University of Southern California Dissertations and Theses

A Study Of Affective Sets: The Effects Of Family And Non-Family Verbal Contexts On Word-Need Stimuli In A Word Association Experiment With Reference To Pleasant And Emotional Tones Of Associated...

(USC Thesis Other)

A Study Of Affective Sets: The Effects Of Family And Non-Family Verbal Contexts On Word-Need Stimuli In A Word Association Experiment With Reference To Pleasant And Emotional Tones Of Associated...

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content T h is d is s e rta tio n h as b e e n 61-6274 m ic ro film e d e x a c tly a s re c e iv e d ABOU— GHORRA, Ib rah im M a n so u r, 1919- A STUDY O F A F F E C T IV E SETS: TH E E F F E C T S O F FA M ILY AND N O N -FA M ILY V ER B A L CON TEX TS ON W ORD-NEED STIM U LI IN A WORD ASSOCIATION E X PE R IM E N T WITH R E FE R E N C E TO PLEA SA N T AND EM OTIONAL TONES O F ASSOCIATED V ER B A L RESPONSES. University Microfilms, Inc., Ann Arbor, Michigan A BO U-GIIORRA, Ib rah im M a n so u r, 1919- 61-6274 U n iv e rs ity of S outhern C a lifo rn ia P h .D ., 1961 P sy ch o lo g y , c lin ic a l University Microfilms, Inc., Ann Arbor, Michigan Copyright by Ibrahim Mansour Abou-Ghorra 1962 A STUDY OF AFFECTIVE SETS: THE EFFECTS OF FAMILY AND NON-FAMILY VERBAL CONTEXTS ON WORD-NEED STIMULI IN A WORD ASSOCIATION EXPERIMENT WITH REFERENCE TO PLEASANT AND EMOTIONAL TONES OF ASSOCIATED VERBAL RESPONSES by Ibrahim Mansour Abou-Ghorra 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 (Psychology) August 1961 UNIVERSITY O F SOUTHERN CALIFORNIA 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 7 , C A L IF O R N IA This dissertation, written by IbraMjR. Mansour.. Abou-Ghorra.......... under the direction of h .ls— Dissertation Com mittee, and approved by all its members, has been presented to and accepted by the Dean of the Graduate School, in partial fulfillment of requirements for the degree of D O C T O R OF P H I L O S O P H Y Dean D ate....... DISSERTATION COMMITTEE . . . . V / J \ tY f l Chairman ACKNOWLEDGMENTS The study to be reported here was made possible through the contributions of many persons to whom the author is most sincerely grateful. Particular gratitude is expressed to Dr. Alfred Jacobs who has patiently trained the author in research and scientific writing, both as a graduate assistant and a doctoral candidate. Thanks are also due to Dr. Georgene Seward, Dr. J. P. Guilford, Dr. Neil Warren, and Dr. D. Welty Lefever for their helpful suggestions throughout the preparation of this study. To Dr. Reginald S. Rood, Dr. Benjamin Siegel, Dr. Fred Cutter, and Mr. Robert Ontell the author wishes to express his thanks for making the gathering of data at Atascadero State Hospital rapid and pleasant. The author also wishes to express his thanks to Dr. Harrison Madden, Dr. William Dienstein, and Dr. Melvin Angfell for aiding in the gathering and preparation of data at Fresno State College. To the subjects, without whose participation this investigation would not have been possible, the author wishes to express his special thanks for their contributions. To the staff of the I.B.M. Western Data Process ing Center at the University of California at Los Angeles, the author is indebted for their generous aid ii in preparing much of the statistical data. The author is especially grateful to William Anderson, Philip Cramer, John Whittlesey, Alice Mark, and Sharon Kurit. To Mrs. Jessie Levine and Mrs* Gertrude Stafford the author is indebted for typing the manuscript and improving its final form. To his wife, Eva, the author remains forever grateful for the many years of patience and the many hours of work which she so generously gave. iii TABLE OF CONTENTS Page ACKNOWLEDGMENTS ................................ ii LIST OF TABLES......................... vi Chapter I. INTRODUCTION TO THE PROBLEM ..... 1 II. BACKGROUND OF THE PROBLEM AND DERIVATION OF THE HYPOTHESES .... 5 The Family Context Hypothesis The Need Satisfaction Hypothesis The Mood Hypothesis The Stimulus Hedonic Tone Facilitation Hypothesis The Emotionality Hypothesis The Stimulus Emotional Tone Facilitation Hypothesis III. INSTRUMENTS AND SUBJECTS.............. 34 The Basic Instrument Definition and Measurement of Variables Definition and Measurement of Criterion Variables The Subjects IV. EXPERIMENTAL PROCEDURES............. 45 Order of Events Instructions Statistical Treatment of the Data V. RESULTS........... 52 Reliability of Individual Judgments Measures of Controlled Variables Ratings of Subject Set Characteristics and Stimulus Variables Measures of Response Variables— Effects of Family Context iv Chapter Page Effects of Pleasant Context Effects of Unpleasant Context Effects of Neutral Context Differences among Responses to Different Need Stimuli Differences between Hospital and College Subjects Correlations between Ratings of Set and Response Variables VI. DISCUSSION............................ 102 VII. SUMMARY.............................. 116 Results Conclusions BIBLIOGRAPHY ................................... 125 APPENDIX A. THE WORD FLUENCY TEST, FORM A . . 129 APPENDIX B. THE SHIPLEY-HARTFORD LIVING INSTITUTE SCALE ............... 132 APPENDIX C. THE GUILFORD-ZIMMERMAN TEMPERAMENT SURVEY ........... 135 APPENDIX D. THE STATISTICAL PROGRAMS AND TESTS...................... 139 v LIST OF TABLES Table Page 1. Correlations between Repeated Judgments of Emotional Tones of Word Stimuli . . 54 2. Correlations between Repeated Judgments of Pleasant Tones of Word Stimuli . . 55 3. Differences between College and Hospital Subjects in Measures of Controlled Variables........................... 56 4. Differences between College and Hospital Subjects in Measures of Subject Set Characteristics................ 58 5. Differences between Ratings of Need Stimuli.............................. 59 6. Difference between Hospital and College Subjects in Their Ratings of Pleasant and Emotional Tones of the Words "Home" and "Parents"................. 60 7. Differences between College and Hospital Subjects in Their Ratings of Emotional Tones of Verbal Stimuli............. 61 3. Differences between College and Hospital Subjects in Their Ratings of Pleasant Tones of Verbal Stimuli............. 62 9. Differences between Family and Neutral Contexts in Eliciting Affective and Non-Affective Associative Word Responses............................ 67 10. Differences between Verbal Contexts in Eliciting Pleasant, Unpleasant, and Neutral Verbal Responses ............. 70 11. Differences between Word Need Stimuli in Eliciting Affective and Non-Affective Verbal Responses..................... 73 vi Table Page 12. Differences between College and Hospital , Subjects in Their Production of Associated Word Responses to Word Need Stimuli......................... 85 13. Differences between Hospital and College Subjects in Production of Pleasant Toned Verbal Responses ............... 86 14. Differences between Hospital and College Subjects in Production of Unpleasant Verbal Responses ..................... 87 15. Differences between Hospital and College Subjects in Production of Neutral Verbal Responses ..................... 88 16. Differences between Hospital and College Subjects in Production of Emotional Verbal Responses ..................... 89 17. Differences between Hospital and College Subjects in Production of Unemotional Verbal Responses ..................... 90 18. Correlations between Judgments of Pleasantness of Verbal Stimuli, Judgment of Pleasantness of Associated Verbal Responses, and Rate of Word Production............................ 94 19. Correlations between Judgments of Emotional Tones of Verbal Stimuli, Judgments of Emotional Tones of Associated Verbal Responses, and Rate of Word Production............. 95 20. Correlations between Ratings of Mood, Emotionality, Need Satisfaction, and Ratings of Pleasant Tones of Verbal Stimuli.............................. 96 21. Correlations between Ratings of Mood, Emotionality, Need Satisfaction, and Ratings of Unpleasant Tones of Verbal Responses............................ 97 vii Correlations between Ratings of Mood, Emotionality, Need Satisfaction, and Ratings of Emotional Tones of Verbal Responses ......................... viii CHAPTER I INTRODUCTION TO THE PROBLEM Set is an ambiguous term. Since its conception, in 1895 (18), it has been used by different psycholo gists to refer to such phenomena as determining tendencies, directing tendencies, contextual stimuli, anchor stimuli, atmospheric conditions, experimental instructions, attitudes, beliefs, intentions, habits, expectancies, anticipations, foresights, tensions, Vectors, needs, preoccupations, and perseverations (22). Equally numerous are the different kinds of sets which Gibson (22) describes in his extensive survey, such as organic set, neural set, motor set, postural set, mental set, voluntary set, unconscious set, situation set, task set (Aufgabe), preparatory set, goal set, temporary set, permanent set, set to per ceive, and set to react. Gibson concluded from his survey that the term "set" seems to be a nearly universal one in psychological thinking despite its ambiguous and indefinite meaning. In an attempt to clarify the meaning of the concept "set," Johnson (7) defines it as fin inter mediate level of integration between the over-all motivational pattern and the specific little acts. 1 In respect to subsequent activities, sets are pre paratory; but at any moment sets are themselves activities of complex organisms. Hence, they are influenced by many conditions in the individual and in his surroundings. Johnson further defines the function of set as facilitation of related responses. He observes that by the use of the word association method it was found that associated word responses to word stimuli vary qualitatively according to the implications of experimental instructions in relation to the word stimulus. For instance, a word stimulus such as man might elicit a specific response such as woman if opposites were called for by the instructions. If, however, the instructions called for class asso ciations, a word response such as human or animal or living being might be expected to occur more readily than a word response such as woman. In the absence of such limiting condition, as in the case of free association, the frequency of occurrence of a word response, far from being a chance event, was found to conform to certain identifiable conditions such as the strength of associative values (13), the special affective state of the individual (19), and the particular motivational states of the organism (13, 27, 28). Johnson speculates that the above findings indicate that set acts as a facilitating factor. In a continuous word association experiment, facilitation refers to two phenomena. It may refer to the fact that a certain word stimulus may generate a greater number of associative word responses within a certain time unit than does another word stimulus. It may also refer to the fact that a stimulus word which possesses a certain attribute may generate a greater number of word responses possessing this particular attribute than does another stimulus word which has a lesser degree of the same attribute. To investigate the facilitation factor in a continuous free word association experiment, the present study was designed to explore the influence of three varying subject set characteristics and three varying experimental set conditions on several consequences of set. The stimulus conditions secure a set through verbal stimuli, upon the assumption that words can and do reinstate the conditions for response provided for by the referent for which they originally stand (13). The subject set character istics whose effects were evaluated in this experiment were studied by choosing subjects who differed in the amount to which they could be characterized as happy or unhappy, as (emotional or non-emotional, and as satisfied or unsatisfied in their needs for nutrition, recognition, and sex. The experimental set conditions whose effects were evaluated in this experiment were studied by presenting verbal stimuli differing in pleasantness, unpleasantness, emotion evoking prop erties, and presumably in the capacity to evoke various combinations of need-associated responses. Consequences of set were the number of word responses associated within a two-minute time unit; the number of pleasant, unpleasant, and neutral toned word responses associated within a two-minute time unit; the number of emotional and unemotional toned word responses associated within a two-minute time unit; and the sum of weighted scores of either pleasant, unpleasant, or emotional tones of verbal responses. The plan of this paper includes a discussion of the background for the hypotheses investigated in this study, which are presented in Chapter II. In Chapters III and IV there is a presentation of the instruments and experimental procedures followed in this research. Chapters V and VI deal with the pres entation and discussion of the experimental findings. CHAPTER II BACKGROUND OF THE PROBLEM AND DERIVATION OF THE HYPOTHESES For many years, students of affectivity such as Beebe-Center (14), Hunt (24), Peters (31), Ruckmick (15), Menzies (14), Rapaport (14), Burt (3), McClelland (10), and Guilford (3) called attention to the impor tance of clarifying the terms used to describe emotional states, and treating affectivity as a multidimensional phenomenon rather than a unitary factor. The present paper is essentially an extension of the effort to clarify the definitions of certain affective variables and to throw some light on their interrelationships. Experimental findings which are relevant to the hypotheses investigated in this research will be discussed under six headings: (1) the Family Context Hypothesis, (2) the Need Satisfaction Hypothesis, (3) the Mood Hypothesis, (4) the Stimulus Hedonic Tone Facilitation Hypothesis, (5) the Emotionality Hypothesis, and (6) the Stimulus ^notional Tone Facilitation Hypoth esis. The Family Context Hypothesis In his book Psychoanalysis as a Science. Hilgard states that the importance of early childhood and family 5 experiences is one of the most familiar and fundamental teachings of psychoanalysis (6:4). If the child's needs are appropriately gratified, the child's early experiences result in a secure individual. If, on the other hand, the child's needs are met in such a way as to induce frustration, the consequent insecurity and anxiety will characterize the individual throughout his life. Jung, in his article on the "Association Method," explains at length how present behavior and verbal associations may be traced to unconscious experiences in infancy. He points out that his use of the word association method enabled him to discover "complexes" formed in infancy. In 1904 he drew attention to the fact that unpleasant emotional word stimuli elicit such responses as a long reaction time, a large GSR deflec tion, and inhibition of response. He further states that such phenomena should be considered indices of repression or the "unconscious resistance to reproduc tion" of early emotionally unpleasant experiences (28:247-249). Jung's indices of repression were called affec tive emotional responses by such investigators as Sears (16), McGinnes (30), Rapaport (14), and McClelland (9). Of special relevance to the present research are McGinnes's and McClelland's views. McGinnes states that early in life most individuals learn that words and symbols having sexual, excretory, or other wise unpleasant or immoral connotations are socially taboo. Since the use of such words by the child will generally result in chastisement by the parents, a conditioned emotional response such as fear will be associated with such words and will persist through adulthood. The stimulus word serves as a cue to deeply embedded anxiety, which is revealed in autonomic reactivity as measured by GSR. Avoidance of further anxiety is contemporaneously aroused in the form of a perceptual defense. In a very penetrating discussion of the affec tive response in relation to parental and early child hood experiences, McClelland also believes that early affective associations are more persistent and much stronger than later associations. Many animal experi ments, and a few studies of human infant behavior, could be used as evidence. From among the many studies which he reviewed, he refers to Hunt's initial study of feeding frustration in young rats as a case in point. He found that if rats were irregularly deprived of food in infancy they tended to hoard more as adults, when deprived of food again, than did rats whose initial feeding frustration occurred after the organism reached maturity. According to McClelland, this may be explained in terms of the association of deprivation cues with anxiety or affective arousal. McClelland believes that since affective arousal is more intense in infancy than in adulthood, and since hoarding is an instrumental response which reduces anxiety, it should follow that when cues are reinstated in adulthood they arouse a greater anxiety in the rats deprived in infancy, which in turn motivates more instrumental hoarding behavior. In humans, comparable results were reported by Dennis (9) and McGraw (9) independently. Although deprivation of social stimulation and intensive training during infancy did not result in retardation or deviation of motor and intellectual development in later years, personality traits such as self-confidence and initiative were impaired in the deprived infants. This suggests the persistence of affective associations through life, as well as their relatively greater strength when aroused in early life. Among the many factors which may contribute to this phenomenon, McClelland (9) discusses the role of the involvement of the autonomic nervous system, children's failure to discriminate time, and more especially the parental generalized threats and promises as well as their irregular rewards and dis cipline. Since children are dependent on their parents for the satisfaction of their needs, deprivation of satisfaction in any form will be associated with the parents or the family context in general. A reinstate ment of family context cues in adulthood should bring a stronger negative affective arousal in those who were more frustrated in infancy than those who were less frustrated in infancy. In other words, where a family context is associated with deprivation, negative or unpleasant affective arousal should be stronger than if a family context is associated with satisfaction. It should also follow that affective arousal is stronger in relation to needs whose initial frustration is associated with the family context, than in relation to needs which were less frustrated or more satisfied within the family context. Therefore, according to McGinnes, response to need stimuli depends upon the degree of family-approved need experiences. In an experiment where nourishment, recognition, and sex experiences are being considered, we may assume that nourishment experiences are supplied by and pleasantly associated with parents, whereas recognition and sex experiences are associated with 10 relatively unpleasant parental pressures to achieve recognition or success and sexual chastity. If this assumption is valid, it will be expected that in a continuous free association experiment a family con text would increase the number of associative responses given to nourishment need stimulus and decrease the number of associative responses given to recognition and sex need stimuli. According to McClelland, a family context should increase the production of unpleasant responses given to recognition and sex need stimuli. From the foregoing discussion, the Family Context Hypothesis may be expressed as follows: Hypothesis I: In a continuous free word associa tion experiment, it is expected that subjects would associate a greater number of word responses to a word stimulus which represents the nutrition need, presented within a family context, than to the same word-need stimulus pre sented singly or within a neutral context; and would associate with fewer word responses to word stimuli which represent the 11 recognition and sex needs, pre sented within a family context, than to the same word-need stimuli presented singly or within a neutral context. It is also expected that sub jects associate a greater number of unpleasant word responses to recognition and sex word-need stimuli presented within a family context them to the same word-need stimuli presented singly or within a neutral context. The Need Satisfaction Hypothesis The degree of need satisfaction has special effects on behavior and affective states. Keys (3), in a study of the effect of degrees of semi-starvation on different aspects of behavior, reports that whereas intelligence did not seem to be markedly impaired, personality characteristics had undergone serious changes which he called semi-starvation neurosis. The main characteristics of this condition were apathy, depression, loss of self-confidence, and a marked decrease in sociability and sexual desire. Food 12 dominated the subjects' conversation, their reading material, and their expressed future vocational aspirations. Brozek, Guetzkow, and Baldwin (21) conducted a quantitative study of perception and association in experimental semi-starvation. Thirty-six young men served as volunteer subjects, losing one-fourth of their initial body weight within six months. They were given a variety of tests including word associa tion test, Rorschach, Rozenzweig's P-F test, and dream content analysis. Only the free word associa tion test provided limited evidence that semi-starva tion resulted in a significant increase in the hunger drive and significantly altered the associative process. The alteration was not of a gross variety such as blocking, failure to follow directions, or peculiar ideation which are characteristic of various types of mental illness. It simply consisted of personalized type of response, neither idiosyncratic, illogical, nor farfetched. It was only unique in a statistical sense, i.e. given by less than 1 per cent of either of two groups of 1,000 subjects each;which were sampled in previous studies of association. Other than this index there was no increase of food related responses in spite of abundant clinical 13 evidence of their preoccupation with food. Of interest, however, is another finding which has more immediate bearing on our study of affective states. The average scores of the depression scale of the MMPI rose from 54.2 during the control period to 73.9 after twenty- four weeks of starvation, and declined to 65.7 after the first twelve weeks of rehabilitation. Postman and Crutchfield (32) studied the inter action of need, set, and stimulus structure in a cognitive task. They investigated the relationship between the intensity of need and the frequency of need-related responses. They dealt with the relation ship between intensity of hunger and frequency of food-related responses to incomplete words under different conditions of stimulus ambiguity and selec tive set. They concluded that there was no simple one-to-one relationship between the intensity of need and the frequency of need-related responses. In a sense Postman and Crutchfield's study confirmed Brozek, et al.'s finding that content analysis of response words does not yield a satis factory index of the effects of need satisfaction or deprivation on verbal behavior. It was Wisp6 (36) who threw considerable light on previous failures. He followed a lead proposed by McClelland and Atkinson (36) and tried to break down the verbal con tent of a word association experiment into categories such as act responses, object responses, affective responses, instrumental responses, and neutral responses. In so doing he obtained a clearer picture of the relationship of the intensity of need depriva