University of Southern California Dissertations and Theses

An Experimental Study Of Aggression Habit Patterns And Sex Role

(USC Thesis Other)

An Experimental Study Of Aggression Habit Patterns And Sex Role

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content AN EXPERIMENTAL STUDY OF AGGRESSION HABIT PATTERNS AND SEX ROLE by Eric Guy Thompson 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) June 1965 UNIVERSITY O F SO U TH ER N CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LO S ANGELES. CALIFO RNIA 8 0 0 0 7 This dissertation, written by under the direction of h..iJL.Dissertation Com mittee, and approved by all its members, has been presented to and accepted by the Graduate School, in partial fulfillment of requirements for the degree of D O C T O R OF P H IL O S O P H Y Dean Date......J . U l i e . . .1965......... DISSERTATION COMMITTEE This work is dedicated to my wife, Linda, and to the memory of my grandmother, Inez Tunison. ACKNOWLEDGEMENTS The writer wishes to make the following acknow ledgements : To Dr. Alfred Jacobs, for consenting to chair the writer's Dissertation Committee, for his perceptive criticism of the present study at all stages of development, and his ready encouragement, without which the work could not have been completed. To Dr. Georgene Seward, for consenting to serve on the writer's Dissertation Committee, and for her many valuable suggestions in the area of her specialization, all of which have contributed to improvement of the present study. To Dr. Aerol Arnold, for graciously consenting to serve on the writer's Dissertation Committee, and his underStending acceptance of the many inconveniences entailed in that role. To Dr. Seymour Feshbach, University of California at Los Angeles, for providing the set of slides used to V gather retponaea, and without which it would have been impossible to complete the experiment. To Dr. Eugene Walker, Westmont College, Santa Barbara, Califoraia, for making possible the test of hypotheses regarding religious background by providing subjects for this study from Westmont College. To Dr. William Domhoff, California State College at Los Angeles, for providing subjects used in the standardization of stimulus materials. To Dr. Benjamin Siegel, Chief Psychologist, Camarillo State Hospital, Camarillo, California, for making time and facilities available, and providing the climate of encouragement needed to spur the writer to completion of these labors. TABLE OF CONTENTS Page Acknowledgments................................ iv List of Tables................................ viii Chapter I. INTRODUCTION........................ 1 II. REVIEW OF THE LITERATURE ........ 6 Behavioral Theories of Aggression Frustration-Aggression Hypothesis Scott's Theory Interaction of Biochemical, Behavioral and Social Determinants Aggression-Hostility-Anger Theory of Buss Feshbach's Theory of Alternate Mediating Responses Behavioral Theory of Sex Differences in Aggression Studies of Sex Differences Cultural Factors in Sex Differences Studies of Acquisition of Aggression Patterns Studies of Imitative Learning and Identification Studies of Imitation of Aggressive Behavior III. HYPOTHESES ............................ hO IV. METHOD ................................ Subjects Materials and Administration Response measures Design Chapter V. RESULTS Page 59 Antiaocial Acceptance scale comparisons Prosocial aggression score comparisons Antisocial aggression score comparisons Total aggression score comparisons Relations between measures of attitude and behavior VI. DISCUSSION.............................. 74 VII. SUMMARY ................................ 81 LIST OF REFERENCES.......... 84 APPENDICES .................................... 93 APPENDIX A Items used in the development of the high anger and low anger instigation l i s t s............ 94- APPENDIX B Composition of high anger and low anger lists of instigating items (revised) .............. 97 APPENDIX C Descriptions of slides used in obtaining siides- anger measures ............................ 100 APPENDIX D Copy of the Emotional Situations test used to measure aggressive responses ................ 104 APPENDIX E Copy of attitude toward antisocial acts scale (AA), used to measure attitude toward antisocial aggressiveness ................ 107 APPENDIX F Checklist of words used to score slides for slides-anger scores ........................ 108 LIST OF TABLES Table Page 1. Means and Standard Deviations of Age and Education of Subject Groups ................ 46 2. Percent of Possible Anger Arousal Potential for High and Low Anger Arousing Lists of Items, Rated for Amount of Anger Arousal Potential . . . ............................ 