing I 23.7 23.9 29.5 29.5 Verbal Scores A-Reducing I 12.7 12.7 14.0 14.1 Neutral I 13.0 12.9 13.2 13.2 A-Producing I 12.1 12.0 16.3 16.3 Mathematical Scores A-Reducing I 16.2 16.3 10.5 10.3 Neutral I 12.6 12.7 10.2 10.1 A-Producing I 12.2 12.3 13.2 12.8 n = 10 287 TABLE 86 Unadjusted Means and Adjusted Means with the Respiration Depth Held Constant for the Ability Test with the Three Types of Instruction Boys Girls Unadj. Adj. Unadj. Adj. Groups________________Means____Means Means______Means Total Scores A-Reducing I 28.1 28.2 24.5 24.4 Neutral I 25.6 25.5 23.4 23.5 A-Producing I 23.7 23.7 29.5 29.5 Verbal Scores A-Reducing I 12.7 12.8 14.0 1.4.0 Neutral I 13.0 12.9 13.2 13.1 A-Producing I 12.1 12.0 16.3 16.3 ! Mathematical Scores A-Reducing I 16.2 16.2 10.5 10.5 Neutral I 12.6 12.6 10.2 10.5 A-Producing I 12.2 12.2 13.2 13.2 n = 10 288 TABLE 87 Unadjusted Means and Adjusted Means with the Heartbeat Rate Held Constant for the Ability Test with the Three Types of Instruction Boys Girls Unadj. Adj. Unadj. Adj. Groups________________Means____Means Means______ Means Total Scores A-Reducing I 28.1 28.1 24.5 24.7 Neutral I 25.6 25.5 23.4 23.3 A-Producing I 23.7 23.8 29.5 29.7 Verbal Scores A-Reducing I 12.7 12.7 14.0 13.9 Neutral I 13.0 13.0 13.2 13.6 A-Producing I 12.1 12.2 16.3 16.2 Mathematical Scores A-Reducing I 16.2 16.2 10.5 10.8 Neutral I 12.6 12.5 10.2 10.1 A-Producing I 12.2 11.2 13.2 13.4 n = 10 289 TABLE 88 Unadjusted Means and Adjusted Means with the Diastolic Blood Pressure Held Constant for the Ability Test with the Three Types of Instruction Groups Boys Unadj. Adj. Means Means Unadj Means Girls Adj. Means Total Scores A-Reducing I 28.1 28.7 24.5 24.9 Neutral I 25.6 25.6 23.4 24.4 A-Producing I 23.7 22.7 29.5 29.0 Verbal Scores A-Reducing I 12.7 13.0 14.0 14.2 Neutral I 13.0 12.8 13.2 13.9 A-Producing I 12.1 11.5 16.3 16.0 Mathematical Scores A-Reducing I 16.2 16.4 10.5 10.7 Neutral I 12.6 12.6 10.2 10.6 A-Prdducing I 12.2 11.8 13.2 13.0 n = 10 290 TABLE 89 Unadjusted Means and Adjusted Means with the Pulse Pressure Held Constant for the Ability Test with the Three Types of Instruction Boys Girls Unadj. Adj. Unadj. Adj. Groups________________Means____Means Means______ Means Total Scores A-Reducing I 28.1 27.8 24.5 25.1 Neutral I 25.6 25.9 23.4 23.9 A-Producing I 23.7 21.9 29.5 30.2 Verbal Scores A-Reducing I 12.7 12.5 14.0 14.4 Neutral I 13.0 13.2 13.2 13.5 A-Producing I 12.1 11.0 16.3 16.7 Mathematical Scores A-Reducing I 16.2 16.1 10.5 10.8 Neutral I 12.6 12.7 10.2 10.4 A-Producing I 12.2 11.3 13.2 13.6 n = 10 291 TABLE 90 Results of the Analysis of Variance for the Mean Respiration Rate for the Three Instruction Groups and the Control Group Source of Variation df Mean Square F A-Instruction 3 3 6 . 1 9 4 . 3 5 * * a B-Proficiency 1 . 2 0 . 0 2 C-Sex 1 4 0 . 0 1 4 . 8 1 * b A X B 3 7 . 4 9 . 9 0 A X C 3 6 . 5 9 . 7 9 B X C 1 7 . 9 0 . 9 5 A X B X C 3 6 . 3 7 . 7 6 Within 6 4 8 . 3 2 — Total 7 9 *Significant at , 0 5 i level ★★Significant at . 