tion to the frequency of need-related responses. He found that the number of food and water responses does not increase linearly with hours of deprivation. In general there is a significant increase in the food-related responses up to a certain point, beyond which these responses decrease. In his own experiment there was an increase in act, object, and affective responses related to food at the tenth hour of depri vation, and a decrease of these responses at the twenty-fourth hour of deprivation. Analysis of these results, in the light of previous findings by other investigators, led Wisp6 to conclude that the rela tionship between physiologically induced deprivation states and need related word responses is probably curvilinear. Since the primary concern of this paper is the study of affective responses, and since the conclusion derived from previous studies is that frustration of needs in contrast to starvation results in an increase 15 of affective responses, the Need Satisfaction Hypothesis which is investigated in the experiment to be reported may be stated as follows: Hypothesis II: In a continuous free word asso ciation experiment, it is expected that subjects would associate significantly a greater number of affective word responses to a stimulus word which represents a more frustrated need than to a stimulus word which represents a more satisfied need. It is also expected that sub jects with lower ratings of need satisfaction would associate significantly a greater number of affective word responses to word-need stimuli than do subjects with higher ratings of need satisfaction. The Mood Hypothesis The over-all degree of satisfaction of various needs within the organism may be expressed in terms of affective states. Among these affective states, 16 mood and emotionality are two conceptualized variables whose properties and relationships to certain behavioral characteristics are yet to be adequately clarified. Ruckmick (15) notes that differentiation of mood and emotion is a difficult task in the absence of pertinent experimental data. As late as 1936 he related that any such distinction is to be made solely on the basis of general uncontrolled observations. He suggests that emotion be limited to short-lived strong affective states with physiological concomitants, and that mood be defined as a more lasting or lingering effect of the aroused emotions. Hunt (25), in his critical review of current approaches to affectivity, states that mood occupies a position between affection and emotion. He defines affection as a general feeling tone, and emotion as a specific feeling which accom panies an autonomic physiological change. Bousfield (19) suggests that the term emotion should be applied to relatively strong motivational states identified through pronounced physiological concomitants; whereas mood may be defined as a generic term denoting a general feeling tone which is resultant of the specific feeling tones associated with the specific motivational systems operating at any one time. Krech and Crutchfield (8) define moods as pervasive transitory 17 emotional states which usually have two principal features: (1) degree of tension, and (2) pleasantness or unpleasantness. Experimental evidence favoring a distinction between mood and emotion was first supplied by Sullivan (14) in the form of a correlation approaching zero between self ratings of mood and several physio logical measures used to define emotion, such as pulse rate and blood pressure. Relevant to the present study is the question, "What are the effects of mood on the production and affective hedonic tones of verbal responses in a word association experiment?" There are two points of view to be considered. The first point of view asserts that mood is a facilitating condition to like responses. A pleasant mood is inducive to production of pleasant responses; unpleasant mood is inducive to production of unpleasant toned responses. The second point of view asserts that mood is a motivational condition and follows the general trend of motivational movement toward securing satisfaction and pleasure. Frustration and deprivation accentuate this movement and bring about a host of pleasurable responses with or without special relevance to goal attainment. 18 In support of the first point of view is a study by Fisher and Marrow (14) of the effects of induced moods of elation and moods of despondency on verbal responses in a word association experiment. Using reaction time as an index, they found that induced depression was correlated with lengthened reaction time, and induced elation correlated with shorter reaction time. They also found that words with unpleasant tones were associated with the longest reaction times, words with pleasant tones were associated with shorter reaction times, and words without strong affective tones were associated with the shortest reaction times. Another important study was reported by Bousfield and his associates (19). They define mood as a correlate of motive satisfaction and frustration, and undertook the task of testing two related hypotheses: (1) with pleasant mood, there is a facilitation of verbal associative responses having pleasant affective value; (2) with unpleasant mood, there is a facilitation of verbal associative responses having unpleasant affective value. Three groups of normal college subjects totaling 994 participated in this experiment. Three hundred and six Ss gave pleasant associations, 351 Ss gave 19 unpleasant associations, and 337 gave neutral asso ciations for eight periods of two minutes each. Each group was divided into high and low mood on the basis of self ratings. Plots indicating the relationship between the mean accumulative totals of items and the time devoted to their production appeared to con- form to the following equation: n = c (1- e” ), where n = total number of items, c- constant, e = base of natural logarithms, m = constant, and t = time. Differences between production of subjects with high mood and production of subjects with low mood confirmed the authors' hypotheses. In support of the second point of view are Mueller's (14) and Cameron's (1) clinical observa tions. Mueller's thesis is as follows: The conscious search after pleasurable sensations is something which, in addition to the unconscious drives, plays a significant role in civilized man. The more his being civilized inhibits him in satisfying his drives, the more obvious is his search for a substitute which he finds by exposing himself to pleasant stimuli. (14:81) Cameron observes that, with frustration, there is an increase in the organism's pleasurable activ ities such as masturbation, autoerotic activities of all sorts, and phantasy solutions. In essence the organism under conditions of frustration seems to be more preoccupied with pleasant experiences and 20 pleasures than with the unpleasant conditions of frustration. Since mood is defined as a correlate of motive satisfaction, its effects on associative verbal responses may best be tested in relation to motive conditions or motive representations. The study to be reported provided for such a condition through instructions to associate to verbal need stimuli as presented singly or within verbal contexts which may represent happy pleasant situations and unhappy unpleasant situations. From the foregoing discus sion, the Mood Hypothesis may be stated in either of two alternative forms: Hypothesis III: In a continuous free word asso ciation experiment, subjects with more pleasant mood are expected to associate signifi cantly a greater number of pleasant words to verbal need stimuli presented within a pleasant context than do sub jects with more unpleasant mood. Those with more unpleas ant mood are expected to associate significantly a 21 greater number of unpleasant words to verbal need stimuli presented within an unpleasant context than do those with more pleasant mood. Hypothesis IV: In a continuous free word asso ciation experiment, subjects with more unpleasant or unhappy mood are expected to associate significantly a greater number of pleasant words to verbal need stimuli presented within either pleasant or unpleasant contexts than do subjects with happier mood. The Stimulus Hedonic Tone Facilitation Hypothesis This hypothesis is specifically concerned with the facilitation effect of hedonic tones of stimulus words on similar tones of response words in a free word association experiment. Hedonic tones refer to the properties of pleasantness and unpleasantness of words as judged by the subjects. Historically, the stimulus hedonic tone facil itation hypothesis was a specific case of the Freudian 22 repression hypothesis. Experimenters were set to examine the contention that pleasantness as such facilitates related responses, while unpleasantness as such inhibits related responses. Although this was not the intention of the Freudian assertion, such was the first attempt at introducing order and sim plicity to the complex phenomenon of repression and affectivity. Facilitation of responses in the word association experiment was defined in terms of reaction time, word fluency, and congruency. An integration of these indices is necessary to understand the full meaning of the hedonic tone hypothesis and to formulate new experimental operations for testing it. Reaction time index. Since the earliest experi ments on free word association which were initiated by Galton (14), psychologists were aware of the role of hedonic tones of verbal stimuli as determinants of certain characteristics of the associative response. Cordes (14) observes that feelings may accompany either the stimulus word or the response word, or may act as a link between the two. In either case it may result in lengthening the reaction time. The nature of this feeling tone was further clarified by Mayer and Orth (14), Wreschner (14), and Menzerath (14). All of these experimenters found that unpleasant associations tended to show longer reaction times than pleasant toned and indifferent toned associations; and that pleasant toned associations tended to show longer reaction times than indifferent toned associations. When these experiments were followed by Jung's famous diagnostic studies of Freudian complexes, the statement was made that pleasant stimuli facilitate the asso ciative response whereas unpleasant stimuli inhibit the associative response. Hull and Lugoff (23), in a systematic study designed to investigate the validity of many of Jung's complex indicators, concluded that, among normals, long reaction time of over two and six- tenths of a second is a valid measure of complexes. Word fluency index. Many experimenters turned to another form of the word association method, namely, the continuous free word association method, to parallel the original Freudian free association technique. This method allows for more freedom of response and more flexibility in measurement. Units of measurement in the continuous asso ciation method include the number of related responses given within a specific time, the length of time taken to respond with certain numbers of responses, and combinations of both. 24 White (35) chose from some of his experiments on pleasant and unpleasant words ten words judged by the subjects as pleasant, and ten words judged as unpleasant. The twenty words were arranged in random order and were presented auditorially to 136 subjects, sixty-seven men and fifty-six women. He instructed his subjects to write each stimulus word which he presented and to write after it as many words as they were made to think of by the stimulus word. White used the one-half minute blocks of time as units of measurements. Scores were the number of words given within each time unit. He analyzed his data in terms of rates of word fluency regardless of the hedonic tones of associated word responses. On the average, more words were associated to pleasant stimuli than to unpleasant stimuli. The difference between the means was statistically significant at the .01 level. From these results he suggested the possibility that, in experiments on memory and recall, superiority of recalled pleasant words to recalled unpleasant words might be a function of the number of associates available for each kind. Congruency index. Young*s experiments (37) using the congruency index are of decided value in 25 clarifying some of the obscurities of this issue. He instructed 555 subjects to write pleasant and unpleasant words during periods of four minutes each. The same task was presented to another group of fifty- six subjects for thirteen days. His analysis of data yielded the following conclusions: (1) about 55 per cent of all recalled words on the average were pleasant, and 45 per cent were unpleasant; (2) the number of words recalled by the subjects increased by practice, but drill did not change the relative proportions of pleasant and unpleasant words listed; (3) facility in recalling pleasant words and facility in recalling unpleasant words were positively cor related (r » .50). Young*s third conclusion is of special relevance to the discussion of the facilitation function of a set regardless of the special attributes of verbal stimuli. Bousfield's similar experiments (19, 20) proved the validity of this conclusion. He assumed that the subjects have a supply of words that fit the instructions and that the rate of production of words is constant, and therefore the number of words pro duced in successive units of time is proportional to the number of words remaining in reserve. For instance, if the subject knows the names of 100 cities, and if 26 his rate of production is 10 per cent per minute, he will produce ten names in the first minute and nine names in the second minute, which is 10 per cent of the remaining ninety names in his reservoir. In the third minute he will produce eight names, and in the fourth minute he will produce seven names, and so forth. Bousfield and Sedgwick (20) found that this rate of production could be expressed in a general mathematical equation, which we introduced on page 19 of this paper: n * c (1- e-mt). Bousfield applied this reasoning to the study of productions of pleasant and unpleasant associates in response to instructions to produce such items. He used nine groups of college subjects. Four groups of seventy-one subjects listed pleasant items, and five groups of eighty subjects listed unpleasant items, for eight continuous periods of two minutes each. The results confirmed Bousfield's hypothesis. He also found that the number of pleasant associates was significantly greater than the number of unpleasant associates at each interval. He concluded that rather than ascribing a determining power to pleasantness and unpleasantness of stimuli as such, it is more feasible to accept an obvious fact such as was demon strated by Flugel (19), that more time is generally 27 spent in pleasant than in unpleasant pursuits, hence a superior reservoir of pleasant associates. He implied, however, that as much as verbal stimuli activate personal frames of reference, they act as determining sets. In the study to be reported, where verbal need stimuli are presented singly and within other contexts, all of which were given hedonic values by the subjects, the Stimulus Hedonic Tone Facilitation Hypothesis may be stated as follows: Hypothesis V: In a continuous free word associa tion experiment, subjects are expected to associate a greater number of pleasant word responses to a word-need stimulus presented within a pleasant context than to the same word-need stimulus pre sented singly or within an unpleas ant or a neutral context. Simi larly, subjects are expected to associate a greater number of unpleasant word responses to a word-need stimulus presented within an unpleasant context than 28 to the same word-need stimulus presented singly or within a pleasant or a neutral context. The Emotionality Hypothesis Similar to the study of moods and hedonic tones of verbal stimuli is the study of emotionality and emotional tones of verbal stimuli. Emotionality has been traditionally defined as a disturbance which may be detected through measures of autonomic nervous activity such as GSR, pulse rate, and blood pressure. As early as 1912, Birnbaum (14) was credited as having observed that for normal subjects there was no difference in reaction time associated with words considered feeling toned and words considered indif ferent. Among psychiatric patients, emotionally toned words elicited a longer reaction time. His findings indicated that the emotional tone of a stimulus word has influence only on persons with strongly accentuated affectivity. Experimenters who investigated Jung's complex indicators, such as Smith (14), Washburn (34), and many others (14), leaned toward the hypothesis that when emotional words reflect the emotional states within the organism, an emotional response occurs. For this reason most subsequent experiments used a PGR index for determining emotionality in relation to their selected verbal stimuli. There was general agreement (14) that where there was an arbitrary judgment of emotional stimuli, results varied and usually failed to demonstrate the consistent facil itation effect of emotionality on verbal responses. Confusion between unpleasantness and emotionality prevailed during the early periods of investigating the effects of emotionality by the use of the word association method. As a result, many experimenters referred to their PGR index as an indicator of unpleasantness. In subsequent studies, the following criteria were used to distinguish between emotional and unemotional subjects. One criterion was psy chiatric diagnosis of subjects characterized by nervousness, excitability, and strong variation of moods. A second criterion was self rating on a scale of emotionality. A third criterion was a high score on a personality test of emotional instability and general maladjustment. Recent investigations make use of combinations of these methods. Evidence that psychological evaluations are correlated with physio logical measures such as PGR is provided by Jacobs (26), who demonstrated that even within one group, such as college students, those who reacted to emo tional words with larger GSR deflections, longer RT, and with blocking in forming associations to emotional words have shown significantly a more inferior pattern of adjustment on the Guilford GAMIN. Jacobs suggests caution, however, in interpreting the relationship of failure in learning tasks and degree of personality maladjustment in the absence of a measure of ability. It is on this basis that in the present study, in addition to the aforementioned psychological criteria, two measures of ability were used to control for the effects of emotionality on production rate and emo tional tones of associative responses. Since emotionality is generally considered a function of many operative factors,. most of which are motivational in nature, it is plausible to test the Emotionality Hypothesis in relation to need stimuli in the following form: Hypothesis VI: In a continuous free word asso ciation experiment, subjects with higher ratings of emotion ality, reacting to word-need stimuli, should associate significantly a greater number 31 of emotional words with greater degree of intensity than subjects of lower ratings of emotionality reacting to the same stimuli within the same time units* The Stimulus Emotional Tone Facilitation Hypothesis It was indicated earlier that most experiments which investigated the effects of stimulus hedonic tones on associative responses were primarily designed to test the effects of emotional tones of verbal stimuli on associative responses. At this time, the unpleasant was considered emotional. However, most findings indicated that both pleasant and unpleasant stimuli produce greater GSR deflections and longer RT than indifferent stimuli in general. Limiting emotionality to the unpleasant was a perplexing con sideration. In addition, GSR and RT measures showed a wide range of variations that impelled the experi menters to carefully consider two variables: (1) the relative degree of stimulus emotional tones, and (2) the necessity of subjective judgments of stimuli, on intensity scales. Ratings of intensity of emotional tones were used by investigators as early as 1929, when Chaney 32 and Lauer (14) found it necessary to introduce such indices. The findings of the very many experiments that followed seem to support the hypothesis that the intensity of the stimulus is more important than its hedonic tone in facilitating the emotional associative response. In all these experiments intensity was determined by either the magnitude of GSR deflections or more frequently by the individual ratings of selected stimuli. The general agreement that arbitrary selection of emotional stimuli was responsible for inconclusive results was challenged by Barret (14). She selected words from lists used by previous experi menters as P, U, and I, and asked a group of normal subjects to rate these words on the same dimension. Agreement between average judgments made by experi menters was high. However, because of a wide range of judgments observed in many instances, it was suggested that individually obtained judgments are more effective in securing relevant results. To test the Stimulus Emotional Tone Facilita tion Hypothesis, in the present study, it was therefore necessary to secure ratings of verbal stimuli as well as verbal responses on an emotional scale, and to state this hypothesis as follows: 33 Hypothesis VII; In a continuous free word asso ciation experiment, it is expected that subjects should associate significantly a greater number of more intense emotional word responses to verbal stimuli with higher ratings of intensity of emo tional tones than to verbal stimuli with lower ratings of intensity of emotional tones. CHAPTER III INSTRUMENTS AND SUBJECTS The Basic Instrument Fifteen one- and three-word stimulus items were presented in four by eleven booklet form, each on a separate page. These items were chosen according to the following criteria. 1— Representation of needs. Three-word need stimuli were arbitrarily selected by the experimenter from Murray's list (11) of need referents. These words were "nourishment," "recognition," and "inter course," which represent the needs for nutrition, recognition, and sex, respectively. 2— Representation of family context. The family context was represented by the two words "home-parents." Such context, as well as all other verbal contexts which were used in this research, refers to two words which precede the word-need stimulus. The word-need stimulus may be presented singly or may be preceded by such a verbal context. Since the two words which constitute the family context were made up of four and seven letters consecutively, all words which constructed the other verbal contexts were made up 34 35 of four and seven letter words also. Thus, a com parison between responses to word-need stimuli presented singly and responses to word-need stimuli presented within a verbal context was made possible. 3— Representation of a pleasant context. For the verbal context to be pleasant, the two consti tuting words must be judged by the respondents as pleasant. In this experiment the two words were "rosy-smiling." These two words were selected as pleasant words from a list of words given by hospital and college subjects in a preliminary study. 4— Representation of an unpleasant context. Similar to the selection of the words which con stituted the pleasant verbal context, two words were selected to represent the unpleasant verbal context. These two words were Mblue-unhappy." 