51 3. Percent of Possible Anger Arousal Potential for High and Low Anger Arousing Lists of Items, Revised, Rated for Amount of Anger Arousal Potential .......................... 51 4. Means, Mean Differences, and Standard Errors of Mean Differences, Ratings of Anger, Emotional Situation Test Items ............ 56 5. Mean Scores and Standard Deviations for Men and Women under High and Low Conditions of Instigation to Aggression, and More and Less Religious Emphasis ........................ 59 6. Analysis of Variance of Antisocial Acceptance Scale Scores.............................. 60 7. Mean prosocial Aggression Slide Scores and Standard Deviations for Men and Women under High and Low Conditions of Instigation to Aggression, and More and Less Religious Emphasis.................................. 62 8. Analysis of Variance of "Prosocial Aggression" Scores............................. 63 9. Mean Antisocial Aggression Slide Scores and Standard Deviations for Men and Women under High and Low Conditions of Instigation to Aggression, and More and Less Religious Emphasis.................................. 65 10. Analysis of Variance of Antisocial Aggression Scores.................................... 66 viii ix Table Page 11. Mean Total Aggression Slide Responses and Standard Deviations for Men and Women under High and Low Conditions of Instigation to Aggression, and More and Less Religious Emphasis................................ 68 12. Analysis of Variance of Total Aggression Scores.................................. 69 13. Mean Emotional Situations Test Scores and Standard Deviations for Men and Women under^ligh and Low Conditions of Instigation to Aggression, and More and Less Religious Emphasis............. 70 14. Analysis of Variance of Emotional Situations Test Scores............................ 71 15. Correlations (Pearson's r) Between Anti social Acceptance Scale Scores, and Slide- Aggression Scores and Emo. Sit. Scores, for Men and Women under High and Low Conditions of Instigation, and More and Less Religious Emphasis ................ 73 CHAPTER I INTRODUCTION I The study being reported proposes to derive a behavioral theory of sex-related differences in aggression ; patterns based on studies of child development and cultural influences, and to subject this theory to experimental test. The theoretical position is limited, and maintains close reference to current behavioral theory and research in the area of aggression (e.g., Berkowitz, 1962; Buss, 1961). The popular social stereotype that men are more aggressive than women finds expression in two recent com prehensive surveys of the literature on aggression. One says: The presence of sex differences in aggression is widely accepted on the basis of everyday observations and empirical evidence. Almost without exception, j the studies that compare aggression in boys and girls have found that boys are more aggressive, especially more physically aggressive. (Buss, 1961, p. 283) { The other says: { i i There is relatively little doubt that a relation exists! between aggressiveness and . . . sex. In most species : of the anisial kingdom . . . the male tends to be more aggressive than his female counterpart. (Berkowitz, 1962, p. 267) ! Attempts to explain alleged sex differences in 1 aggression fall into two classes. Explanations in the [ first class give greatest weight to innate physiological ! differences, such as the average superiority of men in | strength and stature, or the effect of male sex hormone on level of irritability (e.g., Scott, 1958). Explanations of the second class maintain that early social experiences, j particularly in the home, are responsible for enduring habit patterns of aggressive behavior which tend to differ t according to sex role (e.g., Sears, Maccoby, and Levin, 19571 j An earlier study of psychological differences between sexes noted both the different temperamental styles of men and women in a variety of aggression-related behaviors, and the two explanatory possibilities mentioned j (Terman and Miles, 1936). The authors pointed out the need j i for "... parallel examinations of socially and racially different groups widely different in social tradition and circumstance . . , " to resolve the problem (Terman and a • • Miles, 1936, p. 450). These examinations have, to a large extent, been carried out (e.g., Barry, Bacon, and Child, 1957; Mead, 1949; Whiting and Child, 1953). Taken in con- | junction with longitudinal and cross-sectional studies of | child development (e.g., Baldwin, Kalhorn, and Breese, 1945;j Sears et al., 1957; Sears, 1961; Winder and Rau, 1962), it now seems possible to relegate innate factors to a position of relative unimportance in the determination of ) aggressive behavior in humans. This is not to say that differences between sexes in aoumnt of aggressive behavior