0 1 level aAnxiety-reducing instruction group^neutral > Anxiety-producing^control for physiological measures. bGirls Boys : | i 292 TABLE 91 Results of the Analysis of Variance for the Mean Minus the Initial Respiration Depth for the Three Instruction Groups and the Control Group Source of Variation df Mean Square F A-Instruction 3 1.92 0.03 B-Proficiency 1 2.85 0.05 C-Sex 1 5.26 0.09 A X B 3 17.22 0.30 A X C 3 33.92 0.59 B X C 1 5.78 0.10 A X B X C 3 85.03 1.48 Within 64 57.56 Total 79 293 TABLE 92 Results of the Analysis of Variance for the Mean Heartbeat. Rate for the Three Instruction Groups and the Control Group Source of Variation df Mean Square P A-Instruction 3 4736.45 1.07 B-Proficiency 1 1598.11 0.36 C-Sex 1 98.90 0.02 A X B 3 4275.53 0.97 A X C 3 3984.14 0.90 B X C 1 3312.04 0.75 A X B X C 3 2781.91 0.63 Within 64 4419.11 Total 79 294 TABLE 93 Results of the Analysis of Variance for the Mean Minus the Initial GSR for the Three Instruction Groups and the Control Group Source of Variation df Mean Square P A-Instruction 3 431.94 1.41 B-Proficiency 1 41.28 0.13 C-Sex 1 438.36 1.43 A X B 3 48.76 0.16 A X C 3 155.63 0.51 B X C 1 20.03 0.06 A X B X C 3 343.34 1.12 Within 64 306.54 Total 79 295 TABLE 94 Results of the Analysis of Variance for the Mean Systolic Blood Pressure for the Three Instruction Groups and the Control Group Source of Variation df Mean Square F A-Instruction 3 24.47 0.27 B-Proficiency 1 145.56 1.63 C-Sex 1 1109.97 12.43 **a A X B 3 133.12 1.49 A X C 3 119.25 1.34 B X C 1 201.22 2.25 A X B X C 3 40.04 0.45 Within 64 89.32 — Total 79 **Significant at aBoys y Girls .01 level 296 TABLE 95 Results of the Analysis of Variance for the Mean Diastolic Blood Pressure for the Three Instruction Groups and The Control Group Source of Variation df Mean Square F A-Instruction 3 74.04 1.68 B-Proficiency 1 49.20 1.12 C-Sex 1 1.51 0.03 A X B 3 56.27 1.28 A X C 3 39.91 0.91 B X C 1 0.99 0.02 A X B X C 3 16.41 0.37 Within 64 43.99 Total 79 297 TABLE 96 Results of the Analysis of Variance for the Mean Pulse Pressure for the Three Instruction Groups and the Control Group Source of Variation df Mean Square F A-Instruction 3 84.14 0.77 B-Proficiency 1 1.78 0.02 C-Sex 1 1314.80 12.10^a A X B 3 256.31 2.36 A X C 3 164.94 1.52 B X C 1 215.56 1.98 A X B X C 3 63 .14 0.58 Within 64 108.68 Total 79 ♦♦Significant at .01 level aBoys } Girls 298 TABLE 97 Results of the Analysis of Variance for the Mean Oral Temperature for the Three Instruction Groups and the Control Group Source of Variation df Mean Square F A-Instruction 3 0.26 1.77 B-Proficiency 1 0.23 1.53 C-Sex 1 2.70 18.22**3 A X B 3 0.07 0.46 A X C 3 0.35 2.34 B X C 1 0.08 0.58 A X B X C 3 0.02 0.10 Within 64 0.15 __ Total 79 **Significant at .01 level 299 TABLE 98 Results of the Analysis of Variance for the Mean Face Temperature for the Three Instruction Groups and the Control Group Source of Variation df Mean Square F A-Instruction 3 4.32 1.97 B-Profic iency 1 0.09 0.04 C-Sex 1 13.41 6.12 *a A X B 3 0.05 0.02 A X C 3 0.47 0.22 B X C 1 0.02 0.01 A X B X C 3 0.08 0.04 Within 64 2.19 Total 79 ♦Significant at .05 level ^Boys y Girls 300 TABLE 99 Results of the Analysis of Variance for the Mean Finger Temperature for the Three Instruction Groups and the Control Group Source of Variation df Mean Square F A-Instruction 3 40.94 0.82 B-Proficiency 1 5.64 0.11 C-Sex 1 246.41 4.96*a A X B 3 45.93 0.92 A X C 3 53.71 1.08 B X C 1 5.62 0.11 A X B X C 3 14.08 0.28 Within 64 49.66 Total 79 *Significant at .05 level aBoys y Girls 301 UNADJUSTED BOYS 31 ADJUSTED MEAN UNADJUSTED 30 GIRLS ADJUSTED 29 28 27 z < LU z 26, 25 24 23 22 21 ANXIETY- NEUTRAL ANXIETY- REDUCING PRODUCING I INSTRUCTIONS ! j Fig. 10. Unadjusted and adjusted means for the total test | scores with the initial respiration rate held constant for the ■ three types of instructions. MEAN 302 23 22 21 20 19 18 17 16 15 14 13 12 11 1 0 1 — _____________ I ____________________] — _________________ I ANXIETY- NEUTRAL ANXIETY- REDUCING PRODUCING INSTRUCTIONS Fig. 11. Unadjusted and adjusted means for the verbal scores with the initial respiration rate held constant for the three types of instructions. BOYS UNADJUSTED BOYS ADJUSTED MEANS GIRLS UNADJUSTED _ GIRLS ADJUSTED MEAN 303 23 UNADJUSTED 22 BOYS ADJUSTED MEAN 21 UNADJUSTED GIRLS ADJUSTED 20 19 18 17 16 15 14 12 11 10 ANXIETY- NEUTRAL ANXIETY- REDUCING PRODUCING INSTRUCTIONS Fig. 12. Unadjusted and adjusted means for the mathematical scores with the initial respiration rate held constant for the three types of instructions. MEAN 304 UNADJUSTED BOYS ADJUSTED 31 MEAN UNADJUSTED GIRLS 30 ADJUSTED 29 28 27 26 25 24 23 22 21 ANXIETY- NEUTRAL ANXIETY- REDUCING PRODUCING INSTRUCTIONS Fig. 13* Unadjusted and adjusted means for the total test scores with the initial respiration depth held constant for the three types of instructions. 305 23 UNADJUSTED 22 BOYS ADJUSTED MEAN 21 UNADJUSTED GIRLS ADJUSTED 20 19 18 17 < 1 1 . ) 16 15 14 13 12 11 10 ANXIETY- NEUTRAL ANXIETY- REDUCING PRODUCING INSTRUCTIONS Fig. 1^-. Unadjusted and adjusted means for the verbal scores with the initial respiration depth held constant for the three types of instructions. 23 22 21 20 19 18 17 16 15 14 13 12 11 10 F i 306 UNADJUSTED BOYS ADJUSTED . UNADJUSTED GIRLS ADJUSTED * • / „ ____ J I Z-------------L ANXIETY- REDUCING NEUTRAL ANXIETY- PRODUCING INSTRUCTIONS g. 15. Unadjusted and adjusted means for the mathematical with the initial respiration depth held constant for the types of instructions. MEAN 307 UNADJUSTED BOYS ADJUSTED MEAN 31 UNADJUSTED GIRLS ADJUSTED 30 29 28 27 26 25 24 23 22 21 ANXIETY- NEUTRAL ANXIETY- REDUCING PRODUCING INSTRUCTIONS Fig. 16. Unadjusted and adjusted means for the total test scores with the initial heartbeat rate held constant for the three types of instructions. 22 21 20 19 18 17 16 15 14 13 12 11 10 F i < 308 UNADJUSTED BOYS ADJUSTED MEAN UNADJUSTED GIRLS ADJUSTED ANXIETY- NEUTRAL ANXIETY- REDUCING PRODUCING INSTRUCTIONS 17. Unadjusted and adjusted means for the verbal scores initial heartbeat rate held constant for the three types uctions. 23 22 21 20 19 18 17 16 15 14 13 12 11 10 '9* > wi 309 UNADJUSTED BOYS ADJUSTED . MEAN UNADJUSTED GIRLS ADJUSTED / // A ' ANXIETY- REDUCING NEUTRAL ANXIETY- PRODUCING INSTRUCTIONS Adjusted and unadjusted means for the mathematical the initial heartbeat rate held constant for the of instructions. MEAN 310 UNADJUSTED BOYS ADJUSTED 31 MEAN UNADJUSTED GIRLS 30 ADJUSTED . 29 28 27 26 25 24 23 22 21 ANXIETY- NEUTRAL ANXIETY- REDUCING PRODUCING INSTRUCTIONS Fig. 19* Unadjusted and adjusted means for the total test scores with the initial diastolic blood pressure held constant for the three types of instructions. MEAN 311 UNADJUSTED 23 BOYS ADJUSTED MEAN UNADJUSTED GIRLS ADJUSTED ANXIETY- REDUCING NEUTRAL ANXIETY- PRODUCING INSTRUCTIONS Fig. 20. Unadjusted and adjusted means for the verbal scores with the initial diastolic blood pressure held constant for the three types of instructions. 