5— Representation of a neutral context. The two words which constituted the neutral context were selected from a list of nonsense syllables used by Osgood (13)o These two words were "XOMF-KTQHAFX." These five criteria were thus met by the following fifteen stimulus items: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 36 Nourishment: representing the need for nutrition. Recognition: representing the need for recognition. Intercourse: representing the need for sex. Rosy-smiling-nourishment: representing the need for nutrition within a pleasant context. Rosy-smiling-recognition: representing the need for recognition within a pleasant context. Rosy-smiling-intercourse: representing the need for sex within a pleasant context. Blue-unhappy-nourishment: representing the need for nutrition within an unpleasant context. Blue-unhappy-recognition: representing the need for recognition within an unpleasant context. Blue-unhappy-intercourse: representing the need for sex within an unpleasant context. XOMF-KTQHAFX-nourishment: representing the need for nutrition within a neutral context. XOMF-KTQHAFX-recognition: representing the need for recognition within a neutral context. 12. XOMF-KTQHAFX-intercourse: representing the need for sex within a neutral context. 13. Home-parents-nourishment: representing the need for nutrition within a family context. 14. Home-parents-recognition: representing the need for recognition within a family context. 15. Home-parents-intercourse: representing the need for sex within a family context. These fifteen items were presented in a booklet form, each on a separate page. They were randomized and ordered in two different arrangements: 1, 14, 6, 10, 2, 15, 7, 11, 3, 4, 8, 12, 13, 5, 9; and 6, 14, 1, 15, 2, 10, 3, 11, 7, 12, 8, 4, 9, 5, 13. Subjects were instructed to respond to each item for two minutes Definition and Measurement of Variables Stimulus hedonic tone. This variable is defined as a judgment made by the respondents to the above mentioned verbal stimuli on the Pleasant-Unpleasant Scale. The judgments are expressed in numerical values assigned to each of the words which constitute the verbal stimuli. A value of P-2 is assigned to the word which is judged as very pleasant; a value of P-l is assigned to the word which is judged as fairly pleasant 38 a value of (0) is assigned to the word which is Judged as neutral. A value of U-l is assigned to a word which is judged as fairly unpleasant; a value of U-2 is assigned to a word which is judged as very unpleas ant. Stimulus emotional tone. This variable is defined as a judgment made by the respondents of the above mentioned stimulus items on the Emotional- Unemotional Scale. Judgments are expressed in numerical values assigned to each of the words which constitute the verbal stimuli. A value of E-3 is assigned to the word which is judged as very emotional; a value of E-2 is assigned to the word which is judged as fairly emotional; a value of E-l is assigned to the word which is judged as slightly emotional. A value of E-0 is assigned to the word which is judged as unemo tional. Unlike the scale of pleasantness, there was no provision for neutrality of judgments since a word can only be judged as either emotional or unemotional. Mood. Mood is defined as a judgment made by each respondent of his own feelings on the Happiness Scale. A value of H-3 is assigned to the feeling of being very happy; a value of H-2 is assigned to the feeling of being fairly happy. A value of H-l is 39 assigned to the feeling of being fairly unhappy; and a value of H-0 is assigned to the feeling of being very unhappy. To provide for the experimental variation of mood, two groups of subjects were selected so as to differ in their ratings of their moods: hospital and college subjects. The differences between these two groups are reported in Chapter V. Emotional stateo Emotional state is defined as a judgment made by each respondent of his own feelings of emotionality. A value of E-3 is assigned to the feeling of being very emotional; a value of E-2 is assigned to the feeling of being fairly emo tional; a value of E-l is assigned to the feeling of being slightly emotional. A value of E-0 is assigned to the feeling of being unemotional. To provide for the experimental variation of emotionality, two groups of subjects were selected so as to differ in their ratings of their emotionality. Differences between these two groups are reported in Chapter V. An additional measure was indicated by a score obtained on the emotional stability scale of the Guilford-Zimmerman Temperament Survey, which is described in Appendix C. 40 Need satisfaction* This variable is defined as a judgment made by each respondent on the Satisfaction-Frustration Scale of the three needs of nutrition, recognition, and sex. A value of S-4 is to be assigned to a judgment of feeling very satisfied; a value of S-3 is to be assigned to a feeling of being fairly satisfied. A value of S-2 is to be assigned to a judgment of feeling fairly frustrated; a value of S-l is to be assigned to a judgment of feeling very frustrated. Family context. This variable is defined as a judgment made by each respondent of the two-word family context "home-parents" in terms of pleasant ness and emotionality, Pleasant-Unpleasant, and the Emotional-Unemotional Scales which were described above. Definition and Measurement of Criterion Variables Facilitation of production of word responses. This variable is defined as the difference between the number of associated words written by the sub jects in response to one stimulus item within two- minute intervals, and the number of associated words to another stimulus item within the same time unit. 41 Facilitation of hedonic tones of verbal responses. This variable is defined as the differ ence between the sum of values of judgments of pleasant tones of word responses given to one stimulus item, and the sum of values of judgments of pleasant tones of word responses to another stimulus item within identical time units. The scale of judgments of pleasant tones of word responses is the same as that used for judgments of word stimuli which is described above. Facilitation of emotional tones of verbal responses. This variable is defined as the differ ence between the sum of values of judgments of emotional tones of word responses given to one stimulus item, and the sum of values of judgments of emotional tones of word responses given to another stimulus item within identical time units. The scale of judgments of emotional tones of word responses is the same as that used for judgments of word stimuli which is described above. The Subjects A total of 121 male subjects participated in this experiment. Fifty-two subjects were selected from Fresno State College students, and sixty-nine 42 subjects from Atascadero State Hospital patients. Fresno State College students were taken from four classes of students studying mental hygiene, theories of personality, and criminology during the summer and fall semesters of 1959. One summer class was taught mental hygiene by Dr. Harrison Madden on the Visalia Campus. Another summer class was taught criminology by Dr. William Dienstein on the Fresno Campus. The fall classes were taught mental hygiene and theories of personality by the experimenter on the Fresno Campus. Males and females were permitted to participate in the experiment; however, only male subjects were retained in the experimental sample. Atascadero State Hospital patients were selected primarily on the basis of three criteria: college education, adult age, and history of sexual maladjust ment. The experimenter gave a lecture to the entire patient body on research and treatment of sexual maladjustment, after which he asked for volunteers who were of adult age and who had had a college edu cation. Sixty-nine subjects fulfilled these require ments, and all were accepted as participants in the experiment. Their enthusiasm and cooperation were made known to the entire hospital personnel and patient body through their local newsletter. 43 Short informal interviews were conducted with each of the subjects. The interviews lasted from five to twenty minutes, depending upon the individual's interest in the experiment. These interviews were conducted after the completion of the experiment and dealt primarily with the patient's general feeling about the hospital and about his particular sexual problem. Almost all of the patients complained mildly about the quality and ta^te of their food. Three voiced dissatisfaction about providing the personnel with tastier meals than those served to the patients. Almost all of them admitted some per sonal sexual release through masturbation or occasional homosexual and rare heterosexual contact. Only three patients voiced their desire to remain in the hospital. All three were personal friends: a former clergyman, a former teacher, and a former male secretary. The latter two were married but preferred homosexuality. The rest of the patients voiced their dissatisfaction with their segregation from the community; with their inability to control their sexual impulses. They expressed their need for psychiatric help. No such interviews were conducted with the college students, who were assumed to be normals in terms of their social and academic functions. 44 To provide adequate controls for the experiment, both groups of hospital and college subjects were compared with reference to age, education, intelli gence, and word fluency. The results of these comparisons are reported in Chapter V. The following chapter presents the experimental methods and pro cedures which were followed in conducting this study. CHAPTER IV EXPERIMENTAL PROCEDURES Order of Event s All subjects In both college and hospital groups were given seven tasks in the following order: (1) the Word Fluency Test, (2) the Need Word Association Task, (3) Mood Rating Scale, (4) Emotionality Rating Scale, (5) Need Satisfaction Rating Scale, (6) the Shipley- Hartford Living Institute Scale, and (7) the Guilford- Zimmerman Temperament Survey. These tasks were given in two periods, one day apart, except in the case of twenty college subjects who completed the tasks in two periods, two weeks apart. The first five tasks were given during the first period; the latter two were given during the second period. Instructions Preliminary orientation. Both groups of sub jects, who were described in Chapter III, were given a two-minute orientation period designed to achieve cooperation in the experiment. They were told that they would be engaged in a scientific research designed to understand more clearly the psychological problems 45 46 of the patients, and that the results might be helpful to them and to many others in the future. Both groups were told the approximate time involved, and were also told to feel free not to participate if they decided not to. They were assured of their anonymity and the fact that all information would be held confidential. All subjects were willing to participate in the experi ment . The Word Fluency Test Form A. This was the first task which was given to all subjects. The test is composed of two items to be completed within two minutes each. A sample of the test, a description and evaluation of its validity and reliability, are attached to Appendix A. The Need Word Association Task. The Need Word Association Task is the main instrument which was constructed to test the experimental hypotheses. It was explained to the subjects as follows: This is an experiment in Free Association. You are asked not to write your name so that you may feel completely free to write everything that comes to your mind. There are fifteen items, each printed on a separate page. Each item is made up of one word or of three words. You are asked to read each item carefully and begin to write down several different words which you associate freely to the last word. If you have an item made up of one word, your associations 47 will be made to this one word. If you have an item made up of three words, read the first two words and begin to associate to the third word or the last word. Be sure to read all the words which are printed at the top of each page. Write down single separate words in the two columns provided on each page. Do not repeat your words on any one page. You may repeat them, however, on different pages. Do not write sentences, such as "I am a sailor." Write single different words. Do not stop to think, evaluate, criticize, or select your words. Write as freely as you can, and as rapidly as you can. Write whatever words that come to your mind regardless of logicality, meaningfulness, obscenity or absurdity. Write all the different words that come to your mind. For example, if you read at the top of the page the word book, you may start writing words such as read, write, you, pen, penis, etc. A second example is— if you read market, shop. card, you may write game, club, expensive, cheap, whore, etc. Some of the words may have no meaning. Read them just the same and write whatever individual words that come to your mind after reading the last word. A third example is— if you read SMTO-WPQFSY- book, you may write buy, fiction, enjoy, best, badj" etc. Do you have any questions? You will have two minutes for each page. 1 shall give you the signal to start and to stop. The fifteen items should be completed in about thirty minutes. Another thirty or forty minutes will be needed for other tasks after the first part of this experiment. Now begin by filling in the information items which are shown on the cover of the booklet. The information which they were asked to pro vide consisted of identifying themselves as to age, sex, education, and a number, such as their home numbers, which would be repeated on all tasks* There was also additional information that was not used in the final analysis, such as the number of courses and the number of books that they may have read in psy chology. Instructions for ratings of mood and emotion ality. Below this information the subjects were asked to rate themselves on the Emotionality and the Mood Scales, which were described in the preceding chapter. Categories of each scale were either written on the blackboard or presented in large capital letters on a sheet of paper. After this, they were asked to begin at a signal and to start on the first item. Rating responses. When they completed all of the fifteen items, given two minutes for each item, the subjects were asked to rate all the stimulus and response words on an Emotionality Scale. A value of E-3 was to be assigned to a word which was judged by the subject as very emotional; a value of E-2 was to be assigned to a word which was judged as fairly emotional; a value of E-l was to be assigned to a word which was judged as slightly emotional. A value 49 of E-0 was to be assigned to a word which was judged by the subject as unemotional. When they completed their ratings of the words on the Emotionality Scale* they were asked to rate the same words again on the Pleasant-Unpleasant Scale. A value of P-2 was to be assigned to a word which was judged by the subject as very pleasant; a value of P-l was to be assigned to a word which was judged as fairly pleasant. A value of P-0 was to be assigned to a word which was judged by the subject as neutral. A value of U-l was to be assigned to a word which was judged by the subject as fairly unpleasant; a value of U-2 was to be assigned to a word which was judged as very unpleasant. Instructions for ratings of need satisfaction. After completing the rating of words on the Pleasant- Unpleasant Scale) the subjects were asked to rate their three needs of nourishment. recognition, and sex on a scale of satisfaction. They were asked to assign one of the following categories to each of the three needs: very satisfied, fairly satisfied, fairly frustrated, very frustrated. Instructions for the Shipley-Hartford Living Institute Scale. The Shipley-Hartford Scale was 50 given to the subjects during the second period which, as was previously mentioned, was one day apart from the first period for all subjects other than twenty college students, who were not able to take this test for an interim of two weeks. A description and evaluation of this test will be found in Appendix B. The test consisted of two parts: the first on vocabulary, and the second on abstract reasoning. Subjects were allowed ten minutes for each part. However, almost all subjects in both groups finished the first part on vocabulary within three to five minutes. As to the second part, on abstract reason ing, all subjects found it necessary to use all the time allotted to them. Some subjects raised ques tions about this part of the test, others voiced their fear of having done badly, but most of the subjects seemed to have considered it as just a part of the research. Instructions for the Guilford-Zimmerman Temperament Survey. This was the last test given to all subjects. A description and an evaluation of this test are included in Appendix C. The main pur pose for giving this test was to find out whether there were differences between the college group and 51 the hospital group on the emotional stability scale and in the general pattern of adjustment. Statistical Treatment of the Data All tests were scored manually, following the standard procedures required by the manual's instruc tions or by the specifications of the experiment. To facilitate the statistical treatment of the data, recording sheets were prepared for tabulating all numerical values in appropriate columns for each of the experimental groups. Some of these data were transferred to IBM cards for the operation of IBM computers at the Western Data Processing Center, University of California at Los Angeles. Description and evaluation of the statistical programs and tests will be found in Appendix D. CHAPTER V RESULTS The experimental data consisted of four sets of measurements for each of the two experimental groups of subjects: measures of controlled variables, ratings of subject set characteristics, ratings of stimulus variables, and measures of response variables. Measures of controlled variables consisted of four scores of age, education, intelligence, and word fluency. Ratings of subject set characteristics consisted of fifteen scores of self ratings of mood, self ratings of emotionality, scores of ten different scales of the Guilford-Zimmerman , • Temperament Survey, self ratings of satisfaction of the needs for nutrition, recognition, and sex. Ratings of stimulus variables consisted of thirty-two scores of ratings of emotional and pleasant tones of verbal stimuli. Measures of response variables consisted of eleven scores of frequency of word responses; frequency of word responses judged as pleasant, unpleasant, neutral, emotional, and unemotional; sums of weighted scores of word responses judged for intensity of pleasant, unpleasant, and emotional tones. A total of sixty-two scores was obtained for each subject in each of the two experimental groups. 52 53 Data relevant to the hypotheses are presented in table form and will be discussed in the following order: (1) reliability of individual judgments, (2) measures of controlled variables, (3) ratings of sub ject set characteristics and stimulus variables, (4) measures of response variables. Reliability of Individual Judgments Table 1 presents the intercorrelations of repeated judgments of emotional tones of thirty-nine stimulus words presented singly or in a group of three words. Correlations ranged from .20 to .86 with a median of .67 for college subjects; and ranged from .29 to 1„00 with a median of .66 for hospital patients. Table 2 presents the intercorrelations between repeated judgments of pleasant tones of the same thirty-nine stimulus words. Correlations ranged from .12 to .84 with a median of .53 for college students; and ranged from .02 to .86 with a median of .55 for hospital patients. Measures of Controlled Variables Table 3 presents results of measuring the controlled variables for both college and hospital subjects. Table 1 Correlations Between Repeated Judgments Of Emotional Tones Of Word Stimuli I 2 3 4 5 6 7 8 9 10 11 D C 0 E B C D E B C D E B C B C B C B C B C B C B C B C A .43.36 .37.58 .59.82 .60.76 .64.70 .86.83 •53 .63 -^>1 » Co ro .80 .53 .70 .42 .67 .56 .74 .51 .66 .56 .58 .49 < n Ui H B •83 .83.73 .48 .46.53 .64 .75.67 .81 .78 *65 .45 .67 .64 .21 .20 ' (5 U Id Id C •79.72 .59.61 .68.73 -1 □ O 3 U £0 •7i .70 .83 A .50.59 .66.63 .45.54 .65.73 .47.58 .52.50 .69 .69 .67 .59 .81 .63 •59 .70 .80 .73 .64 .64 .70 .70 .67 .66 J </) < H 0 .5 b .66.51 .29 .42.59 .73 .86.80 • 70 .80 .65 •53 .90 .75 1 .00 1.00 1- o — u C .70.55 .53.58 .73.76 to m O D D .67 .87 .83 I to No t e: College Subjects: n = 52 ! with df = 50, R = .27 at p ■ .05 and r - .35 at p = .01 Hospital Subjects: n = 69; with df = 67, r = .24 at p = .05 and r - .31 at p = .01 Key: I- Nourishment: a) presented singly, b) presented as home-parents-nourishment, c) presented as rosy-smiling- nou rishment, d ) presented as blue-unhappy-n o u r1s hm ent, e ) presented as xomf-Ktqhafx-n o u rISHMENT. 2- Recognition: a) presented singly, b) presented as home-parents-recogn1t 1 on, c) presented as rosy-smiling- RECOGNITION, d ) PRESENTED AS BLUE-UNHAPPY-RECOGNITI ON, e ) PRESENTED AS XOMF—KTQHAFX-RECOGNITI ON. 3- Intercourse: a ) presented s in gly, b ) presented as home-parents-intercourse, c) presented as rosy-smiling- INTERCOURSE, d ) PRESENTED AS BLUE-UNHAPPY, INTERCOURSE, E) PRESENTED AS XOMF-KTQHAFX-INTERCOURSE. 4- Home: a) presented as home-parents-nourishment, b) presented as home-parents-recognition, c) presented as HOME-PARENTS-INTERCOURSE. 5“ PARENTS: PRESENTED AS IN NUMBER 1 +. 6- Ro s y: a ) presented as rosy-smiling-nou rishment, b ) presented as rosy-smiling-recognition, c) presented as ROSY-SMILING-INTERCOURSE. 7“ SMILING: PRESENTED AS IN NUMBER 6. 8- Bl u e: a ) presented as blue-unhappy-nourishment, b ) presented as blue-unhappy-recognition, c) presented as SLUE-UNHAPPY-lNTERCOURSE. 9“ UNHAPPY: PRESENTED AS IN NUMBER 8. [0- XOMF: a) PRESENTED AS XOMF-KTQHAFX-NOUR1SHMENT, b) PRESENTED AS XOMF-KTQHAFX-RECOGNITI ON, c) PRESENTED AS XOMF-KTQHAFX-INTERCOURSE. II" KTQHAFX: PRESENTED AS IN NUMBER 10. Ol Table 2 Correlations Between Repeated Judgments Of Pleasant Tones Of Word Stimuli 1 2 3 it 5 6 7 8 9 10 11 B C D E B C D E B C D E B c b c B C B C B C B C B C B C A .55.68 .28.1+8 .27.1+6 . Ill .55 .8l.6| .78.75 .60 .68 •57 -53 •53 .56 .36 .25 .28 .21 .32 .38 .29 .38 .19 .ito LI H 0 U S .70 .21.39 .36 .50.55 .53 -72.611 •73 .52 .58 .50 .H7 •70 .8ii •53 tJ U3 *J C .32.3^ .63.113 .66.81 J Q O D O <0 D .12 •73 .83 A .55.56 .65.78 .60.51 .66.61 .57.55 .20.31 •59 .62 .5ll .38 •55 .56 .60 .69 .116 .26 •77 .86 .81 .61 .69 .[It (0 < H B .1+7 .65.60 .38 .L8.L5 .67 .111.53 .68 .80 .113 .3 8 .02 .8ii .68 . l6 : - U C L C -^3-73 .29.119 (0 CD O D D .80 •57 .112 1 ; X (/> ;Note: College Subjects: n = 52; with df = 50, r = .27 at p = .0^ and r = .35 at p = .01 Hospital Subjects: n = 69; with df = 67, r = .2ll at p = .05 and r = .31 at p = .01 Ke y: I- Nourishment: a ) presented sin gly, b ) presented as home-parents-nourishment, c ) presented as rosy-smiling- NOURISHMENT, D) PRESENTED AS BLUE-UNHAPPY-NOURISHMENT, E) PRESENTED AS XOMF-KTQHAFX-NOURISHMENT. 2- Recognition: a) presented singly, b) presented as home-parents-recognition, c) presented as rosy-smiling- RECOGNIT1 ON, D) PRESENTED AS BLUE-UNHAPPY—RECOGNITI ON, E) PRESENTED AS XOMF-KTQHAFX—RECOGNITI ON. 3" Intercourse: a) presented singly, b} presented as home-parents-intercourse, c) presented as rosy-smiling- INTERCOURSE, o) PRESENTED AS BLUE-UNHAPPY-INTERCOURSE, E) PRESENTED AS XOMF-KTQHAFX-INTERCOURSE. It- Ho m e: a ) presented as home-parents-nourishment, b ) presented as home-parents-recognit ion, c ) presented as HOME-PARENTS—INTERCOURSE. 5" PARENTS: PRESENTED AS IN NUMBER k. 6- Rosy: a) presented as rosy-smiling-nourishment, b) presented as rosy-smiling-recognition, c) presented as rosy-smiling-intercourse. 