is due solely to differential training. Phylogenetic studies of the relationship between hormones and aggressive,; dominant behavior patterns have shown that biological factors must be taken into account at all levels, including the human level (e.g., Seward, 1946). However, in human affairs what is important is not the wide individual dif ferences in potential aggressiveness, determined by bio logical factors, as much as the differences in aim, style, and meaning of aggressiveness, which are shaped by social forces, and attached to social roles, including sex roles. The study being reported argues that there are sex differences in aggressive habit patterns that are a function of a sex role, conceived as a learning process, rather than i sexuality, conceived as a function of biological factors such as hormone level. The nature of the hypothesized difference in aggressive habits is as follows: male aggression tends to be more antisocial than female aggression, and female aggression tends to be more prosociali than male aggression. The difference is conceived as being i the result of learning, and therefore does not have to conflict with the biologically based argument that men are t more aggressive than women due to differences in hormone i levels. Men could be totally more aggressive than women, and still manifest the hypothesized differences in habit patterns. I * This difference in sociality of aggression was noticed by Sears (1961), who developed scales of prosocial aggression, and found the predicted sex differences. Essentially the same attitude toward aggression as that labeled prosocial was defined in the study of the authori tarian personality, and called authoritarian aggression (Adorno, Frenkel-Brunswik, Levinson, and Sanford, 1950). Sears (1961) used juvenile subjects, in contrast to the adult subjects used by Adorno et al. (1950), who did not find significant sex differences on this variable. Pro- social aggression can be defined as a " . . . tendency to be on the lookout for, and condemn, reject and punish people who violate conventional values." (Adorno et al.. 1950, p. 228). Antisocial aggression would then be de fined as a tendency to violate conventional values and resist or defy attempts of others to check or punish such behavior. Evidence of differential acceptance of such an attitude by men and women has been demonstrated in factor analytic studies of personality inventories (e.g., Mees, 1959). There is also evidence that persons from environ ments with relatively high degree of religious emphasis, such as seminaries, will tend to he more conventional in their values than persons from settings not character istically thought of as having a deliberate religious emphasis (Adorno et al., 1950; Terman and Miles, 1936). To test this kypothesis that aggression habit patterns are related to sex role, and differ with respect to sociality or conformity, the following experiment is proposed: Two samples of adults, drawn from populations of presumably different degrees of conventionality, will each be randomly divided into groups and differentially instigated to aggression. The hypothesis can then be tested by measuring the relative differences between sexes of manifested prosocial and antisocial aggressiveness. CHAPTER II | REVIEW OF THE LITERATURE Behavioral Theories of Aggression i Frustration-Aggression Hypothesis The Frustration-Aggression (F-A) Hypothesis (Dollard, Miller, Doob, ftowrer, and Sears, 1939) defines aggression as, "(A) . . . sequence of behavior, the goal- response to which is the injury of the person toward whom it is directed." (Dollard et al.. 1939, p. 9). The behavior referred to may be overt or covert, and may not appear to be goal-oriented. Frustration is defined, as, j " . . .an interference with the occurrence of an insti- gated goal-response at its proper time in the behavior se quence ..." (Dollard et al., 1939, p. 7). The hypo thesized relationship, characterized by extraordinary simplicity and generality, is that "... the occurrence of aggressive behavior always presupposes the existence of j 1 frustration and, contrariwise, the existence of frustration! ! always leads to some form of aggression." (Dollard et al..I i 1939, p. 1). The above transitive, necessary and suf- l ficient relationship was quickly challenged on grounds, the! Antecedent, frustration, did not necessarily lead to aggression, but that alternative responses such as regres- sion, or problem solving behavior, might be the consequent ones (e.g., Levy, 1941; Maslow, 1941). The need for mod ification was admitted, and the theory was reformulated to make the likelihood of aggression a function of degree of frustration. Other, competing responses were not ruled out, as had been implied before (Miller, 1941). The mod ification did not quiet all dissatisfaction with the theory, but it is not the purpose of the present study to consider the long, tangled history of argument and experimentation. This literature is amply reviewed by others (Berkowitz, 1958; Berkowitz, 1962; Buss, 1961). Two important modifi cations that should be mentioned are the equating of anger with instigation to aggression, and inclusion of a medi ating process of interpretation of meaning or significance, as an important modifier of aggression (Berkowitz, 1962). Scott's Theory. Scott (1958) proposed a definition of aggression as, ". . . fighting . . . the act of initiating an attack." His theory draws heavily on animal studies and physio logical mechanisms as bases of explanation of aggression phenomena. His theory of human sex differences in aggres sion, and his observations on social control of aggression, | j are based on the concept of greater stale irritability being a function of male sex hormones. The view is that social control must provide outlets for male aggression, such as rought sports, and daring, socially uesful, adventures. Interaction of Biochemical. Behavioral and Social Determinants Scott (1958) and Seward (1946) discussed inter action of hormone level and aggression in different species. In reviewing animal studies across phylogenetic levels, Seward (1946) demonstrated that aggressive behavior, as an expression of dominance, is influenced in all cases by sex hormone variation. The effect of sex hormones on aggressive behavior appears to be an interaction of direct stimulation of aggressive behavior by male hormones and inhibition of aggressive behavior by female hormones. The effect of these hormonal variations is most clearcut across individuals in species of relative phylogenetic simplicity, such as fish, birds, and reptiles. In higher forms such as apes and monkeys, the degree of individual variability due to learning is greater. Personality and social pheno mena begin to appear, as in the case of chimpanzees, and typical behavior patterns of species may be differentiated more completely by analogous human typologies, e.g., howler monkeys. Maslow (1942) found suggestive evidence of a relationship between dominance behavior and typically male sex behavior, using human subjects. Interviews with college girls revealed a strong positive relationship between degree of dominance and degree of sexuality. Less I i dominant, aggressive girls not only had less sexual experi-j ence, but also expressed less sexual need fantasies. Sexual) behavior of more dominant women was found to be more aggres- i sive than that of less dominant women, but this finding did) not extend to the case of secure women. Peelings of security-insecurity were unrelated to degree of dominance. j Dominant-secure subjects were less inclined to use sex as an expression of power, and more likely to stress the cooperative aspects of sex. Similar patterns of behavior among gibbons are discussed by Seward (1946), who warns of ! the tendency to create a false dichotomy between humans and) i i animals in regard to emergent phenomena such as personality: and society. The Aggression-Hosti1itvrAnger Theory of Buss In a comprehensive review of the literature Buss (1961) put forth objections to the F-A hypothesis. He felt that Berkowitz (1958) and others were enlarging both frustration and aggression into omnibus terms (Grayson and Tolman, 1950). Antecedents of aggressive j | behavior, not clearly frustrating, were being interpreted i as frustrating. He'noted that experimental frustration ! procedures were often confounded with verbal attack on the j I subjects (e.g., Feshbach, 1955). F-A theory proponents 10 argue that verbal attack fruetrates ongoing behavior oriented to the goal of maximizing self-esteem* Gilleapie (1961) compared effecta of fruatration and verbal attack in a atudy in which men and women college atudent8 were verbally attacked, fruatrated by having a taak interrupted in another condition, and allowed to complete the taak with no verbal aaaault at the end in the control condition. Aggreaaion waa meaaured by a word aaaociation teat, acrambled aentencea teat and attitude questionnaire. Attacked subjects scored significantly higher than control subjects in all six conditions of the study. Frustrated subjects scored higher than controls in only three of the six possible conditions (2 sexes X 3 measures). It would appear that verbal attacks are stronger instigators of aggressive behavior than frustrations, defined in the ordinary sense of preventing someone from doing something he is set to do. Buss (1961) classifies antecedents of aggression as frustrations or noxious stimuli. Noxious stimuli are subdivided into attacks, where the