312 23 UNADJUSTED 22 BOYS ADJUSTED MEAN 21 UNADJUSTED GIRLS ADJUSTED 20 19 18 17 6 15 14 13 12 11 10 ANXIETY- NEUTRAL ANXIETY- REDUCING PRODUCING INSTRUCTIONS Fig. 21. Unadjusted and adjusted means for the mathematical scores with the initial diastolic blood pressure held constant for the three types of instructions. MEAN 313 UNADJUSTED BOYS ADJUSTED MEAN UNADJUSTED 31 GIRLS ADJUSTED 22 ANXIETY- NEUTRAL ANXIETY- REDUCING PRODUCING INSTRUCTIONS Fig. 22. Adjusted and unadjusted means for the total test scores with the initial pulse pressure held constant for the three types of instructions. MEAN 314 23 UNADJUSTED BOYS ADJUSTED 22 MEAN UNADJUSTED GIRLS 21 ADJUSTED 20 19 18 17 16 15 14 13 12 11 10 ANXIETY- NEUTRAL ANXIETY- REDUCING PRODUCING INSTRUCTIONS Fig. 23- Unadjusted and adjusted means for the verbal scores with the initial respiration depth held constant for the three types of instructions. 23 22 21 20 19 18 17 16 15 14 13 12 11 10 F 315 UNADJUSTED BOYS ADJUSTED MEAN UNADJUSTED GIRLS ADJUSTED ANXIETY- REDUCING NEUTRAL INSTRUCTIONS ANXIETY- PRODUCING 2k. Unadjusted and adjusted means for the mathematical ith the initial respiration rate held constant for the pes of instructions. TABLE 100 Means and Standard Deviations for Face Temperature Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxiety-Reducing 96.3 0.9 95.5 1.4 95.9 1.1 Neutral 96.2 1.2 95.8 1.3 96.0 1.2 Anxiety-Producing 00 * in & > 1.4 94.9 1.7 95.4 1.6 Total 96.1 1.2 95.4 1.4 95.8 1.3 Control for Physiological Measures 95.6 1.0 94.* 2.0 95.0 1.5 Total 96.0 1.1 95.2 1.6 95.6 1.4 n = 10 316 TABLE 101 Means and Standard Deviations for Finger Temperature Boys Girls Total Instruction Groups Mean S.D. Mean S.D. Mean S.D. Anxiety-Reducing 86.8 . CO 87.7 5.4 87.8 7.0 Neutral 88.4 6.9 84.3 5.8 86.3 6.4 Anxiety-Producing 86.1 8.4 82.0 3.9 84.0 6.2 Total 87.1 8.0 84.7 5.1 86.1 6.5 Control for Physiological Measures 88.4 7.9 81.4 6.3 84.9 7.1 Total 87.4 8.0 83.9 5.4 85.8 6.7 n = 10 317 TABLE 102 Means and Standard Deviations for Systolic Blood Pressure Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anx i ety-Reduc ing 120.7 7.6 109.3 10.5 115.0 9.1 Neutral 114.4 10.8 110.7 8.7 112.6 9.8 Anxiety-Producing 121.1 11.4 109.1 7.6 115.1 9.5 Total 118.7 10.0 109.7 9.0 114.2 9.5 Control for Physiological Measures 115.7 7.5 113.1 11.1 114.4 9.3 Total 118.0 9.4 110.6 9.5 114.3 9.4 Control for Test Performance 115.8 10.6 110.2 9.4 113.0 10.0 Total 117.5 9.6 110.5 9.5 114.0 9.5 n = 10 318 TABLE 103 Means and Standard Deviations for Pulse Pressure Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxi ety-Reduc ing 50.3 8.6 41.3 4.8 45.8 6.7 Neutral 50.1 15.5 43.0 13.0 46.6 14.3 Anxiety-Producing 57.8 12.9 42.9 8.6 50.4 10.8 Total 52.7 12.3 42.4 8.8 47.6 10.6 Control for > Physiological Measures 49.5 7.4 48.6 9.8 49.0 8.6 Total 51.9 11.1 44.0 9.1 47.9 10.1 n = 10 319 TABLE 104 Means and Standard Deviations for Mean Diastolic Blood Pressure Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxi ety-Reduc ing 72.0 7.2 68.7 8.3 70.0 7.7 Neutral 65.9 8.5 68.9 4.6 67.4 6.6 Anxiety-Producing 65.1 5.3 66.9 6.3 66.0 5.8 Total 67.4 7.0 67.8 6.4 67.6 6.7 Control for Physiological Measures 66.6 5.8 64.9 5.2 65.8 5.5 Total 67.2 6.7 67.1 6.1 67.2 6.4 Control for Test Performance 67.2 8.6 64.8 5.6 66.0 7.1 Total 67.2 7.1 66.6 6.0 67.0 6.5 n = 10 320 TABLE 105 Means and Standard Deviations for Oral