7" Smiling: presented as in number 6, 8- Bl u e: a ) Presented as blue-unhappy-nourishment, b ) presented as blue-unhappy-recognition, c ) presented as blue-unhappy-intercourse, 9" Unhappy: presented as in number 8. Wi 10- XOMF: A) PRESENTED AS XOMF-KTQHAFX-NOUR[SHMENT, b) PRESENTED AS XOMF-KTQHAFX—RECOGNITI ON, c) PRESENTED AS W XOMF-KTQHAFX-INTERCOURSE. 1|- KTQHAFX: PRESENTED AS IN NUMBER 10. Table 3 D ifferences Between College And Hospital Subjects In measures Of Controlled Variables Variables College Hospital t Ko l-Smirnov Chi Sq . S ignificance Mean Mean Statistic ESTI M Level Age 26.3 32.1 4 .|o .01 Education (y e a r s) 15-6 13-3 3.20 .01 Intelligence (WAIS IQ Est) 114 1 12 1.76 N .S Shipley-Hartford Scale Vocabulary Score 31.25 31 -55 - 13.27 2.087 .4o Abstract Reasoning 29,98 27.57 - I 9.50 4.514 •20 Total Score 61.23 59.06 - 22.17 5.794 .10 Word Fluency Test Part I 19.20 20.20 + | 6 .|6 3.099 .30 Part 11 21 -35 22.96 + 16.19 3 . n o .30 Total Score 40.5o 43.16 + 2 2 .4 | 5-995 .10 Note: For i: test, df =II9 ,for chi square, df = 2. 57 Age♦ The ages of the subjects ranged from eighteen to forty-nine. However, hospital patients were significantly older than college students. Education. Both groups of subjects had received college education ranging from one to eight years. College students, however, had significantly more years of education than hospital patients. Intelligenceo Converting the Shipley-Hartford scores to equivalent WAIS scores by the use of Sines and Simmons's table (33) resulted in a range of I.Q. scores from 96 to 126, with a mean of 114 and a standard deviation of 6.9 for college students. For hospital subjects I.Q. scores ranged from 97 to 128, with a mean of 112 and a standard deviation of 12.5. Differences between the two groups were not significant. Word fluency. Results of the word fluency test showed no significant difference between college and hospital subjects. Ratings of Subject Set Characteristics and Stimulus Variables Tables 4, 5, 6, 7, and 8 present the results of measuring differences between college and hospital subjects with respect to ratings of subject set char acteristics and stimulus variables. Table 4 D ifferences Between College And Hospital Subjects |n Measures And Ratings Of Subject Set Characteristics College Mean HOSPITAL Mean Ko l-Smirnov STAT 1 ST 1C Chi Sq . Es t. P , (df = 2) Mood (Self ratings) 3.06 2.80 - 19.23 4.386 .2000 Emotionality (self ratings) GU LFORD-ZIMMERMAN TEMPERAMENT SURVEY 2.69 3-12 + 24.67 7.216 .0500 General Activity(En e r g y)-Slowness G |6. |2 17.^5 + 27.70 9.103 .0200 Restraint - Impulsivity R 18.85 17.25 - 23.86 6.750 .0500 Ascendence - Submissiveness A 15.27 16.12 + 30.13 IO.766 .0050 Sociability - shyness S 17.27 20.14 + 43.65 22.594 .0005 Emotional Stability - Instability E 18.27 17 • 64 - 22.85 6.195 .0500 Objectivity - Subjectivity 0 19.98 I8.96 - 12.74 1 .924 .4000 Friendliness - Hostility F 16.92 17.38 + 13.32 2.(05 .4000 Reflectiveness - Unreflectiveness T 19-77 18.12 - 32.05 12.184 .0050 Cooperativeness - Criticalness P 19.67 20.13 + 16.56 3.250 .2000 Masculinity - Femininity Need Satisfaction (self ratings) M 20.31 19.71 “ 1 2.4o 1.824 .5000 Nutrition 3.67 3.26 - 36.37 15■690 .0005 Recognition 3-13 2.99 - 1 1 . 9 6 1.695 .5000 Sex 3.08 2 .( 4 - 45.04 2 4 .o 6o .0005 CO 59 Table 5 D ifferences Between Ratings of Need Stimuli Group + R-N P + 1 -N Judgments ! p + 1 OF -R Pleasant p + Tones P-U ? P-X + - p + x - u p College Subjects 9 5 N . S 22 6 .01 |6 3 .01 50 1 .01 to 0 .01 23 L .01 Hosp1tal Subjects 3 13 H .S ^0 2 .01 to 2 .01 65 I .ot 66 2 .01 18 12 N .3 Group 4- Judgments R-N P 1 OF 1 + Emotional Tones l- N I-R p + p 4- Judgments R-N p of Need-Satisfaction S-N S-R + - p + - p College Subjects 22 2 .01 39 2 .01 33 6 .01 32 8 .01 3 23 .01 13 \ b N .3 Hospital Subjects in k .01 59 1 .01 to 2 .01 6 22 .01 3 53 .01 7 to .01 Key: R: Recognition N: Nourishment I : Intercourse P: Pleasant context U: Unpleasant context X: Neutral context S: Sex + : Number of cases v/here R exceeds N in R-N , I exceeds N in l - N , etc. Number of cases where R is less than N in R-N, I is less than N in I-N , etc. p : Level of significance as indicated by the tables of sign t e s t. TALLE 6 D ifference Between Hospital And College Subjects In Their Ratings Of Pleasant And Emotional Tones Of The Words "HOME" And "PARENTS" Context Hospital Mean College Mean Kol-Sm1RNOV STAT 1 ST 1C Chi Square Es t i m. p. (DF = 2 ) Pleasant Tone: "Home" Home-Parents-Nour ishment 3.17 3-12 + 5.99 ■**3 .9000 Ho m e-Pare nts-Recognition 3.21+ 3 .0 6 + 9.67 1 . 1 1 .6000 Home-Pa re nts-Intercourse 3.30 3 .08 + 1 9.31 * 3+.U6 .2000 Pleasant To n e: "Parents" Ho m e-Parents-No u rishment 3.17 3-10 + 7. 1 + 2 .66 ■ .8000 Home-Parents-recogn IT 1 O N 3.26 2.96 - 13.52 2.18 .1+000 Ho me-Pare nts-Intercourse 3.2!+ 2.98 - 23.19 6.1+2 .0500 Emotional To n e: "Home" Ho me-Pare nts-Nourishment 2.0 6 1-35 + 1+6.26 25.58 .0005 Ho me-Pare nts-Recognit i on 2.09 1-77 + 13.68 2.23 .1+000 Home-Pa re nts-|ntercourse 2.06 1-73 + 16.10 3. to .3000 Emotional To n e: "Parents" Ho m e-Pare nts-Nourishment 2.17 1 .8 | + I 1 * . 18 2.ho .1+000 Ho me-Pare nts-Recognition 2.21 1.90 + 17.36 3 .6 0 .2000 Ho m e-Pare nts-!ntercourse 2.lk 1.83 + [7 .0 3 3.1+6 .2000 < T > O Table 7 Differences Between College And Hospital Subjects In Their Ratings Of Emotional Tones Of Verbal Stimuli Verbal Stimuli College Mean Hospital Mean Kol-SmiRNOV STATISTlc Chi Square Es t i m. P. (df = 2) Nourishment . 3 | 1 .13 + 18.65 4.123 .2000 Ho me-Pa re nts-Nourishment U.27 5.58 + 1 )6.57 25.726 .0005 ROSY-SM1L 1NG-NOUR1SHMENT 3.17 *•77 + 50.33 30.050 .0005 Blue-Unhappy-Nourishment 3.02 4.32 + 42.17 21.091 .0005 Xomf-Ktqhafx-Nourishment 1 -33 1 .3^ + 10.51 1 .309 .6000 Recognition 1.25 1 .83 + 34.48 14.098 .0010 Ho m e-Parents-Recognition 5-35 6.17 + 16.78 3.338 .2000 Rosy-Smiling-Recogn1tion 3.56 4 .12 + 12.68 1 .907 .1)000 BluE-Unhappy-Recognition 3.02 4.26 + 44.09 23.058 .0005 Xo mf-Ktqhafx-Recognition : 1 .60 1.80 + 29.65 10.1)30 .0100 1NTERCOURSE : 2.00 2.74 + 47.49 26.752 .0005 Home-Pare nts-)ntercourse : 5.67 6.52 + 19-93 4.710 .1000 Ro sy-Sm i lIng-Intercourse : 4.8l 5.46 + 13 . So 2.19I) .1)000 Blue-Unhappy-Intercourse : 4.27 4.99 + 33.03 12.937 .0050 Xo m f-Ktqhafx-|ntercourse : 2.19 2.29 + 9 -31 * 1.033 .6000 Table 8 D ifferences Between College And Hospital Subjects In Their Ratings Of Pleasant Tones Of Verbal Stimuli Stimulus Hospital College Ko l-Smirnov Chi Square p . Mean Mean Statistic Estim. (df = 2) Nourishment Home-Parents-nourishment Ro s y-Smil ing-Nourishment Blue-Unhappy-Nourishment Xo m f-Ktqh afx-Nourishment Recognition Ho m e-Pare nts-recognition Ro s y-Smil ing-Recognition Bl u e - U n h a p p y - R e c o g :i I T I on Xomf-Ktqhafx-Recogn1tI on INTERCOURSE Home-Parents-!ntercourse ROSY-SMILING-i NTERCOURSE Blue-Unhappy-|ntercourse XOMF-KTQHAFX-INTERCOURSE 3.20 9.38 9.52 6.94 7.20 3.39 9.67 9.45 7.07 7.22 3.00 10.19 10.19 7.35 7 - 75 2.88 + 2 4 . 19 6.9I} .0500 9.10 + 14.13 2.37 . 4ooo 8.67 + 31 .44 11.72 .0050 5-83 + 46.52 25.66 .0005 6 .6 7 + 27.06 8.69 .0250 3.00 + |4.55 2.51 .3000 9.27 + 17.75 3 -7 ^ .2000 3.77 + 27.56 9.01 .0250 ... .'0 + I0 .9 0 4.23 .2000 6.92 + 13-10 2.03 .4ooo 3.23 + 53-79 34.32 .0005 9.17 + 55-21 36.16 .0005 9.08 + 55.69 36.78 .0005 6.85 + 22.80 6 . 16 .0500 7.44 + 38.35 17.44 .0005 9) to 63 Mood, College students rated themselves as happier than hospital patients; however, the difference was not significant (.20 level). Emotionality. Hospital patients rated them selves significantly more emotional than did college students. They also scored significantly lower on the emotional stability scale of the Guilford-Zimmerman Temperament Survey. Guilford-Zimmerman Temperament Survey. Hospital patients scored significantly lower than college students on the restraint, emotional stability, and reflectiveness scales. They scored significantly higher than college students on the general activity, social boldness, sociability, and cooperativeness scales. They did not differ significantly from college students on the objectivity, friendliness, and mascu linity scales. Need satisfaction. Table 4, page 58, shows that hospital subjects rated themselves significantly lower than college students with respect to the satisfaction of the needs for nourishment and sex. No significant difference was found for their rating of satisfaction of the need for recognition. 64 Table 5, page 59, shows that for college sub jects the need for sex was significantly less satisfied than the needs for nutrition and recognition. The need for recognition was significantly less satisfied than the need for nutrition. For hospital patients the needs for sex and recognition were significantly less satisfied than the need for nutrition; however, the satisfaction of the need for recognition did not differ significantly from the satisfaction of the need for sex. Family context. Table 6, page 60, shows that the word "home" was rated significantly more emotional by hospital patients than by college students in asso ciation with the word nourishment, and not so in asso ciation with the words recognition and intercourse. Also there was no significant difference between ratings of the pleasant tone of the word "home" by college and hospital subjects. As a general trend, however, hospital patients rated the word "home" as more emotional and more pleasant than did the college students. No significant differences were found between ratings of the emotional and pleasant tones of the word "parents" by either group of subjects. As a 65 general trend, however, hospital patients rated the word nparents" as more emotional and more pleasant than did college students. Emotional verbal stimuli. Table 5, page 59, shows that for college as well as for hospital sub jects the word "intercourse" was significantly more emotional than either recognition or nourishment. The word "recognition" was significantly more emotional than the word nourishment. Table 7, page 61, shows that hospital patients tended to rate the stimulus words as more emotional than did the college students. Pleasant tones of verbal stimuli. Table 5, page 59, shows that for both college and hospital subjects the word "intercourse" was significantly more pleasant than either recognition or nourishment. The word "recognition" did not differ significantly from the word "nourishment" with respect to pleasant ness. Table 8, page 62, shows that hospital patients tended to rate the stimulus words as more pleasant than did the college students. 66 Pleasant tones of verbal contexts. Table 5, page 59, shows that the "pleasant context" was rated significantly more pleasant than either the neutral or the unpleasant contexts by both college and hospital subjects. Measures of Response Variables— Effects~oT~Family Context Table 9 shows the differences between responses associated to word-need stimuli presented singly and word responses associated to the same word-need stimuli presented within a family context. The follow ing observations may be made by examining the table. Frequency of word responses. Family context seems to increase, decrease or not influence the frequency of associative word responses to word-need stimuli presented singly, depending upon the particu lar need and the group of subjects concerned. For both college and hospital subjects, significantly more subjects associated a greater number of verbal responses to the stimulus word "nourishment" presented within a family context than to the same need stimulus word presented singly. On the other hand, signifi cantly more subjects in both experimental groups associated fewer word responses to the stimulus word "intercourse" presented within a family context than D ifferences Between Family and Neutral Contexts in Eliciting Affective and Non-Affective Associative Word Responses 1 Pr 2P 3U Hp0 5E 6 Eo F-N F-X F-N F-X F-N F-X F-N F-X F-N F-X F-N F-X + - p + - p + - P + - p + - P + - p + - p + - p + - P + p + - p + - p College Gr o u p: A: 35 13 .01 30 16 N.S 28 20 N.S Ho 10 .01 20 18 N.S 20 |H N.S 25 16 N.S 25 19 N.S 25 22 n .s zk 27 N.S 30 16 N.S 25 20 N.S B: |H 33 .01 17 3^ .05 17 27 N.S 23 28 N.S 17 12 N.S 20 16 N ,S 18 26 N.S 15 zk N.S |8 26 N.S 31 19 N.S 17 27 N.S 10 32 .01 C: 15 31 ! .01 15 3^ .01 I 1 * 31 .05 20 29 N .S 9 2H .05 II 23 N.S 18 19 N.S El 27 .05 19 28 N.S zk 25 N.S |H 25 N.S 13 27 .05 Hospital Group A: 39 19 .01 45 23 .05 38 22 N .S H3 23 .05 2H 23 N .S 21 17 N.S 22 15 N .S 16 16 N .S H6 21 .01 H6 18 .01 29 29 N.S 33 31 N.S B: 26 31 n.s 23 kz .05 2 k 37 N.S 29 30 N.S 20118 N.S 25 2] N.S 16 17 N .3 13 21 N.S 30 30 N.S 28 30 N.S 2H 29 N.S 2H 32 N.S C: [3 39 .01 iH H5 .0[ 22 J J II .s 2H 33 I J .S 20 17 N.S 21 17 N , S 20 15 N.S 18 17 N.S E5 H2 .01 22 H3 .05 22 25 N .S 29 22 N.S A: Nourishment B: Recognition C: Intercourse 1 PR: Production of words - Number of word responses associated to word NEED STIMULI WITHOUT REFERENCE TO judgments of affective tones 2P: Number OF WORD RESPONSES JUDGED AS PLEASANT 3U: Number of word RESPONSES judged as UNPLEASANT H Po; Number OF WORD RESPONSES JUDGED AS NEUTRAL 5E = Number OF WORD RESPONSES JUDGED AS EMOTIONAL 6 Eo: Number OF WORD RESPONSES JUDGED AS UNEMOTIONAL F: FAMILY CONTEXT N: Word need stimulus presented singly X: Neutral, CONTEXT + : Number OF CASES WHERE F EXCEEDS N IN F-N, F EXCEEDS X IN F-X, ETC. - ; Number OF CASES WHERE F IS LESS THAN N IN F~N, F IS LESS THAN X IN F-X,, ETC. p : Level of significance as indicated by the tables of sign test 68 to the same stimulus word presented singly. No significant change occurred with reference to the stimulus word "recognition" for hospital subjects. For college subjects the difference was significant, indicating the inhibiting effect of family context. Pleasant toned responses. With reference to the sex need stimulus word, significantly more college subjects associated fewer pleasant word responses to the word "intercourse" presented within a family con text than to the same stimulus word presented singly. Such difference failed to reach significance among hospital patients. The number of pleasant responses to "recognition" and "nourishment" did not change significantly when presented in the family context. Unpleasant toned responses. Similarly, with reference to the sex need stimulus word, significantly more college subjects associated fewer unpleasant word responses to the word "intercourse" presented within a family context them to the same word stimulus presented singly. The number of unpleasant responses to "recognition" and "nourishment" did not change significantly when presented in the family context. 69 Emotionally toned responses. A significantly greater number of hospital subjects associated fewer emotional word responses to the sex need stimulus word "intercourse" presented within a family context than to the same word stimulus presented singly. On the other hand, a significantly greater number of hospital patients associated a greater number of emo tional words to the stimulus word "nourishment" pre-o sented within a family context than to the same stimulus word presented singly. Indifferent toned responses. The frequency of unemotional and pleasant neutral word responses to word-need stimuli presented within a family context did not differ significantly from their frequencies to word-need stimuli presented singly for either of the two experimental groups. Effects of Pleasant Context Table 10 shows the differences between responses associated to word-need stimuli presented singly and those associated to the same word stimuli presented within pleasant, unpleasant, and neutral contexts. When the pleasant context is compared with the unpleasant and neutral contexts it may either facili tate, inhibit, or not influence the frequency and Table |0 D ifferences Between Verbal Contexts in Eliciting Pl e a s a n t, 'Unpleasant and Neutral Verbal Responses 1 p 2 u 3 x P-N P-U P-X U-N U-P U-X X-N X-P X-U + - p + - p + - p + - p + - p + - p + - P + - p + - p College Gr o u p: A : 20 N ,S 26 i r .05 29 19 N.S 27 |1* N .S 25 13 N.S 28 8 .01 2l* 17 N.S 19 23 N .S 20 22 N.S B: 31 15 .05 3* 10 .01 38 1 1 .01 2l* 1 1 .05 22 17 N.S 17 12 N.S 23 19 N.S 2l* 17 N.S 18 17 N.S C: 15 27 N .S 18 25 N.S 16 30 N .S 2[ 10 N.S 29 7 .01 23 8 .05 30 10 .01 31 13 .05 20 16 N .S Hospital Gr o u p: A: 1 *2 15 .01 1 *4 15 .01 37 22 N . S 31 * 1 1 .01 38 6 .01 30 13 .05 20 Ik N.S 18 18 N.S 18 14 N .S B: 1 *6 15 .01 ^7 12 .01 38 15 .01 36 9 .01 35 15 .01 31 1 1 .01 22 12 N.S 21 10 N.S 20 21 N .S C: |8 1 *3 .01 38 23 N .S 23 hi .05 23 17 N .S 31 * 15 .01 22 19 N.S 17 12 N.S 25 9 .01 16 12 N .S Ke y: I P: Pleasant Context 2 U: Unpleasant Context 3 X: Indifferent Context P-N: P-Us P-X: + : Pleasant responses given to word need stimuli presented within a pleasant, context minus pleasant responses GIVEN TO WORD NEED STIMULI PRESENTED SINGLY, Pleasant responses given to word need stimuli presented within a pleasant context minus pleasant responses GIVEN TO WORD NEED STIMULI PRESENTED WITHIN AN UNPLEASANT CONTEXT. Pleasant responses given to word need stimuli presented within a pleasant context minus pleasant responses given to word need stimuli presented within an indifferent context. Number of cases where P exceeds N in P-N, P exceeds U in P-U, P exceeds X in P-X, etc. Number of cases where P is less than N in P-N, P is less than U in P-U, P is less than X in P-X, etc. Level of significance as indicated by the tables of sign t e s t. Nourishment B: Recognition C: Intercourse (a ,B ,C : Word need stimuli) -0 o 71 value of pleasant word responses associated to word- need stimuli. Such an influence varies according to the particular need and the group of subjects concerned. Whereas the pleasant context seems to generally facilitate the production of pleasant word responses associated to the word stimuli which repre sent the needs for nutrition and recognition, it seems to generally inhibit the production of pleasant word responses associated to the stimulus word which represents the need for sex. Where such effects were not statistically significant, there was a consistent trend in the direction indicated above. It should also be noted that the significant effects were more noticeable among hospital subjects than among college subjects. Effects of Unpleasant Context When the unpleasant context is compared with the pleasant context it seems to generally facilitate the production of unpleasant word responses associated to word-need stimuli. Such effect varies according to the particular need and the group of subjects con cerned. Significantly more hospital subjects asso ciated a greater number of unpleasant word responses to each of the three word-need stimuli presented 72 within an unpleasant context than to the same word stimuli presented within a pleasant context. Such significant difference was obtained for college sub jects, with reference to the sex need stimulus only. For the nutrition and recognition need stimuli, the same direction of differences was observed without reaching an acceptable level of statistical signifi cance. Effects of Neutral Context When the neutral context is compared with the pleasant and unpleasant contexts it seems to generally facilitate the production of neutral word responses to word-need stimuli. However, such effect is not significant in most cases for either hospital or college subjects. Only with reference to the sex need stimulus were there significantly more hospital and college subjects who associated a greater number of neutral word responses to the word "intercourse’ * presented within a neutral context than to the same stimulus word presented within a pleasant context. Differences among Responses to Different Need Stimuli Table 11 presents the results of the differences between the affective values of word responses which Table 11 D ifferences Between Word need Stimuli in Eliciting Affective And No n-Affective Verbal Responses Pleasant Responses Unpleasant Responses Neutral Responses Group R-N l-N l-R R-N l-N 1 -R R-N l-N l-R + - p + - p + - p + - p + - p + - P + - p- + - p + - P College Subjects 28 21 n .s 28 22 n .s 22 26 N.S 15 23 N.S 21 18 N.S 25 10 .05 21 23 N.S 21 25 N.S 17 26 N.S Hospital « Subjects zk 38 N.s 39 22 .05 kl 20 .01 17 3^ ■05 zk 25 N .S 25 15 N.S 21 16 N.S 18 20 N.S 15 23 N.S Sum of responses G iven Under All F ifteen Conditions College Subjects 32 [8 n .s 30 21 n .s 25 26 N ,S Id 3° N.S 17 27 N .S 31 19 N .S 13 37 .01 12 37 .01 16 32 .05 Hospital ‘Subjects zk 1*5 .05 36 31 N.S 1*3 25 .05 33 25 N ,S 30 27 N.S 27 30 N .S 17 32 .05 13 1*9 .01 15 29 .05 Sum of Pleasant and Unpleasant Responses Emotional Responses UNEMOTIONAL Responses College Subjects 32 17 .05 32 17 .05 26 23 N .S 17 28 N.S 19 26 n .s 27 |8 N.S Hospital Subjects 39 26 N .S 51* 12 .01 51 16 .01 21 39 .05 18 ^3 .01 25 31 N.S Sum of responses G iven Under All F ifteen Conditions iCOLLEGE Subjects 26 23 n .s 22 26 n .s 26 27 N.S 30 22 N ,S 31 21 N.S 28 23 N.S 8 kz .01 7 kk .01 zk 27 N.S Hospital Subjects 18 50 .01 33 35 N.s **•7 20 .01 k\ zk .05 50 15 .01 1*6 21 .01 18 ^7 .01 7 58 .01 27 36 N .S K e y: R: Recognition N: Nourishment I: Intercourse +: Number of cases where R exceeds N in R-N, I exceeds N in l-N , e t c. Number of cases where R is less than N in R-N, I is less than N in l-N , e t c. p : Level of significance as indicated by the tables of sign t e s t. •J C 4 74 were associated to each of the three word stimuli which represent the needs for nutrition, recognition, and sex. The following is a summary of comparisons. Nourishment and recognition (presented singly). When the stimulus word "nourishment" (presented singly) was compared with the stimulus word "recognition" (presented singly) the following results were obtained: 1* With reference to production of pleasant word responses, no significant difference was found for either college or hospital subjects. 2. With reference to the production of unpleas ant word responses, significantly more hospital subjects associated fewer unpleas ant word responses to the stimulus word "recognition" than to the stimulus word "nourishment." No such significant differ ence was found for college subjects. 3. With reference to the production of neutral word responses, the two word-need stimuli did not differ significantly for either college or hospital subjects. 4. With reference to the production of emotional word responses, significantly more college 75 subjects associated a greater number of emotional word responses to the stimulus word "recognition" than the stimulus word "nourishment." This difference did not reach significance for hospital subjects. 5. With reference to the production of unemo tional words, significantly more hospital subjects associated a greater number of unemotional words to the stimulus word "nourishment" than to the stimulus word "recognition." This difference did not reach significance for college subjects. Nourishment and recognition (combined presenta tions) . When the stimulus word "nourishment" was compared with the stimulus word "recognition” in terms of the results obtained from their five presentations singly and within different verbal contexts, the following results were obtained: 1. With respect to the production of pleasant word responses, significantly more hospital subjects associated a greater number of pleasant word responses to the stimulus word "nourishment" than to the stimulus word "recognition." The reverse direction was obtained from college subjects although the difference did not reach an acceptable level of significance. With respect to the production of unpleasant word responses, the two word stimuli did not differ significantly for either hospital or college subjects. It may, however, be observed that both groups show opposite directions, i.e. hospital subjects tend to associate more unpleasant words to the stimulus word "recognition'* than to the stimulus word "nourishment," whereas college subjects tend to associate more unpleasant words to the stimulus word "nourishment" than to the stimulus word "recognition." With respect to the production of neutral word responses, significantly more college and hospital subjects associated a greater number of neutral word responses to the stimulus word "nourishment" than to the stimulus word "recognition." With respect to the production of emotional word responses, significantly more hospital subjects associated a greater number of emotional word responses to the stimulus word "recognition" than to the stimulus word "nourishment." Such difference did not reach an acceptable level of signifi cance for college subjects. 