Temperature Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxiety-Reducing 99.6 0.4 100.2 0.2 99.9 0.3 Neutral 99.9 0.4 100.3 0.5 100.1 0.4 Anxiety-Producing 99.9 0.3 99.9 0.2 99.9 0.3 Total 99.8 0.4 100.1 0.3 100.0 0.3 Control for Phys iological Measures 99.6 0.3 100.2 0.5 99.9 0.4 Total 99.8 0.4 100.1 0.4 99.9 0.4 Control for Test Performance 99.2 0.7 99.4 0.4 99.3 0.5 Total 99.6 1.4 100.0 0.4 99.8 0.4 n = 10 321 TABLE 106 Means and Standard Deviations for Mean Respiration Rate Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxiety-Reducing 18.6 3.0 21.3 3.2 19.9 3.1 Neutral 17.8 3.3 19.8 2.5 18.8 2.9 Anxiety-Producing 16.9 2.2 18.1 3.1 17.5 2.7 Total 17.7 2.9 19.7 2.9 18.7 2.9 Control for Physiological Measures 17.0 2.7 17.0 2.4 17.0 2.6 Total 17.6 2.8 19.0 2.8 18.3 2.8 n = 10 322 TABLE 107 Means and Standard Deviations for the Mean Minus the Pretest Respiration Depth Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anx i e ty-Reduc ing -2.2 6.7 0.5 8.1 -0.9 7.4 Neutral 1.7 4.8 -1.4 6.9 0.1 5.9 Anxiety_I>r' oducing -1.1 9.7 -0.6 7.0 -0.9 8.3 Total -0.5 7.1 0.5 7.3 -0.5 7.2 Control for Phys iological Measures 1.0 6.4 -1.9 9.1 -0.4 7.7 Total C M • 0 1 6.9 -0.6 7.8 -0.5 7.3 n = 10 323 TABLE 108 Means and Standard Deviations for the Mean Respiration Depth Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxiety-Reducing 28.7 10.4 21.4 10.2 25.0 10.3 Neutral 25.0 11.3 27.7 15.2 26.4 13.2 Anxiety-Producing 26.0 8.1 20.6 9.0 23.3 8.5 Total 26.6 9.9 23.2 11.5 24.9 10.7 Control for Physiological Measures 22.0 6.3 19.6 8.5 20.8 7.4 Total 25.4 9.0 22.3 10.7 23.9 9.9 n = 10 324 TABLE 109 Means and Standard Deviations for the Heartbeat Rate Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxi e ty-Reduc ing 73.9 9.1 84.8 11.7 79.4 10.4 Neutral 77.0 9.5 88.2 12.4 82.6 11.0 Anxiety-Producing 74.2 9.3 83.7 9.5 79.0 9.4 Total 74.0 9.3 85.6 11.2 80.3 10.3 Control for Physiological Measures 72.3 8.7 88.0 10.5 80.2 9.6 Total 74.4 9.2 86.2 11.0 80.2 10.1 Control for Test Performance 61.2 3.7 74.0 11.6 67.6 7.7 Total 71.7 8.1 83.7 11.1 77.7 9.6 n = 10 325 TABLE 110 Means and Standard Deviations for the Mean Minus the Pretest Reading for GSR Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxiety—Reducing -7.8 14.3 1 h- a 0 0 21.1 0 0 » i 17.7 Neutral -6.0 10.9 -0.2 15.8 -3.1 13.4 Anxiety—Producing -15.4 16.2 -4.3 9.8 -9.8 13.0 Total -9.7 13.8 -2.1 15.6 -5.9 14.7 Control ±or Physiological Measures 2.4 16.7 0.0 25.7 1.2 21.2 Total -6.7 14.6 -1.6 18.1 -4.1 16.3 n = 10 326 I ! TABLE 111 Means and Standard Deviations for the GSR Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anx i e ty-Re du c ing 74.5 22.6 69.7 33.0 72.1 27.8 Neutral 81.3 39.9 86.7 39.9 84. Q 39.9 Anxiety-Producing 79.9 48.9 69.3 24.4 74.6 36.7 Total 79.6 37.2 75.2 32.5 76.9 34.8 Control for Physiological Measures 82.6 35.5 90.0 32.3 86.3 33.9 Total 78.6 36.7 78.9 32.4 79.2 34.6 n = 10 i | i 327 MEAN 328 22 21 20 19 18 17 16 BOYS GIRLS ANXIETY- REDUCING NEUTRAL ANXIETY- PRODUCING CONTROL INSTRUCTIONS Fig. 25. Mean respiration rate. MEAN 329 1.5 1.0 IZJ BOYS P~T GIRLS 0.5 -0.5 - 1.0 v\ / 7\ -1.5 -2.0 -2.5 -3.0 f t ANXIETY- REDUCING / / / NEUTRAL ANXIETY- PRODUCING NSTRUCTIONS 7 CONTROL FOR PHYSIOLOGICAL MEASURES Fig. 26. The mean minus the pretest respiration depth. MEAN 330 89' 87 85 83 81 79 77 76 73 71 69 67 65 63 61 BOYS GIRLS ANXIETY- REDUCING NEUTRAL ANXIETY- PRODUCING INSTRUCTIONS CONTROL FOR CONTROL FOR PHYSIOLOGI - TEST PERFORM- CAL MEASURES ANCE Fig. 27. Mean heartbeat rate. MEAN 331 -2 -10 -12 -14 -16 -18 -20 □ b ANXIETY- REDUCING NEUTRAL ANXIETY- PRODUCING CONTROL FOR PHYSIOLOGICAL MEASURES INSTRUCTIONS BOYS GIRLS Fig. 28. Mean minus the pretest reading for galvanic skin response. MEAN 332 121 120 119 118 117 116 115 114 113 112 111 110 109 108 BOYS GIRLS ANXIETY- NEUTRAL ANXIETY- CONTROL FOR CONTROL FOR REDUCING PRODUCING PHYSIOLOGICAL TEST PERFORM- MEASURES ANCE INSTRUCTIONS Fig. 29. Mean systolic blood pressure. MEAN 73 BOYS 72 GIRLS 71 70 69 68 7 67 66 65 64. 63 62 ANXIETY- NEUTRAL ANXIETY- CONTROL FOR REDUCING PRODUCING PHYSIOLOGICAL MEASURES INSTRUCTIONS 333 □ □ CONTROL FOR TEST PERFORM ANCE Fig. 30. Mean diastolic blood pressure. MEAN 334 6CL 58 56 54 52 50 48 46 44 42- 40- 38- BOYS GIRLS ANXIETY- REDUCING NEUTRAL ANXIETY- PRODUCING CONTROL INSTRUCTIONS Fig. 31. Mean pulse pressure. MEAN 335 BOYS GIRLS 100.-25 100.15 100.05 99.95 99.85 99.75 99.65 99.55 99.45 99.35 99.35 99.15 ANXiETY- NEUTRAL ANXIETY- CONTROL FOR CONTROL FOR I REDUCING . PRODUCING PHYSIOLOGICAL TEST PERFORM- MEASURES ANCE Fig. 32. INSTRUCTIONS Mean oral temperature. 96.45 96.25 96.05 95.85 95.65 95.45 95.25 95.15 94.95 94.75 94.55 94.35 Fig. 336 BOYS GIRLS ANXIETY- NEUTRAL ANXIETY- CONTROL REDUCING PRODUCING INSTRUCTIONS 33. Mean face temperature. MEAN 337 88 ! 87 86 85 84 83 82 81 BOYS GIRLS ANXIETY- REDUCING NEUTRAL ANXIETY- PRODUCING CONTROL INSTRUCTIONS Fig. 3^- Mean finger temperature. TABLE 112a Means and Standard Deviations for the Psychological Test of Anxiety Means and Standard Deviations for _____________ S-R Inventory of Anxiousness____________ Boys Girls Total Instruction Groups Mean S.D. Mean S.D. Mean S.D. Anxiety-Reducing 2 8 . 1 6 . 7 3 5 . 6 6 . 7 3 1 . 8 6 . 7 Neutral 2 7 . 9 5 . 8 3 2 . 3 5 . 7 3 0 . 1 5 . 8 Anxiety-Producing 2 7 . 1 4 . 6 3 2 . 8 7 . 1 3 0 . 0 5 . 8 Control for Physiological Measures 2 8 . 0 5 . 9 3 1 . 0 5 . 9 2 9 . 5 5 . 9 Control for Test Performance 3 1 . 0 7 . 2 3 4 . 3 r— 1 e o f — f 3 2 . 6 8 . 6 Toi:al 2 8 . 4 6 . 0 3 3 . 2 7 . 1 3 0 . 8 6 . 6 n = 10 338 TABLE 112b Means and Standard Deviations for the Psychological Test of Anxiety Means and Standard Deviations for the Affect Adjective Check List Boys Girls Total Instruction Groups Mean S.D. Mean S.D. Mean S.D. Anxiety-Reducing 11.1 5.7 13.6 5.9 12.4 5.8 Neutral 9.0 3.9 15.2 4.7 12.1 4.3 Anxiety-Producing 9.3 5.5 16.1 10.4 12.7 7.9 Control for Physiological Measures 12.7 5.4 12.0 i 4.8 12.4 5.1 Control for Test Performance 11.9 4.4 11.6 5.5 11.8 4.9 Total 10.8 5.0 13.7 6.2 12.2 5.6 n = 10 339 TABLE 112c Means and.Standard Deviations for the Psychological Test of Anxiety Means and Standard Deviations for the Test Anxiety Scale Instruction Groups Boys Mean S.D. Mean Girls S.D. Total Mean S.D. Anxiety-Reduc ing 6.9 4.1 10.1 3.6 8.5 3.8 Neutral 6.2 2.4 9.1 3.2 7.6 2.8 Anxiety-Producing 5.7 3.5 9.3 2.8 7.5 3.2 Control for Physiological Measures 6.6 3.8 9.8 2.4 8.2 3.0 Control for Test Performance 7.4 4.0 9.5 4.1 8.4 4.1 Total 6.6 3.6 9.6 3.2 8.1 3.4 n = 10 1 340 TABLE 112d Means and Standard Deviations for the Psychological Test of Anxiety Means and Standard Deviations for the Manifest Anxiety Scale Instruction Groups Boys Mean S.D. Girls Mean S.D. Total Mean S.D. Anxiety-Reduc ing 9.0 5.8 10.2 3.1 9.6 4.4 Neutral 8.1 2.6 11.1 4.3 9.6 3.4 Anxiety-Producing 8.7 4.5 9.5 3.6 9.1 4.1 Control for Physiological Measures 7.8 3.8 11.4 2.0 9.6 2.9 Control for Test Performance 8.5 5.3 10.3 4.6 9.4 4.9 Total 8.4 4.4 10.5 3.5 9.