5. With respect to the production of unemotional word responses, significantly more college and hospital subjects associated a greater number of unemotional words to the stimulus word "nourishment" than to the stimulus word "recognition." Nutrition and sex (presented singly). When the stimulus word "nourishment" presented singly was com pared with the stimulus word "intercourse," the follow ing results were obtained: 1. With reference to the production of pleasant word responses, significantly more hospital subjects associated a greater number of pleasant words to the stimulus word "inter course" than to the stimulus word "nourish ment." Such difference did not reach significance for college subjects. 2. With reference to the production of unpleas ant word responses, the two word-need stimuli did not differ significantly for 78 either hospital or college subjects. 3* With reference to the production of neutral word responses, the two word-need stimuli did not differ significantly for either hospital or college subjects. 4. With reference to the production of emotional word responses, significantly more hospital and college subjects associated emotional word responses to the stimulus word "inter course" than to the stimulus word "nourish ment ." 5. With reference to the production of unemo tional words, significantly more hospital subjects associated a greater number of unemotional words to the stimulus word "nourishment" than to the stimulus word "intercourse." The difference did not reach significance for college subjects. Nutrition and sex (combined presentations). When the stimulus word "nourishment" was compared with the stimulus word "intercourse" in terms of the results obtained from their five presentations singly and within different verbal contexts, the following results were obtained: With reference to the production of pleasant word responses, the two word-need stimuli did not differ significantly for either hospital or college subjects. With reference to the production of unpleas ant word responses, the two word-need stimuli did not differ significantly for either hospital or college subjects. With reference to the production of neutral word responses, significantly more hospital and college subjects associated a greater number of neutral word responses to the stimulus word "nourishment" than to the stimulus word "intercourse." With reference to the production of emotional word responses, significantly more hospital subjects associated a greater number of emotional word responses to the stimulus word "intercourse" than to the stimulus word "nourishment." The difference did not reach significance for college subjects. With reference to the production of unemo tional word responses, significantly more hospital and college subjects associated a greater number of unemotional word responses 80 to the stimulus word "nourishment" than to the stimulus word "intercourse." Sex and recognition (presented singly). When the stimulus word "intercourse" was compared with the stimulus word "recognition, both being presented singly, the following results were obtained: 1. With reference to the production of pleasant word responses, the two need stimuli did not differ significantly for college subjects, but differed very significantly for hospital subjects. Significantly more hospital sub jects associated a greater number of pleasant word responses to the stimulus word "inter course" than to the stimulus word "recog nition. " 2. With reference to the production of unpleas ant word responses, the two word-need stimuli did not differ significantly for hospital subjects, but differed signifi cantly for college subjects. Significantly more college students associated a greater number of unpleasant word responses to the stimulus word "intercourse" than to the stimulus word "recognition." 81 3. With reference to the production of neutral word responses, the two word-need stimuli did not differ significantly for either experimental group. 4. With reference to the production of emotional word responses, the two words did not differ significantly for college subjects, but differed significantly for hospital subjects. Significantly more hospital subjects asso ciated a greater number of emotional word responses to the stimulus word "intercourse" than to the stimulus word "recognition.” 5. With reference to the production of unemo tional word responses, the two word-need stimuli did not differ significantly for either experimental group. Sex and recognition (combined presentations). When the stimulus word "intercourse" was compared with the stimulus word "recognition" in terms of the results of their five presentations singly and within different verbal contexts, the following observations were obtained: 1. With respect to the production of pleasant word responses, the two need stimuli did not differ significantly for college sub jects, but differed significantly for hospital subjects. Significantly more hospital subjects associated a greater number of pleasant word responses to the stimulus word "intercourse" than to the stimulus word "recognition." With respect to the production of unpleasant word responses, the two word-need stimuli did not differ significantly for either experimental group. With respect to the production of neutral word responses, significantly more college and hospital subjects associated a greater number of neutral word responses to the stimulus word "recognition" than to the stimulus word "intercourse." With respect to the production of emotional word responses, the two words did not differ significantly for college subjects, but differed significantly for hospital sub jects. Significantly more hospital sub jects associated a greater number of emotional words to the stimulus word "Intercourse" than to the stimulus word 83 "recognition." 5. With respect to the production of unemotional word responses, the two need stimuli did not differ significantly for either experimental group. The lengthy preceding presentation of differ ences among responses to the three word stimuli may be expressed by a simple comparison of needs with respect to the five categories of affective responses by pool ing the results of similar responses into one score for each category, and by presenting them in rank order of magnitude in relation to the three word stimuli. The results of this procedure were; (1) the largest number of pleasant, unpleasant, and emotional word responses, as well as the smallest number of indiffer ent responses, were associated to the sex need stimulus; (2) the smallest number of pleasant, unpleasant, and emotional word responses, as well as the largest number of indifferent responses, were associated to the nutrition need stimulus, nourishment; (3) the median position for each response category belonged to word responses associated to the stimulus word "recognition." These findings represent a general trend rather than statistically consistent significant differences. 84 Differences between Hospital and College Snb.iects When hospital patients were compared to college students with respect to the production of word responses and the production of affective toned word responses to word-need stimuli, the results presented in Tables 12 to 17 were obtained. Production of word responses. Table 12 shows that in eight out of fifteen experimental conditions college subjects produced a significantly greater number of associated word responses than did hospital subjects. It also shows that, with reference to the sum of all word responses, college subjects produced a significantly greater number of associated word responses than did hospital subjects. Production of pleasant responses. Table 13 shows that hospital patients produced significantly a greater number of pleasant words with greater intensity than did college students in response to the two word-need stimuli "nourishment” and "inter course." With the exception of one out of five experimental conditions dealing with the stimulus word "recognition," the two groups of subjects did not differ significantly in their production of Table 12 D ifferences Between College And Hospital Subjects In Their Production Of Associated Word Responses To Word Need Stimuli I 2 3 4 5 6 7 8 9 10 II |2 13 |4 15 Hospital Mean 17.67 20.39 19-19 l 8 . 18.52 |6.|0 15.64 18.87 18.25 17.65 18.48 17.17 15-97 19-33 19-1^ College Mean 21.35 24.71 24.10 22.15 2| .87 21.19 19-71 23.33 22.06 21.90 21.23 19.12 17.88 22.57 22.12. Kol-Smirnov Statistic 19.06 29.89 27.94 [9.18 26.60 30.60 26.28 28.73 24.86 25.72 19.96 11.45 l4.96 20.29 23.86 Chi Square Estimate 4,30 |0.4o 9.10 4.36 5.03 II.II 8.19 9-68 7.33 7,85 4.72 I.56 2.56 4.83 6.75 Significance Level .2000 .0100 .0200.2000 .1000 .0050 .0250 .0100 .0500 .0250 .1000 .5000 .3000 .1000 .0500 Results Of The F ifteen Conditions Comb ined: Hospital Subjects Mean = 271 College Subjects Mean = 325 t (df = U 9) = 3-379 p. - .01 Ke y : 1- Nourishment 6- Recognition 11- Intercourse 2- Home-Parents-Nourishment 7“ Home-Parents-Recognition 12- HOME-PARENTS-INTERCOURSE 3- Ro s y-Smil ing-nourishment 8- ROSY-SM1L 1NG-RECOGH1T 1 ON 13- Ro sy-Smiling-Intercourse 4- BlUE-UMHAPPY-NOUR1SHMENT 9- Blue-Unhappy-Recognition l4- Blue-Unhappy-Intercourse 5- XOMF-KTQHAFX-NOUR1SHMENT 10- Xomf-Ktqhafx-Recognition 15- XOMF-KTQHAFX-INTERCOURSE 00 0 1 Table 13 D ifferences Between Hospital And College Subjects |n Production Of Pleasant Toned Verbal Responses 1 2 3 4 5 6 7 8 9 10 1 [ 12 13 15 Hospital Mean College Mean Kol-Smirnov Statistic Chi Square Estimate S ignificance Level 16,78 16.71+ 19.43 20.87 14.29 16.43 17-03 15-93 19.68 15.75 20.93 20.84 18.62 18.59 20.38 12.62 n.60 17.56 (4.38 10.27 1X98 13.42 13.71 18.75 12.67 15.77 15-35 1^.58 13.52 14.48 25.81 26.23 17-11 39-8o 28.83 12.26 16.56 23.47 18.03 19 2 9 . 6 0 28.12 $5.78 30.60 35.42 7.90 8.16 3.47 18.79 7-31 1.78 3-25 6.53 3.86 4.50 10.39 9.38 7-88 11.11 14.88 .0250 .0250 .2000 .0005 .0500 .5000 .2000 .0500 .2000 .2000 .0100 .0100 .0250 .0250 .00(0 results Of The Fifteen Conditions Combined: Hospital Mean = 272 .14 College Mean = 2(3-66 t (of = 119) = 3.96 p . = .01 No t e: Values are sums of weighted Ke y: 1- Nourishment 2- Xomf-Ktqhafx-Nour1shment 3- Home-Parents-Nourishment Rosy-Smiling-Nourishment 5- Bl ue-Unhappy-Nourishment 6- Recognition 7- Xomf-Ktqhafx-Recognition 8- Ho m e-Parents-Recognition 9- Ro sy-Smiling Recognition 10- Blue-Unhappy Recognition 11- Intercourse 12- XOMF-KTQHAFX-INTERCOURSE 13- HOME-PARENTS-INTERCOURSE |4- Rosy-Smiling-Intercourse 15- Blue-Unhappy-Intercourse GO 0 5 Table |4 D ifferences Between Hospital And College Subjects In Production Of Unpleasant Verbal Responses 1 2 3 4 5 6 7 8 9 10 11 12 13 |4 15 Hospital Mean 2.10 2.42 2.74 l .19 4-55 2.23 3-28 2.80 2.29 5.17 2.36 3.07 2.81 1.88 4.23 College Mean 3-23 3.08 3-17 3.23 4.50 2.52 2.75 2.71 3.06 3.92 3.98 3.29 2.33 2.65 4.90 Ko l-Smirnov Statistic 19.09 19.62 14.63 33.05 13.68 [I.3 7 9.42 6.88 13.32 13.13 20.57 7.02 12.21 9.59 16.61 Chi Square Estimare 4.32 4.57 2.54 12.96 2.22 1.53 1.05 .56 2.10 2.04 5.02 .58 1-77 1.09 3-27 S ignificance Level .2000 .2000 .3000 .0050 .4ooo .5000 .6000 Co O O O O O O * = r 0 0 0 . 1000 O O O CO .5000 .6000 .2000 Results Of The Fifteen Conditions Combined: Hospital Mean = 43.25 College Mean = 49.32 t (df = 119) = .89 P . = N.S . Values are sums of weighted sco res. Note : Ke y: i - Nourishment 2- XOMF-KTqHAFX-NOURISHMENT 3- Home-Pare nts-Nourishment 4- Rosy-Smiling-Nourishment 5- BLUE-UNHAPPY-NOURISHMENT 6- Recognition 7- Xomf-Ktqhafx-Recognition 8- Ho m e-Parents-Recognition 9- Ro s y-Smiling-Recognition 10- Blue-Unhappy-Recognition 11- Intercourse 12- XOMF-KtQHAFX-INTERCOURSE 13- HOME-PARENTS-INTERCOURSE I1*- Rosy-Smiling-Intercourse 15- Blue-Unhappy-!ntercourse 00 ■Nj Table 15 D ifferences Between Hospital And College Subjects In Production Of Neutral Verbal Responses 1 2 3 it 5 s O 7 8 9 10 II 12 13 |i+ 15 Hospital Mean 3.71* H.38 3-52 4.06 3.81 3.25 3.87 3.00 2.94 3.72 2.1+6 3-01 3-19 2.1+8 2.90 College Mean 8.71 9.56 10.38 9.9^ 9.58 7.90 8.71 7.62 7.06 8.1+0 6.96 8.17 6.1+8 6-37 7.98 Ko l-Smi: rnov Statistic 38.71 37-71* 39.72 1 +1+.89 35.^8 35.1*5 31*.36 36.87 29.2*1 39-33 M -75 1+7.1+9 27.60 ^9.33 1+3.12 Chi Square Estimate 17.77 I0-89 18.71 23.59 1* 1.93 1^.91 1H.01 16.13 10.1U 18.31+ 20.67 26.75 9.05 28.86 22.05 S ignifi cance Level .0005 .0005 .0005 .0005 .0010 .0010 .oo;to .0005 .0100 .0005 .0005 .0005,. .0250 .0005 .0005 Results Of The F ifteen Conditions Co mb ined: Hospital Mean = 1+9.51+ College Mean - 12^.02 t (df = 119) = 5-3^ p . = .01 Note: Values Are Sums of Weighted Scores. Ke y: 1- Nourishment 2- Xomf-Ktqhafx-Nour1shment 3- Home-Parents-Nourishment 1 +- Rosy-Smiling-Nourishment 5- Blue-Unhappy-Nourishment 6- Recognition 7- Xo m f-Ktqhafx-Recognition 8- Home-Parents-Recognition 9- Rosy-Smiling-Recognition 10- Blue-Unhappy-Recognition II- Intercourse |2- Xomf-Ktqhafx-|ntercourse 13- Home-Parents-Intercourse 14- Rosy-Smiling-Intercourse 15- Blue-Unhappy-Intercourse Table |6 D ifferences Between Hospital And College Subjects In Production Of Emotional Verbal Responses 1 2 3 4 5 6 7 8 9 10 II 12 13 14 15 Hospital Mean College Mean Kol-Smirnov Statistic Chi Square Estimate S ignificance Level 19.79 25.12 23.40 21.66 20.00 23.86 24.57 25.85 27.13 26.19 29.99 26.75 28.09 30.76 30.74 [9.22 20.20 18.71 18.88 18.81 24.15 23.00 24.81 21.13 20.25 25.35 23.37 23.00 23,73 23.39 10.29 21.57 18.63 1^.71 9-20 11.82 12.63 11-76 23.42 19.34 18.46 15.69 22.94 22.24 18.63 I.2 5 5 A 2 4 .o4 2.52 1.00 [.6 6 1.89 1.63 6.46 4.44 3.99 2.87 6.24 5.73 4 .o4 .6000 .1000 .2000 .3000 .7000 .5000 .4ooo .5000 .0500 .2000 .2000 .3000 .0500 .1000 .2000 results Of The Fifteen Conditions Combined: Hospital Mean = 394.70 College Mean = 335-58 t ( df = 119) = 3.97 p. = .01 No t e: Values are sums of weighted Ke y: I - Nourishment 2- Home-Parents-Nourishment 3- Rosy-Smjling-Nourishment 4- Blue-Unhappy-Nourishment 5- Xom f-Ktqhafx-No u rishment 6- Recognition 7- Home-Parents-Recognition 8- Rosy-Smiling-Recognition 9- Blue-Unhappy-Recognition !0- Xomf-Ktqhafx-Recognition 11- Intercourse 12- Home-Parents-|ntercourse 13- Rosy-Smiling-|ntercourse | 4 - Blue-Unhappy-1ntercourse 15- Xomf-Ktqhafx-|ntercourse 00 < 0 Table 17 D ifferences Between Hospital And College Subjects In Production Of Unemotional Verbal Responses 1 2 3 4 5 6 7 8 9 10 ll 12 13 14 15 Hospital Mean 6.17 7.42 5*97 6,54 6.12 4.68 4.62 3*91 4 .8 i 3*93 3*99 4.49 4.07 3*^5 3*12 College Mean 9.96 IO.73 11.15 10.62 9*79 7.65 9*23 7.27 8.31 9.25 7.50 8*33 6 .o4 4.98 ■ 8.10 Ko l-Smirnov Statistic 24.44 25.33 3^ *5° 26 *3** 25.33 I9-98 32.11 24. 4 ( 27.76 43.12 44.12 38.38 23.83 19.15 45.60 Chi Square Estimate 7.09 7*61 14 . 12 8.23 7*61 4.74 12.23 7.07 9 .(4 22.05 23.09 17.47 6.73 4.35 24.66 S ignificance Level .0500 .0250 .0010 .0250 .0250 .1000 .0050 .0500 .0250 .0005 .0005 .0005 .0500 .2000 .0005 Results Of The Fifteen Conditions Combined; Hospital Mean = 73*32 College Mean = 127.61 t (df = 119) = 15.92 p. = .01 No t e* . Values Are Sums Of Weighted Scores Ke y: 1- Nourishment 2- Home-Parents-Nourishment 3- Ro s y-Smiling-Nourishment 4- Blue-Unhappy-Nourishment 5- Xo m f-Ktqhafx-Nourishment 6- Recognition 7- Ho me-Parents-recognition 8- Ro sy-Smiling-Recognition 9- Blue-Unhappy-Recognition 10- Xomf-Ktqhafx-Recognition 11- Intercourse 12- Home-Parents-Intercourse 13- Rosy-Smiling-Intercourse 14- Blue-Unhappy-!ntercourse 15- Xomf-Ktqhafx-|ntercourse 91 pleasant responses to the stimulus word "recognition." It should be noted, however, that, as a general trend, hospital subjects produced a slightly greater number of pleasant responses with greater intensity in response to the word "recognition" than did the college subjects. The difference between the means of sums of production of pleasant responses by college and hospital subjects was significant beyond the .01 level. Production of unpleasant responses. Table 14, page 87, shows that hospital patients did not differ significantly from college students in the production of unpleasant responses. However, it should be noted that college students produced consistently slightly greater number of unpleasant responses with greater intensity than did hospital patients. In spite of this fact, the difference between the means of sums of production of unpleasant responses by college and hospital subjects was not significant. Production of both pleasant and unpleasant responses. The difference between the means of sums of production of both pleasant and unpleasant responses by college and hospital subjects was significant beyond the .01 level. Hospital subjects produced the greater number of responses (Table 14, page 87). 92 Production of neutral responses. Table 15, page 88, shows that college subjects associated a significantly greater number of neutral words to each word-need stimulus than did hospital subjects. Simi larly, the difference between the means of sums of their production of neutral responses was significant beyond the .01 level. Production of emotional responses. Table 16, page 89, shows that, with the exception of one out of fifteen experimental conditions, both groups of sub jects did not differ significantly in the production of emotional word responses. The one exception was that of the stimulus word "recognition" presented within an unpleasant context. Hospital subjects associated a significantly greater number of emotional words with greater intensity than college subjects. It should also be noted that, as a general trend, hospital subjects produced slightly more emotional word responses with greater intensity than college subjects. This trend became meaningful as the differ ence between the means of sums of production of emo tional responses proved significant beyond the .01 level. 93 Production of unemotional responses. Table 17, page 90, shows that, with the exception of one out of fifteen experimental conditions, college subjects produced a significantly greater number of unemotional word responses than did hospital subjects. The one exception was that of the stimulus word "recognition" presented within a family context, where significance reached the .10 level only. The difference between the means of sums of production of unemotional responses by college and hospital subjects was significant beyond the .01 level, with college subjects producing the greater number of such responses. Correlations between Ratings of Set and Response Variables Tables 18 through 22 present the correlations between ratings of set variables and ratings of response variables. The following is a summary of the findings which are relevant to the hypotheses investigated in this study: 1. Correlations between judgments of pleasant ness of verbal stimuli and judgments of pleasantness of verbal responses ranged from -.03 to .52 with a median of .27 for college subjects, and ranged from -.08 to .26 with a median of .04 for hospital Table |8 Correlations Between Judgments Of Pleasantness Of Verbal Stimuli, Judgment Of Pleasantness Of Associated Verbal Responses, And Rate Of Word Production 1 2 3 ^ 5 6 7 8 9 10 11 12 13 l4 15 College Subjects : R1 : .28 .27 .21 .21 -.03 .27 .52 .03 .06 .23 .44 •09 .35 .22 .29 R2 : .03 -.50 -.111 -.09 -■19 -.09 .00 -.12 .00 -•37 -• 15 -.21 -.13 -.10 .01 R3 : -.06 .08 .05 -.05 .ib -.lb -.06 -.15 -.13 -.23 .04 -.10 .06 -.11 -.04 Hospital Subjects : Rl : .01 -.02 .26 .01 • 15 -.08 .09 -.10 .01 .o4 .02 .22 .|6 .05 .11 R2 : .24 -.15 -.07 -.01 -.05 -•13 .21 -.01 -.09 .08 -.07 -.11 -.03 .04 .02 R3 : ~.3b -.i4 -.01 .09 .10 -.14 •13 -.02 .23 -.09 -.06 -.01 -.o4 .06 -.20 No t e: College Subj ects: n = 52; WITH OF = 50. r - .27 AT P = .05 AND r = .35 at p = .01 Hospital Su bj ects; N = 69; WITH DF = 67 , Ft - .24 AT P r .05 AND R 31 at p = .01 Ke y: Rl- Pleasant Responses 1- Nourishment 2- Xomf-Ktqhafx-Nourishment 3- Home-Parents-Nourishment 4- Rosy-Smi ling-Nourishment 5" BLUE-UNHAPPY-NOURISHMENT R2- Unpleasant Responses 6- Recognition 7- Xomf-Ktqhafx-Recognit1 ON 8- Hom e-Pare nts-Recognition 9- Rosy-Smiling-Recognition |0- Blue-Unhappy-Recognition R3- Word Production II- Intercourse |2- Xomf-Ktqhafx-Intercourse 13- Home-Parents-Intercourse |4- Rosy-Smiling-|ntercourse 15- Blue-Unhappy-1:ntercourse Table 19 CORRELATIONS BETWEEN JUDGMENTS OF EMOTIONAL TONES OF VERBAL STIMULI, JUDGMENTS OF Emotional Tones Of Associated Verbal Responses, And Rate Of Word Production 2 3 b 5 6 7 8 9 10 11 12 13 ill 15 college Subjects : Rl : .53 -^5 .52 .60 .50 .29 ,57 .57 ,39 .61 ,36 .‘lO .56 .1*5 .35 R2 : .15 .19 .19 .16 -.09 .03 .10 .06 .29 .03 .11 .05 .08 -.c8 Hospital Subjects Rl R2 .1*6 .27 .1*1 .38 .50 .28 .23 .25 .39 .1*5 .29 .1*0 .26 .30 .53 .06 -.05 -.11 .00 .15 .00 -.07 -.16 .05 -.08 -.15 .00 -.09 -.02 .03 Note: College Su~jects: n = 52; with df = 50, r = .27 AT P = .05 and r = .35 at p - .0| Hospital Subjects: n - Ke y: Rl- Emotional Responses I - N0urishment 2- Home-Parents-Nourishment 3- Rosy-Smil1ng-Nourishment 1 *- BLUE-UNHAPPY-NOUR I SHMENT 5- Xomf-Ktqhafx-NourISHMENT WITH DF = 67, R - .2! * AT P - .05 AND R = .31 AT .01 R2- Word Production 6- Recognition 7- Home-Parents-Recognition 8- Ro sy-Smiling-Recognition 9- Blue-Unhappy-RecognitiON 10- Xomf-Ktqhafx-Recognition I I - Intercourse |2- Home-Parents-Intercourse 13- Rosy-Smiling-Intercourse |1 *- Blue-Unhappy-Intercourse 15- Xomf-Ktqhafx-|ntercourse Table: 20 Correlations Between Ratings Of Mo o d , Emot ionality, Need Satisfaction And Ratings of Pleasant Tones Of Verbal Responses 1 2 3 1 } 5 6 7 8 9 10 1 1 12 13 |4 ■15 College Subjects Mood .06 -.0! -.03 • 07 .06 ■ 17 .08 .23 •03 .14 .11 -13 .11 • 17 .02 Emot1ONAL1TY -.03 -.12 .00 -.08 .16 -.06 -.01 -.03 .18 .05 .05 .12 . 1 1 -.o4 .05 Need Satisfaction Nutrition .29 .24 .23 .29 .36 .24 ■ 23 .16 • 33 .19 .28 • 37 •33 .22 .21 Recognition .18 .01 -.09 .00 .50 ■ 17 .60 -.17 -.14 -.07 .39 •38 -.03 .25 -•.10 Sex -.29 -.ok 1 O -.ok .19 .26 -.07 .29 .41 .44 .26 • 25 .45 .06 .42 Hospital Subjects Mood .01 -.01 -.14 .01 .o4 .06 . 16 .02 - .06 .07 -.12 -.05 -.13 .03 -.01 Emotionality -.02 -.01} -.18 -.11 -.13 -.16 -.05 .06 -.05 -•13 .05 -■17 -. 10 -.07 I b Need Satisfaction Nutrition -.09 1 b .03 -.11 .01 -.11 .08 -.01 -.07 .01 -.13 •03 -. 