-5 4.0 n = 10 341 342 TABLE 113 Results of Analysis of Variance for the S-R Inventory of Anxiousness Source of Variation df Mean Square F A-Instruction 3 38.58 0.81 B-Proficiency 1 38.80 0.81 C-Sex 1 566.57 11.85' A X B 3 8.49 0.18 A X C 3 15.24 0.32 B X C 1 16.03 0.34 A X B X C 3 93.38 1.93 Within 64 47.81 Total 79 **Significant at .01 level aGirls Boys 343 TABLE 114 Results of Analysis of Variance for the Affect Adjective Check List Administered Prior to the Major Testing with the Instructions to Score it According to Their Usual Feelings During an Examination (Part II of the Student Survey) Source of Variation df Mean Square F A-Instruction 3 4°57 0.13 B-> Proficiency 1 22.07 0.65 C-Sex 1 315.78 9.3o**a A X B 3 105.26 2.10 A X C 3 V 61.80 1.82 B X C 1 71.02 2.09 A X B X C 3 53.83 1.58 Within 64 33.97 Total 79 ♦♦Significant at .01 level aGirls ^ Boys I 344 TABLE 115 Results of Analysis of Variance for the Test Anxiety Scale Source of Variation df Mean Square P A-Instruct ion 3 6.15 0.48 B-Proficiency 1 11.58 0,90 C-Sex 1 179.90 13.92**a A X B 3 8.86 0.68 A X C 3 2.06 0.16 B X C 1 0.54 0.04 A X B X C 3 9.77 0.76 Within 64 12.92 Total 79 **Significant at .01 level aGirls > Boys 345- TABLE 116 Results of Analysis of Variance for the Manifest Anxiety Scale Source of Variation df Mean Square F A-Instruction 3 0.84 0.05 B-Proficiency 1 8.98 0.51 G-Sex 1 65.82 3.74 A X B 3 10.09 0.57 A X C 3 6.86 0.39 B X C 1 0.01 0.00 A X B X C 3 58.33 3.31* Within 64 17.62 Total 79 ♦Significant at .05 level 346 TABLE 117 Results of the Analysis of Variance for the Affect Adjective Check List Administered After the Instruction Stimuli Source of Variation df Mean Square F A-Instruction 3 .27 0.66 B-Prof-iciency 1 .45 0.02 C-Sex 1 12.67 0.52 A X B 3 12.37 0.50 A X C 3 15.88 0.64 B X C 1 0.20 0.01 A X B X C 3 5.21 0.21 Within 64 24.59 Total 79 347 TABLE 118 Data for AACL Administered as a Part of the Student Survey S tandard Treatment Error of Groups Means S.D. Mean Boys Anx.-Reducing 11.1 5.7 .1.8 Neutral 9.0 3.9 1.2 Anx.-Producing 9.3 5.5 1.7 Control for Phy. Measures 12.7 5.4 1.7 Control for Test Performance 11.9 4.4 1.4 Girls Anx.-Reducing 13.6 5.9 1.9 Neutral 15.2 4.7 1.5 Anx.-Producing 16.1 10.4 3.3 Control for Phy. Measures 12.0 4.8 1.5 Control for Test Performance 11.6 5.5 1.7 Boys and Girls Anx.-Reducing 12.4 Neutral 12.1 Anx.-Producing 12.7 Control for Phy. Measures 12.4 Control for Test Performance 11.8 n = 10,20 348 TABLE 119 AACL Administered After the Anxiety Treatment Treatment Groups Means Standard Error of S.D« Mean Variance Anx.-Reducing 6.7 Boys 3.4 1.1 11.3 Neutral 6.9 3.7 1.2 13.7 Anx.-Producing 8.5 5.5 1.7 30.3 Control for Phy. Measures 7.8 3.3 1.0 10.8 Control for Test Performance 10.3 5.0 1.3 17.6 Anx.-Reducing 9.1 Girls 4.6 1.5 21.2 Neutral 9.5 5.2 1.6 27.2 Anx.-Producing 7.8 5.9 1.8 33.5 Control for Phy. Measures 5.2 4.6 1.5 21.4 Control for Test Performance 9.5 6.0 1.6 26.9 Anx.-Reduc ing Neutral Anx .-Produc ing Control for Phy. Measures Control for Test Performance 7.9 8.2 8.2 6.5 9.9 Boys and Girls n = 10,20

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Effects Of Success And Failure On Impulsivity And Distractibility Of Three Types Of Educationally Handicapped Children

PDF

Auditory Perceptual Si Factors As Non-Predictors Of Reading Achievement In An Upper-Class And Upper-Middle-Class Population

PDF

Maternal Child-Rearing Attitudes And Developmental Growth Of Rubella Deaf-Blind Children

PDF

The effect of noise on intellectual performance as related to personality and social factors in upper division high school students

PDF

Testing Attitudes Toward Physical Handicap: Stereotyping And Objectivityin A Partially Integrated Elementary Program

PDF

Effects Of Associative And Category Cuing On Recall Of Items From Exhaustive And Nonexhaustive Lists

PDF

The Conditioning Of A Discriminative Stimulus Measured As An Orienting Reaction In Profoundly Retarded Blind Children

PDF

An Experimental Study Of Two Moral Responses Of Two Socioeconomic Groups Based On Four Paradigms

PDF

Prescriptive Teaching As A Supplement To Behavior Modification In The Remediation Of Learning Disorders

PDF

The Relative Efficiency Of Prompting And Confirmation Learning Paradigms

PDF

Immediate Memory And Its Correlates With School Achievement

PDF

Development And Analysis Of Some Tactual Measures Of Intelligence For Adolescent And Adult Blind

PDF

Delay Of Feedback And The Acquisition And Retention Of Verbal Material Inthe Classroom

PDF

The Effects Of Two Types Of Experimenter Intervention And Schedules Of Reinforcement On Verbal Operant Conditioning Of Affective Self-References

PDF

Ethnicity And Measures Of Educability: Differences Among Navajo, Pueblo And Rural Spanish-American First Graders On Measures Of Learning Style, Hearing Vocabulary, Entry Skills, Motivation And H...

PDF

Effects Of Preschool Enrollment And Parent Participation On Academic Growth

PDF

Confirmation Of Expectancy And Changes In Teachers' Evaluations Of Student Behaviors

PDF

The Visual Fusion Threshold (Vft) Test As A Measure Of Perceptual Efficiency In Kindergarten And First Grade, And As A Possible Predictor Of Later Reading Retardation

PDF

Reading Achievement And In-Grade Retention Rate Differentials For Mexican-American And Black Students In Selected States Of The Southwest

PDF

Academic And Non-Academic Achievement Of Freshmen And Seniors At Azusa Pacific College

Asset Metadata

Creator Chambers, Alma Clyde (author)

Core Title Anxiety, Physiologically And Psychologically Measured, And Its Consequences On Mental Test Performance

Degree Doctor of Philosophy

Degree Program Educational Psychology

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

Tag education, educational psychology,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Wolf, Richard M. (committee chair), Martin, David W. (committee member), McIntrye, Robert B. (committee member), Meyers, Charles Edward (committee member), Rigney, Joseph W. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-593384

Unique identifier UC11359867

Identifier 6805859.pdf (filename),usctheses-c18-593384 (legacy record id)

Legacy Identifier 6805859.pdf

Dmrecord 593384

Document Type Dissertation

Rights Chambers, Alma Clyde

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