10 .00 .00 Recognit 1 on -.15 .11 .12 -.ok .24 -.13 -.07 -.18 -.08 -.06 -.09 •03 .28 “•11 .08 Sex .04 .17 -.06 -.03 -.01 -.09 -.04 .02 -.10 .01 .10 •13 -.05 • 17 .06 Note: College Subjects: n *• 52; with df = 50, R = *27 at p = .05 and r = .35 AT P = .0| Hospital Subjects: n - 69; with df = 67, r = .24 at p = .05 and r = .31 at p = .01 Ke y: I - NOURISHMENT 6 2- XOMF-KTQHAFX-NOURISHMENT 7 3- Home-Parents-Nourishment 8 4- Rosy-Smiling-Nourishment 9 5“ BLUE-UNHAPPY-NOURISHMENT |0 - Recognition - Xo m f-Ktqhafx-Recognition - Ho m e-Parents-Recognition - Ro s y-Sm i ling-Recognit ion - Bl u e-Unhappy-RecogniTI ON 11- Intercourse 12- Xomf-Ktqhafx-Intercourse 13- HOME-PARENTS-INTERCOURSE |4- Rosy-Smiling-Intercourse 15- BLUE-UNHAPPY-INTERCOURSE Table 21 Correlations Between Ratings Of Mo o d, Emotionality, Need Satisfaction And Ratings Of Unpleasant Tones Of Verbal Responses 1 2 3 ^ 5 6 7 8 9 |0 || |2 13 |4 15 College Subjects Mood -.12 -.22 -•31 -.zb t.02 -.09 -.05 -.15 -.20 -.19 -.18 -.06 -.13 -.01 -.02 Emotionality -.08 -.15 .05 • 13 .00 -.o4 .05 -.17 .05 .05 .06 -.14 -.o4 .05 .02 Need Satisfaction Nutrition • 17 .20 • 13 -.04 .07 -.06 • 31 .41 • *7 .36 .o4 • 17 -.12 .18 -.23 Recognition .67 .82 .85 .47 .30 .68 -.06 -.10 -.14 -.08 • 37 .48 •51 .22 .56 Sex -.26 -.30 -.28 • 17 .27 -•15 •39 .24 .03 .16 .24 .17 -.18 • 19 -.12 Hospital Subjects Mood -.oU -.07 .07 -.04 .01 -.11 -.42 -.o4 .01 -.06 -.02 -.16 .08 -.06 .09 Em o t1onality .24 ■ 3^ • 31 .24 • 19 • 35 • 33 • 35 .27 .27 .24 • 31 •31 * • 30 .09 Need Satisfaction Nutrition -.l4 -.09 -.23 -.12 -•03 -.20 -.26 "•32 -.09 -.26 .3° -.26 -.24 -.20 -.12 Recognition -.02 .09 -.22 -.02 -.17 .08 -.02 -.14 .04 -.17 - .06 -.12 -.26 -.05 -.11 Sex -.2b 1 O -.32 -.12 -.21 -.08 -.26 -.19 -.29 -.21 -.26 -.21 • 15 -.19 -.22 No t e: College Subjects: n 52; WITH DF = R = .27 AT P - .05 AND R = .35 AT P - ,0| Hospital Subjects: n = 69; with df - 67, r = .24 at p = .05 and r = ,e| at p Ke y: 1 - NOURISHMENT .01 2- XOMF-KTQHAFX-NOURISHMENT 3- Ho me-Parents-Nourishment 4- Rosy-Smiling-Nourishment 5- BLUE-UNHAPPY-NOURISHMENT 6- RECOGNITION 7- Xo mf-Ktqhafx Recognition 8- Home-Parents-Recognition 9- Ro sy-Sm i l1ng-recognition 10- Blue-Unhappy-Recogn1t 1 ON 11- Intercourse 12- XOMF-KTQHAFX-1NTERCOURSE 13- HOME-PARENTS-INTERCOURSE |4- Rosy-Smi LI NG-lNTERCOURSE 15- BLUE-UNHAPPY-INTERCOURSE < 0 Table: 22 CORRELATIONS BETWEEN RATINGS OF MOOD , EMOTIONALITY, NEED SATISFACTION And Ratings Of Emotional Tones Of Verbal Responses I 2 3 4 5 6 7 8 9 10 11 12 13 14 15 College Subjects Mood .08 .00 .03 .07 .03 .12 .05 .08 .13 .10 .06 .12 .02 .05 .07 Emotionality •13 .12 .18 .14 .21 .00 .08 .24 .16 .14 -23 .18 .15 .23 .20 Need Satisfaction Nutrition .29 •23 .26 .32 .24 .39 • 19 .28 .26 • 31 .29 •35 • 23 .29 .21 Recognition -.23 -.01 -.11 -.02 -.12 -.04 -.06 -.15 -.07 -.08 -.03 .02 .01 .06 -.08 Sex -•35 -.06 -.06 -.07 -.15 -.It -.06 -.17 -.07 -.01 -.20 -•17 -.12 -.13 -.07 Hospital Subjects Mood .05 -.05 .00 .09 .08 .07 .07 -.04 .o4 -.01 -.01 -.02 .07 .03 -.04 Em o t1onality .26 •15 .20 .18 .22 .27 .22 .23 .21 .18 • 31 •23 • 30 .21 .18 Need Satisfaction Nutrition -.12 -.15 - .|4 -.08 -.09 -.13 .02 -.03 -.13 -.16 -.28 -.23 -.01 -. 16 -.15 Recognition -.18 - . | 4 -.16 -,o4 -.11 -.24 -.21 -.25 -.24 -.23 -.23 -.12 -.17 -.08 -.09 Sex -.01 -.18 -.09 .02 .08 - .|4 -.06 -.08 -.10 -.12 -.09 -.15 -.04 -.08 .00 No t e: College Subjects: n = 52; WITH DF = 50, R = 27 AT p = .05 AND R = -35 AT p = .01 Hospital Subjects: N = 69 ; WITH DF = 67, R ^ .24 AT p = .05 AND R = .3L AT p = .01 Key: I - NOURISHMENT 2- Home-Parents-Nourishment 3- Ro s y-Smil ing-Nourishment 4- Blue-Unhappy-Nourishment 5- Xomf-Ktqhafx-Nourishment 6- Recognition 7- Home-Parents-Recognition 8- Ro sy-Sm il ing-Recognition 9- Blue-Unhappy-Recognition 10- Xomf-Ktqhafx-Recognition [ I - |NTERCOURSE 12- HOME-PARENTS-INTERCOURSE 13- Rosy-Smiling-|ntercourse |4- Blue-Unhappy-|ntercourse 15- XOMF-KTQHAFX-1NTERCOURSE ( £ > 00 subjects. Seven out of fifteen correlations were significant for college subjects, whereas one out of fifteen correlations was significant for hospital subjects (Table 18, page 94). Correlations between judgments of emotional tones of verbal stimuli and judgments of emotional tones of verbal responses ranged from .29 to .61 with a median of .50 for college subjects, and ranged from .24 to •53 with a median of .38 for hospital sub jects. All correlations were significant for both groups of subjects (Table 19, page 95). Correlations between ratings of mood and ratings of pleasant tones of verbal responses ranged from -.03 to .23 with a median of .08 for college subjects, and ranged from -.14 to .16 with a median of .01 for hospital subjects. All correlations were insignifi cant for both groups of subjects (Table 20, page 96). Correlations between ratings of need satis faction and ratings of pleasant responses ranged from -.17 to .60 with a median of .29 for college subjects, and from -.18 to .17 with a median of .03 for hospital sub jects. Six out of fifteen correlations were significant for college subjects. None of the correlations were significant for hospital subjects (Table 20, page 96). Correlations between ratings of need satis faction and ratings of unpleasant responses ranged from -.18 to .24 with a median of .07 for college subjects, and ranged from -.26 to .15 with a median of -.14 for hospital subjects. None of the correla tions were significant for college students, and only one out of fifteen correlations was significant for hospital patients (Table 21, page 97). Correlations between ratings of emotionality and ratings of unpleasant responses ranged from -.17 to .13 with a median of .04 for college subjects, and ranged from .09 to .35 with a median of .30 for hospital sub jects. None of the correlations were significant for college subjects, whereas thirteen out of fifteen correlations were significant for hospital subjects (Table 21, page 97). Correlations between ratings of emotionality and ratings of emotional tones of verbal responses ranged from *00 to .24 with a median of .16 for college subjects, and ranged from .15 to .31 with a median of .22 for hospital subjects. One out of fifteen correlations was significant for college subjects, whereas six out of fifteen correlations were significant for hospital subjects (Table 22, page 98). CHAPTER VI DISCUSSION A discussion of the results will be made with reference to each of the seven major hypotheses. Hypothesis I stated that, in a continuous free word association experiment, it is expected that sub jects would associate a greater number of word responses to a word stimulus which represents the nutrition need presented within a family context than to the same word-need stimulus presented singly or within a neutral context, and would associate fewer word responses to word stimuli which represent the recognition and sex needs presented within a family context than to the same word-need stimuli presented singly or within a neutral context. It is also expected that subjects would associate a greater number of unpleasant word responses to recognition and sex word-need stimuli presented within a family context than to the same word-need stimuli presented singly or within a neutral context. The first prediction was partly supported by the results presented in Table 9, page 67, and dis cussed in Chapter V. For college subjects the family context increased the number of responses associated 102 103 with the word "nourishment" and decreased the number of responses associated with the words "recognition" and "intercourse." The same results were obtained from hospital subjects except for failure of pre diction with reference to the recognition need stimulus. The second prediction was not supported. In fact, it was partly negated. Family context decreased significantly the production of unpleasant words asso ciated with sex need stimulus among college students. These results are difficult to interpret and further research is needed. Hypothesis II stated that, in a continuous word association experiment, it is expected that subjects would associate significantly a greater number of affective word responses to a stimulus word which represents a more frustrated need than to a stimulus word which represents a more satisfied need. It is also expected that subjects with lower ratings of need satisfaction would associate significantly a greater number of affective word responses to word- need stimuli than do subjects with higher ratings of need satisfaction. Table 5, page 59, shows that the nutrition need is rated as most satisfied and the sex need as least satisfied by both experimental groups. The 104 recognition need occupies a median position which is significantly different from that of nutrition for both groups, significantly different from that of sex for hospital patients, and insignificantly differ ent from that of sex for college students. On the basis of these findings, the prediction from Hypothesis II was that the greatest number of affective responses should be associated with the sex need stimulus and the smallest number of affective responses should be associated with the nourishment need stimulus. The number of affective responses to the recognition need stimulus should occupy a median point between that given to the sex need stimulus and that given to the nourishment need stimulus. Results of the experiment support this pre diction. Findings follow the indicated trend. However, the correlations between need satisfaction values and values of affective responses were not significant. This fact might indicate that the above reported findings are not related to the need satis faction variable. It might also be due to the limitations of the rating method and to the narrow range of ratings. The second prediction which was made from Hypothesis II concerns the difference between the more satisfied subjects and the less satisfied sub jects with respect to their production of affective responses in association to word-need stimuli. The more satisfied subjects in this experiment are repre sented by college students and the more frustrated subjects are represented by hospital patients, as shown by the results presented in Table 4, page 58. According to the stated hypothesis, hospital patients should associate a significantly greater number of more intense affective responses than college students. This prediction was supported by the results which are presented in Table 14, page 87, where the sums of pleasant and unpleasant responses, as well as the sums of emotional responses, were compared. The difference between the means of sums of ratings of pleasant and unpleasant responses was significant beyond the .01 level. Similarly, the difference between the means of sums of ratings of emotional responses was significant beyond the .01 level. Hospital patients did associate a greater number of more intense pleasant, unpleasant, and emotional word responses to word-need stimuli than college students. Hypothesis 111 stated that, in a continuous free word association experiment, subjects with a more pleasant mood are expected to associate a 106 significantly greater number of pleasant words to verbal need stimuli presented within a pleasant context than do subjects with a less pleasant mood. Those with a more unpleasant mood are expected to associate a significantly greater number of unpleasant words to verbal need stimuli presented within an unpleasant context than do those with a more pleasant mood. On the other hand, Hypothesis IV stated that, in a continuous free word association, subjects with a more unpleasant or unhappy mood are expected to associate a significantly greater number of pleasant words to verbal need stimuli presented within either pleasant or unpleasant contexts than do subjects with a happier mood. Results presented in Tables 12 and 13, pages 85 and 86, clearly negate Hypothesis III and support Hypothesis IV. Hospital subjects who are of an unhappier mood than college students, as shown by the results of the mood rating scale and more sig nificantly by the Guilford-Zimmerman Emotional Stability vs Depression Scale, associated consis tently and significantly more pleasant words to word-need stimuli than did college subjects. The production of unpleasant words by hospital subjects 107 was consistently of smaller magnitude. However, it should be considered that the above findings, though not significantly different from those of college subjects, may be due to the special group of patients used in this experiment. The psychodynamics of sexual deviants whose marked history of pleasure outlets have brought them into conflict with the law, may be different from the psychodynamics of other psychiatric groups tested by previous inves tigators (Chapter II). The particular hospital group used in this study rated the word stimuli as more pleas ant, and this in itself may account for their greater number of pleasant associates. It is conceivable that other psychiatric groups might rate sex stimuli as being unpleasant, and might associate a smaller number of pleasant responses and a greater number of unpleas ant responses. The congruency of mood and response has been repeatedly demonstrated by Bousfield (19) and White (35), and therefore cannot be discarded solely on the basis of the present findings. As in the case of studying the effects of need satisfaction on the production of pleasant word responses, failure in obtaining significant correla tions between ratings of mood and ratings of pleasant tones of verbal responses to word-need stimuli, which 108 are presented in Table 20, page 96, may support Young's (37) point of view that judgments and pro ductions of affective words may be independent of personal dynamic experiences. It could also be interpreted in terms of the low reliability of ratings or in terms of restricted range of obtained scores. Hypothesis V stated that, in a continuous free word association experiment, subjects are expected to associate a greater number of pleasant word responses to a word-need stimulus presented within a pleasant context than to the same word-need stimulus presented singly or within an unpleasant or a neutral context. Similarly, subjects are expected to asso ciate a greater number of unpleasant word responses to a word-need stimulus presented within an unpleas ant context than to the same word-need stimulus presented singly or within a pleasant or a neutral context. Predictions made for this hypothesis were: (1) that significantly more pleasant word responses would be associated to word-need stimuli presented within a pleasant context than pleasant word responses associated to the same word-need stimuli presented within unpleasant and neutral contexts; (2) that significantly more unpleasant word responses would be 109 associated to word-need stimuli presented within an unpleasant context than word responses associated to the same word-need stimuli presented within a pleasant or a neutral context; and (3) significantly more neutral word responses would be associated to word- need stimuli presented within a neutral context than neutral word responses associated to the same word- need stimuli presented within a pleasant or an unpleasant context. Examination of results presented in Table 10, page 70, shows that all of the three predictions are supported to a much greater than chance degree. Significant differences between experimental condi tions were obtained in a larger number of instances with reference to the effect of the unpleasant context than with respect to the pleasant and neutral contexts. The neutral context was the least effective in facilitating its similar responses. The general trend was in line with the predicted results, but actual findings were such that the second prediction was significantly supported in only eight out of twelve instances, the first prediction was signifi cantly supported in seven out of twelve instances, and the third prediction was significantly supported in only two out of twelve instances. It should also 110 be noted that seven out of fifteen correlations between judgments of pleasantness of verbal stimuli and judg ments of pleasantness of associated verbal responses were significant for college subjects, whereas only one out of fifteen such correlations was significant for hospital subjects (Table 18, page 94). It is difficult to account for these results. One interpretation may be made in terms of the low reliability of judgments of affective tones of verbal material; another interpretation may be that a restricted range of scores such as obtained in this experiment lowered the correlation values. Hypothesis VI stated that, in a continuous free word association experiment, subjects with higher ratings of emotionality, reacting to word-need stimuli, should associate a significantly greater number of emotional words with greater degree of intensity than subjects of lower ratings of emotionality, reacting to the same stimuli within the same time units. The results which are presented in Table 15, page 88, show that there was no significant difference between hospital and college subjects in their pro duction of emotional words in association to word-need stimuli. However, hospital patients who rated them selves higher on emotionality than did college students Ill showed a trend of associating consistently more intense emotional words than college students. The means of sums of responses given by each group dif fered significantly beyond the .01 level. Correlations between ratings of emotionality and ratings of emotional tones of verbal responses, which are presented in Table 22, page 98, give additional insight in evaluating the predictions which were derived from Hypothesis VI. For college subjects, none of the fifteen correlations were sig nificant; for hospital subjects, six out of fifteen correlations were significant. Of special interest is the fact that the correlations between ratings of emotionality and ratings of emotional tones of verbal responses associated with word-need stimuli presented singly were all significant for hospital subjects. These findings suggest that the predictive value of ratings of emotionality with reference to production of emotional responses is greater for the higher emotional group than for the lower emotional group; a suggestion which has been noted in clinical writings of investigators such as Freud (2). Also relevant to the discussion of Hypothesis VI is another finding which is presented in Table 21, page 97. Thirteen out of fifteen correlations between 112 ratings of emotionality and ratings of unpleasant responses were significant for hospital patients, whereas only one out of fifteen such correlations was significant for college students. These results suggest that ratings of emotionality are more reliable predic tors of production of unpleasant responses for subjects with higher ratings of emotionality than for subjects with lower ratings of emotionality. It is this finding which might have been implied by clinical observations made by Freud and modern clinical psychologists (3, 9). Hypothesis VII stated that, in a continuous free word association experiment, it is expected that sub jects should associate significantly a greater number of more intense emotional word responses to verbal stimuli with higher ratings of intensity of emotional tones than to verbal stimuli with lower ratings of intensity of emotional tones. On the basis of the results presented in Table 5, page 59, where sex need stimulus was judged as significantly more emotional than recognition and nutrition need stimuli, and where the recognition need stimulus was judged as significantly more emotional than the nutrition need stimulus, the following pre dictions were made: (1) significantly more emotional words with greater intensity should be elicited by 113 the stimulus word "intercourse” than by either of the two stimulus words "recognition" and "nourishment," and (2) significantly more emotional words with greater intensity should be elicited by the stimulus word "recognition" than by the stimulus word "nourishment." According to the results which are presented in Table 11, page 73, the two predictions were partly supported. Among college students, significantly more subjects associated a greater number of emotional words with the stimulus words "intercourse" and "recognition" than with the stimulus word "nourishment." No sig nificant difference was found among associations to the stimulus word "intercourse" and associations to the stimulus word "recognition." Among hospital patients, significantly more subjects associated a greater number of emotional words to the stimulus word "intercourse" than to the stimulus words "recognition" and "nourishment." No significant difference was found among associations to the stimulus word "recog nition" and associations to the stimulus word "nour ishment ." However, the evidence which is obtained from examining the correlations between judgments of emotional tones of verbal stimuli and judgments of emotional tones of associated verbal responses 114 decidedly supports Hypothesis VII. These correlations, which are presented in Table 19, page 95, are all significant for both college and hospital subjects. These findings are more important for evaluating Hypothesis VII than those reported above, and serve as further evidence to Jacobs's contention (26) that emotional words are more similar to each other and therefore more likely to be elicited by the stimulus to some other emotional words than to the non- emotional. To sum up the discussion of the results of this experiment, the final conclusions which may be derived from the analyses of the data and the evaluation of the seven major hypotheses may be stated as follows: 1. There is evidence to support the definition of "set” as a facilitating condition for production of responses which are similar and related to set characteristics. 2. The relationship between judgments of affective attributes of verbal stimuli and affective attributes of associated verbal responses seems to be different from, and greater than, the relationship between judgments of affective attributes of personality and Judgments of affective attributes of verbal responses. As a consequence, ratings of affective attributes of verbal stimuli may be considered as better predictors of affective verbal responses than ratings of personality traits. The predictive value of ratings of emotion ality with reference to production of emotional and unpleasant verbal responses is greater for the more emotional subjects than for the less emotional subjects. Ratings of emotional tones of verbal stimuli are more reliable and better predictors of production of like responses than ratings of pleasant tones of verbal stimuli. CHAPTER VII SUMMARY Purpose of the study. It was the purpose of this study to investigate the facilitation function of affective sets by the use of the continuous free word association method. Facilitation) in this context, means (1) that a stimulus word might elicit a greater number of word responses than another word stimulus, and (2) that a stimulus word which possesses a certain attribute might elicit a greater number of word responses possessing this particular attribute than does another stimulus word which has a lesser degree of the same attribute. To investigate the facilitation function of affective sets seven hypotheses, in the form of experi mental predictions, were derived from examining previous experimental and theoretical studies. These hypotheses were: Hypothesis I: the Family Context Hypothesis. In a continuous free word associa tion experiment, it is expected that subjects would associate a greater number of word responses to a word stimulus which represents 116 117 the nutrition need, presented within a family context, than to the same word-need stimulus pre sented singly or within a neutral context; and would associate with fewer word responses to word stimuli which represent the recognition and sex needs, pre sented within a family context, than to the same word-need stimuli presented singly or within a neutral context. It is also expected that sub jects associate a greater number of unpleasant word responses to recognition and sex word-need stimuli presented within a family context than to the same word-need stimuli presented singly or within a neutral context. Hypothesis II: the Need Satisfaction Hypothesis. In a continuous free word asso ciation experiment, it is expected that subjects would associate significantly a greater number of 118 affective word responses to a stimulus word which represents a more frustrated need than to a stimulus word which represents a more satisfied need. It is also expected that sub jects with lower ratings of need satisfaction would associate significantly a greater number of affective word responses to word-need stimuli than do subjects with higher ratings of need satisfaction. Hypothesis III: the Mood Hypothesis. In a continuous free word asso ciation experiment, subjects with more pleasant mood are expected to associate signifi cantly a greater number of pleasant words to verbal need stimuli presented within a pleasant context than do subjects with more unpleasant mood. Those with more unpleasant mood are expected to associate signifi 119 cantly a greater number of unpleasant words to verbal need stimuli presented within an unpleasant context than do those with more pleasant mood. Hypothesis IV: the Mood Hypothesis. In a continuous free word asso ciation experiment, subjects with more unpleasant or unhappy mood are expected to associate significantly a greater number of pleasant words to verbal need stimuli presented within either pleasant or unpleasant contexts than do subjects with happier mood. Hypothesis V: the Stimulus Hedonic Tone Facilitation Hypothesis. In a continuous free word associa tion experiment, subjects are expected to associate a greater number of pleasant word responses to a word-need stimulus presented within a pleasant context than to the same word-need stimulus pre 120 seated singly or within an unpleasant or a neutral context. Similarly, subjects are expected to associate a greater number of unpleasant word responses to a word-need stimulus presented within an unpleasant context than to the same word-need stimulus presented singly or within a pleasant or a neutral context. Hypothesis VI: the Emotionality Hypothesis. In a continuous free word asso ciation experiment, subjects with higher ratings of emotion ality, reacting to word-need stimuli, should associate significantly a greater number of emotional words with greater degree of intensity than sub jects of lower ratings of emotionality reacting to the same stimuli within the same time units. 121 Hypothesis VIIj the Stimulus Emotional Tone Facilitation Hypothesis, In a continuous free word asso ciation experiment, it is expected that subjects should associate significantly a greater number of more intense emotional word responses to verbal stimuli with higher ratings of intensity of emo tional tones than to verbal stimuli with lower ratings of intensity of emotional tones. Experimental procedures. This study was designed to explore the influence of three varying subject set characteristics and three varying experimental set conditions on several consequences of set. The subject set characteristics were studied by choosing subjects who differed in the amount to which they could be characterized as happy or unhappy, as emotional or unemotional, and as satisfied or frustrated in their needs for nutrition, recognition, and sex. The experi mental set conditions were studied by presenting verbal stimuli differing in pleasantness, emotion evoking properties, and presumably in the capacity to evoke various combinations of need-associated responses. 122 Consequences of set were the total number of associated word responses; the number of pleasant, unpleasant, neutral, emotional, and unemotional word responses; and the sum of weighted scores of either pleasant, unpleasant, or emotional tones of verbal responses associated within a two-minute time unit. Fifty-two male adult college students and sixty- nine male adult college educated hospitalized sexual deviants participated in this experiment. They were tested and compared with reference to intelligence; verbal fluency; ratings of mood; ratings of emotion ality; ratings on the Guilford-Zimmerman Temperament Survey; ratings of satisfaction of the needs for nutrition, recognition, and sex; and production of affective and non-affective words associated to word- need stimuli presented singly or within a family context, a pleasant context, an unpleasant context, and a neutral context. Subjects did not differ significantly with respect to intelligence and verbal fluency, but differed significantly with respect to all other criteria of comparison. Results Hypothesis I was partly supported. Hypothesis II was supported. 123 Hypothesis III was negated. Hypothesis IV was supported. Hypothesis V was partly supported. Hypothesis VI was partly supported. Hypothesis VII was supported. Conclusions 1. There is evidence to support the definition of a "set” as a facilitating condition for production of responses which are similar and related to set characteristics. 2. Ratings of affective attributes of verbal stimuli are better predictors of associated affective verbal responses than ratings of personality traits. 3. Ratings of emotionality are better predictors of emotional and unpleasant verbal responses for the more emotional subjects than for the less emotional subjects. 4. Ratings of emotional tones of verbal stimuli are more reliable and better predictors of production of like responses than ratings of pleasant tones of verbal stimuli. B I B L I O G R A P H Y BIBLIOGRAPHY 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. Books Cameron, N. A. The psychology of behavior dis orders, a behavioral interpretation. Boston: Houghton Mifflin, 194^1 Freud, S. Collected papers. Vol. I. New York, London: Hogarth Press and the Institute of Psychoanalysis, 1949. Guilford, J. P. Personality. New York: McGraw- Hill, 1959. _______. Psychometric methods. 2nd ed. New York: McGraw-Hill, 1954. _______. Fundamental statistics in psychology and education. 3rd ed. New York: McGraw- Hill, 1956. Hilgard, E. R. Psychoanalysis as science. Stanford, Calif.: Stanford Univ. Press, 1952. Johnson, D. M. The psychology of thought and .judgment. New York: Harper, 19J>5. Krech, D., and Crutchfield, R. Elements of psychology. New York: Alfred Knopf, 1958. McClelland, D, C. Personality. New York: Sloane, 1951. ' The achievement motive. New York: Appleton-Century-Crofts, 1953. Murray, H. A. Explorations in personality: a clinical and experimental study of fiity~~men of college age. New York: Oxford Univ. Press, 1938. Osgood, Co E. The measurement of meaning. Urbana: Univ0 ot Illinois Press, 1957. ______ _. Method and theory in experimental psychology. New York: Oxford Univ. Press, 1953. 125 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 126 Rapaport, D. Emotions and memory. New York: International Universities Press, 1950. Ruckmick, C. A. The psychology of feeling and emotion. New York: McGraw-Hill, 19367 Sears, R. Snrvey of objective studies of psycho analytic concepts. New York: Social Science Research Council, 1943. Siegel, S. Nonparametric statistics for behavioral sciences. New York: McGraw-Hill, 1956. Woodworth, R. S., and Schlossberg, H. Experimental psychology. New York: Holt, 1954. Periodicals Bousfield, W. A. The relationship between mood and the production of affectively toned associates. J. gen. Psychol.. 1950. 42. 67-85. . An empirical study of the production ot affectively toned items. J. gen. Psychol., 1944, 30, 205-215. Brozek, J., et al. A quantitative study of perception and association in experimental semi-starvation. J. Pers., 1950-1951, 19, 245-264. Gibson, J. A. A critical review of the concept of set. Psychol. Bull.. 1941, 38, 781-817. Hull, C., and Lugoff, L. S. Complex signs in diagnostic free association. J. exp. Psychol., 1921, 4, 11-136. Hunt, W. A. Ambiguity of descriptive terms of feeling and emotion. Amer. J. Psychol., 1935, 47, 165-166. _______. A critical review of current approaches to affectivity. Psychol. Bull., 1939, 36, 807-810. 127 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. Jacobs, A. Formation of new associations to words selected on the basis of reaction time-GSR combination. J. abnorm. soc. Psychol.. 1955. 51, 371-377. Jung, C. G. On psychophysical relations of the associative experiment. J. abnorm. Psychol.. 1907, 1, 247-255. _______• The associative method. Amer. J. Psychol.. 1910. 21, 238-249. Kohs, S. C. The association method in relation to the complex and complex indicators, Amer. J. Psychol.. 1914, 25, 544-594. McGinnes, E. Emotionality and perceptual defense. Psychol. Rev.. 1949, 56, 244-251. Peters, H. The judgmental theory of pleasantness and unpleasantness. Psychol. Rev.. 1935. 42. 354-386. Postman, L., and Crutchfield, R. The interaction of need, set, and stimulus structure in a cognitive task. Amer. J. Psychol., 1952, 65, 196-217. Sines, L., and Simmons, H. The Shipley-Hartford Scale and the Dopplet Short Form as estimates of WIAS I.Q. in a state hospital population. J. clin. Psychol.. 1959, 15, 452-453. Washburn, M. F., et al. Memory revival of emotions as a test of emotional and phlegmatic tempera ments. Amer. J. Psychol.. 1925, 36, 456-459. White, M. M. Some factors influencing recall of pleasant and unpleasant words. Amer. J. Psychol., 1936, 48, 134-139. Wisp6, L. G. Physiological need, verbal frequency and word association. J. exp. Psychol.. 1953, 46, 229-234. Young, P. T. A study upon the recall of pleasant and unpleasant words. Amer. J. Psychol.. 1937, 49, 581-596. APPENDIX A THE WORD FLUENCY TEST, FORM A THE WORD FLUENCY TEST, FORM A The Word Fluency test was prepared by Paul R. Christensen and J. P* Guilford. Factors Involving Production b£ Units Word fluency: The ability to produce rapidly words fulfilling specified symbolic require ments. Word Listing I (E writes words, each containing a specified letter.) Two items of information can be mentioned con cerning predictions from scores for word fluency. They have been found related to creative performance of students of sciences and of arts in college. They have also shown small correlations (.2 to .3) with leadership criteria. The reasons for these findings are not very obvious and call for specula tion. In a highly speeded test of word fluency, possibly a motivational component, rather than the more obvious aptitude factor, is responsible for the relationships. (3:381-382) The following directions are given on the Word Fluency test: In this test you are to write words that contain a certain letter of the alphabet. This will be a different letter in each item of the test. SAMPLE ITEM: Write words containing the letter 0. load________ provide ______________ pot ' fought ______________ over________ loss_______ ______________ too 129 130 All the words written above contain the letter "0" at least once. WAIT FOR THE SIGNAL BEFORE TURNING THIS PAGE. Avoid using a word more than once; avoid even different forms of the same word, such as "bond" and "bonded." Your score will be the number of words that you write containing the given letter during limited time, so work rapidly. There are two parts to this test. You will have 2 minutes for each part. Are there any questions? STOP HERE. WAIT FOR FURTHER INSTRUCTIONS. APPENDIX B THE SHIPLEY-HARTFORD LIVING INSTITUTE SCALE THE SHIPLEY-HARTFORD LIVING INSTITUTE SCALE This test is made up of two parts: part one is a vocabulary test, and part two is a reasoning test. Instructions for part one: In the test below, the first word in each line is printed in capital letters. Opposite it are four other words. Draw a line under the one word which means the same thing« or most nearly the same thing, as the first word. A sample has been worked out for you. If you don't know, guess. Be sure to underline the one word in each line that means the same thing as the first word. SAMPLE: LARGE red big silent wet Instructions for part two: Complete the following. Each dash ( ) calls for either a number or a letter to be filled in. Every line is a separate item. Take the items in order, but don't spend too much time on any one. SAMPLE: (1) 1 2 3 4 5 6 Evaluation WAIS and Shipley scales were intercorrelated, and it was found that Shipley total score was the best predictor of WAIS full scale IQ (r = .90). . . . Because of the ease and economy of administer ing and scoring . . . , a conversion table was provided whereby WAIS full scale IQ equivalents may be obtained for Shipley total scores. (33:453) The conversion table mentioned above is on the following page. 132 S-H total WAIS full score scale IQ S-H total score 80 128 56 79 127 55 78 127 54 77 126 53 76 125 52 75 124 51 74 124 50 73 123 49 72 122 48 71 121 47 70 121 46 69 120 45 68 119 44 67 118 43 66 118 42 65 117 41 64 116 40 63 115 39 62 115 38 61 114 37 60 113 36 59 112 35 58 112 34 57 111 33 WAIS full scale IQ S-H total WAIS full score______scale IQ 110 110 109 108 107 107 106 105 104 104 103 102 101 101 100 99 98 98 97 96 95 95 94 93 92 92 91 90 89 89 88 87 86 86 85 84 83 83 82 81 80 80 79 78 77 77 76 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 133 APPENDIX C THE GUILFORD-ZIMMERMAN TEMPERAMENT SURVEY THE GUILFORD-ZIMMERMAN TEMPERAMENT SURVEY The GZTS was designed to cover in a single inventory most of the 13 traits of the three original inventories. Another aim was to reduce or to avoid some of the higher intercorrelations found in the original ones. The highest inter correlation had been between scores for traits C and D. Those two traits are therefore repre sented by a single score in the GZTS. The traits, with some slight changes in symbols and names, are: G General activity R Restraint vs. rhathymia A Ascendance S Sociability E Emotional stability (combination of D and C) 0 Objectivity F Friendliness (previously called "agreeableness") T Thoughtfulness (previously called P "thinking introversion") Personal relations (previously called M "cooperativeness") Masculinity (of emotions and interests) Three validation scoring keys have been developed for use with the GZTS inventory. A gross-falsifica- tion score is designed to detect the examinee who willfully attempts to make good scores. A subtle- falsification score is designed to detect the examinee who perhaps unwittingly achieves the same kind of result. A careless-deviancy score is designed to detect the examinee who responds in an erratic manner and shows this by giving an unusually large number of extremely rare responses. Some Evaluation of the Guilford Inventories The split-half estimates of reliability of the scores from the three original inventories range from .80 to .94 and of scores from the GZTS, .75 to .87. Some information is available to answer the ques tion of whether the scores measure the traits they 135 136 were designed to measure. The 13 scores from the original inventories correlate moderately with self-ratings and with ratings made by associates. The ratings are of unknown validity and may well be less valid than the scores. A recent factor analysis involving a liberal sampling of all items in the three inventories provides a general verification of earlier conclusions regarding the nature of the factors and the kinds of items needed to measure them. Although these inventories were not designed for the purpose of discriminating abnormals from normals or of aiding in the diagnosis of patho logical groups, there is some information in this connection, much of it in reasonable directions. Psychotics generally were found to be less sociable and less reflective. Psychoneurotics were char acterized by depression, emotional instability, and low rhathymia (overly self-controlled). Delinquents and psychopathic personalities were high on rhathymia, and schizophrenics and organic cases tended to be low on sociability. The suggestion that there is a relation of the scores for D (depression) and C (cycloid) to neurotic tendency is supported by a correlation of the sum of the two with the Maudsley Medical Questionnaire equal to .88. It is also reported that the E (emotion ality score) from the GZTS has correlated .80 with the Bernreuter (BPI) score for neurotic tendency. In spite of these reasonable relationships, it would be premature to propose that patterns of the scores from these inventories might be used as aids in the diagnosis of pathology and of category of pathology. Although it may be more reasonable to think that abnormals are more accurately described in terms of concepts that pertain to the normal population than to think that pathological concepts apply to normals, the relationship of normal traits to pathology is not entirely clear. A good hunch is that the essence of pathology involves much more. Pratt (1952) concludes, however, that scores for the factors S, D, C, R, M, and N have significant dis criminating power among ten nosological (diagnostic) categories. From his findings, it would seem that the syndrome types that characterize pathological groups may be describable in part in terms of specified patterns of such trait scores. (3:185- 186) APPENDIX D THE STATISTICAL PROGRAMS AND TESTS QUESTIONNAIRE ANALYSIS PROGRAM I FOR THE IBM-709 (Mod. 4) This program was developed by John R. B. Whittlesey for the Western Data Processing Center, University of California at Los Angeles. It is usually referred to as WD QUAP-4. Two quotations from the author may serve to describe and evaluate this program. Introduction This program, for the IBM-709 Data Processing System, has been prepared to eliminate the clerical effort required to tabulate and summarize the results of questionnaires and surveys of the multiple-choice answer variety. Each question naire may have from as many as 66 questions each with 100 possible choices up to 999 questions each with five or less choices. The questionnaire may be filled out by as many as 100,000 participants. Classification groups may be established and the results for the various classifications may be prepared separately, or as a composite group. Output is from magnetic tape onto the IBM-717 printer. There is a potential of 36 possible output formats, including formats for percentages of responses at each level, frequency counts, partial frequency counts, means and standard deviations, the Kolmogorov-Smirnov and Chi Square tests between pairs of classification groups, and other standard statistical measures. The Kolmogorov-Smirnov Two-Sample Test Function and Rationale The Kolmogorov-Smirnov two-sample test is a test of whether two independent samples have been drawn from the same population (or from populations with 139 140 the same distribution). The two-tailed test is sensitive to any kind of difference in the dis tributions from which the two samples were drawn— differences in location (central tendency), in dispersion, in skewness, etc. The one-tailed test is used to decide whether or not the values of the population from which one of the samples was drawn are stochastically larger than the values of the population from which the other sample was drawn, e.g., to test the prediction that the scores of an experimental group will be "better" than those of the control group. Like the Kolmogorov-Smirnov one-sample test, this two-sample test is concerned with the agree ment between two cumulative distributions. The one-sample test is concerned with the agreement between the distribution of a set of sample values and some specified theoretical distribution. The two-sample test is concerned with the agreement between two sets of sample values. If the two samples have in fact been drawn from the same population distribution, then the cumula tive distributions of both samples may be expected to be fairly close to each other, inasmuch as they both should show only random deviations from the population distribution. If the two sample cumula tive distributions are "too far apart" at any point, this suggests that the samples come from different populations. Thus a large enough deviation between the two sample cumulative distributions is evidence for rejecting H . Summary of procedure. These are the steps in the use of the Kolmogorov-Smirnov two-sample test: 1. Arrange each of the two groups of scores in a cumulative frequency distribution, using the same intervals (or classifications) for both dis tributions. Use as many intervals as are feasible. 2. By subtraction, determine the difference between the two sample cumulative distributions at each listed point. 141 3. By inspection, determine the largest of these differences; this is D. For a one-tailed test, D is the largest difference in the predicted direction. 4. The method for determining the significance of the observed D depends on the size of the samples and the nature of : a. When n. * n„ « = N, and when N .5? 40 is used. It gives critical values of KD (the numerator of D) for various levels of significance, for both one-tailed and two-tailed tests. b. For a two-tailed test, when n^ and n2 are both larger than 40 .... In such cases it is not necessary that n, = n^. Critical values of D for any given large values of n^ and n^ may be computed .... c. For a one-tailed test where n. and n„ are large, the value of X with dr » 2 which is associated with the observed D is computed .... The significance of the resulting value of X 2 may be determined .... This chi-square approximation is also useful for small samples with n^ £ n^, but in that application the test is conservative. If the observed value is equal to or larger than that given in the . . . for a particular level of significance, H may be rejected at that level of significance. Power-Efficiency When compared with the t test, the Kolmogorov- Smirnov test has high power-efficiency (about 96 per cent) for small samples. It would seem that as the sample size increases the power-efficiency would tend to decrease slightly. The Kolmogorov-Smirnov test seems to be more powerful in all cases than either thex^ test or the median test. 142 The evidence seems to indicate that whereas for very small samples the Kolmogorov-Smirnov test is slightly more efficient than the Mann-Whitney test, for large samples the converse holds. (17:127-128, 135-136) This program was prepared by Philip. A. Cramer. To extract the following quantities from raw data; means, standard deviations, correlation coefficients, and covariance terms; and to save these results for further study by multiple regression analysis, factor analysis, partial correlation analysis, or analysis of covariance. The program will calculate these quantities for "M" variables with "N" observations, subject to certain restrictions. The program calculates the following quantities: CORRELATION ANALYSIS Purpose Method Means M) Standard Deviations M) 143 Covariance N N N ( r xn) ( d kci x k=l J D D = N2 - N or N2 (for i = 1,2, . (for j = i,i + 1 M) * M) Correlation Coefficients (for i » 1,2, . (for j = i,i + 1 M) M) It is possible to get decimal output of the raw 3ums and sums of squares and cross products, means, standard deviations, covariances, correla tion coefficients. For further analysis, it is possible to produce a binary tape of the means, standard deviations, and either the matrix of correlation coefficients, the matrix of covari ances, or both. All computations in the summations and computa tions of means and standard deviations, and matrix of covariance are performed in fixed point. Double precision is used where necessary. The correlation coefficients are calculated in double precision floating point. All intermediate steps in the calculations are performed with no rounding or truneating. Accuracy 144 THE SIGN TEST Function The sign test gets its name from the fact that it uses plus and minus signs rather than quantita tive measures as its data. It is particularly useful for research in which quantitative measure ment is impossible or infeasible, but in which it is possible to rank with respect to each other the two members of each pair. The sign test is applicable to the case of two related samples when the experimenter wishes to establish that two conditions are different. The only assumption underlying this test is that the variable under consideration has a continuous distribution. The test does not make any assump tions about the form of the distribution of differences, nor does it assume that all subjects are drawn from the same population. The different pairs may be from different populations with respect to age, sex, intelligence, etc.; the only requirement is that within each pair the experi menter has achieved matching with respect to the relevant extraneous variables. . . . (17:68) Power-Efficiency The power-efficiency of the sign test is about 95 per cent for N = 6, but it declines as the size of the sample increases to an eventual (asymptotic) efficiency of 63 per cent. For dis cussions of the power-efficiency of the sign test for large samples, see Mood (1954) and Walsh (1946). (17:75)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The Effects Of Prior Part-Experiences On Visual Form Perception In The Albino Rat

PDF

A Study Of The Effectiveness Of Teaching Methods Of Study To Selected High School Freshmen

PDF

Behavioral Seriousness And Impulse-Control Balance In Delinquency

PDF

Positive And Negative Verbal Reinforcement In The Conditioning Of The Verbal Behavior Of Adolescent Underachieving Delinquents With Dependent, Neutral, And Aggressive Stimuli

PDF

Predicting Success In The Study Of Descriptive Linguistics

PDF

Verbal Reports Of Emotional States And Onsets And Offsets Of Conditioned Stimuli

PDF

A Study Of The Effects Of Generalized Expectancies Upon Accuracy Of Interpersonal Perception

PDF

A Study To Compare The Effectiveness Of Individual And Group Counseling Approaches With Able Underachievers When Counselor Time Is Held Constant

PDF

The Effects Of Group Experiences On The Aged

PDF

Conflict And Generalized Conflict In Verbal And Motor Responses

PDF

A Study Of Recall Set Utilizing Early Childhood Recollections And Recent Recollections

PDF

The Relationship Of Dependency To Verbal Learning Without Awareness

PDF

A Problem-Centered Approach To Introductory Psychology

PDF

Anxiety Level And The Repression-Sensitization Dimension In Desensitization Therapies

PDF

The Effects Of Sex, Assigned Therapist Or Peer Role, Topic Intimacy, And Expectations Of Partner Compatibility On Dyadic Communication Patterns

PDF

An Experimental Study Of The Effect Of Anxiety And Stress On The Group Intelligence Test Performance Of Elementary School Children

PDF

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

PDF

A Study Of The Relationship Of Temperament Variables To The Ability To Make Certain Judgments Of Emotional Behavior

PDF

Transfer Of The Partial Reinforcement Extinction Effect Across Tasks In Normal And Retarded Boys

PDF

Validity Concomitants Of Various Scoring Procedures Which Attenuate The Effects Of Response Sets And Chance

Asset Metadata

Creator Abou-Ghorra, Ibrahim Mansour (author)

Core Title A Study Of Affective Sets: The Effects Of Family And Non-Family Verbal Contexts On Word-Need Stimuli In A Word Association Experiment With Reference To Pleasant And Emotional Tones Of Associated...

Degree Doctor of Philosophy

Degree Program Psychology

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

Tag OAI-PMH Harvest,Psychology, clinical

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Jacobs, Alfred (committee chair), Lefever, David Welty (committee member), Warren, Neil D. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-231041

Unique identifier UC11358037

Identifier 6106274.pdf (filename),usctheses-c18-231041 (legacy record id)

Legacy Identifier 6106274.pdf

Dmrecord 231041

Document Type Dissertation

Rights Abou-Ghorra, Ibrahim Mansour

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