subject is the discrimi nated object of some agent's attacking behavior, and annoy ances, in which the subject is exposed to noxious stimuli, but only fortuitously. As noted earlier, most experimental studies of aggression involve verbal assault (derogation, insult) as the presumed effective antecedent instigating condition, rather than frustration in the sense of a set of blocking operations. This distinction seems worthwhile, since the nature of the antecedents seems to have a great deal to do with the response. It is much easier to explain the results of aggression studies after the fact than to predict outcome. An example of frustration, in the re stricted sense of blocking, leading to regressive rather than aggressive behavior is the study of Barker, Dembo and Levin (1941) in which children's play styles regressed significantly when they were shown inaccessible, highly desirable toys behind a wire mesh barrier. Two attempts to preserve the concept of frustration as a broad explanation of behavior are those of Brown and Farber (1951) and Amsel (1958). The former investigators attempt to explain a wide variety of antecedent stimulus situations as giving rise to conflicting drive-mediating states. The quality of these states is then defined as "frustrating," and the state itself defines the concept of frustration. The latter investigator attempts to define extinction as frustration of the anticipation of reward or reinforcement. Studies of extinction do not suggest that instigation to aggression is an invariable concomitant, although it is a possible concomitant (e.g., Gillespie, 1961). Buss (1961) lists four' main variables that influ ence the probability of the occurrence of aggressive 12 behavior, which ha* been instigated by frustration* They i are the strength and arbitrariness of frustration, the degree of frustration tolerance of the subject, and the instrumental value of aggression as a response in the situation. Independent ratings of hypothetical degrees of frustration strength and aggression strength obtained from college students indicate a positive relationship between these two variables (Allison and Hunt, 1959). Experimental studies of interruption of a goal-oriented task at differ ent points on the goal-gradient indicate increases in what appears to be Misplaced aggression' (slamming in the plunger on a marble game) (Haner and Brown, 1955). A study utilizing number of frustrations as a variable, also lends support to the frustration-aggression relationship, except for the possible confounding of the results by the instru mental value of the response (Otis and McCandless, 1955). This instrumental value could account for the results in terms of a reinforcement contingency promoting learning: The more times the subject responded to the frustration aggressively (pushing his toy car past the experimenter's), the more times he was reinforced (winning the game), and the more likelihood of his responding in that fashion again. Arbitrariness, or, as conceptualized by F-A theo rists, interpretation (Berkowitz, 1962) was demonstrated by describing hypothetical situations in 'unexplained'' and 13 explained ways. Explanation seems to this writer to consist of structuring the antecedents of the frustrating (or even noxious) stimulus in such a way that they are con-j gruent with the values of the object of frustration, there by leading to his acquiescence. Arbitrariness was found to be directly related to degree of hypothesized anger (insti-i j gation to aggression) (Allison and Hunt, 1959; Cohen, 1955;j Pastore, 1952). The instrumental value of aggression as a response ^ to frustration is differentiated by Buss (1961) from anger or rage responses, which have as their goal infliction of pain on some object (Dollard et al., 1939) or purely ex pressive consequences (e.g., Berkowitz, 1962; Maier, 19h9; i Sears et al.. 1957). Those six items, from the Doob and Sears questionnaire which was rated for anger provokingnessj and degree of frustration by college students (Dollard et al., 1939), which were most clearly frustrating (as op- ; posed to noxious) were evaluated in terms of the likelihood of aggressive behavior being successful in removing the frustration. Those situations in which aggression would be most effective elicited higher ratings of probability of aggressive behavior than the others CBuss, 1961). Differential frustration tolerance as a determiner of aggressive reactions to frustration was demonstrated by j Block and Martin (1955) and Davitz (1952). In the former i study, children were judged in a variety of situations for ; degree of 'emotional control.' Using the Barker et al. (1941) situation, Black and Martin (1955) found children judged to have high frustration tolerance were less likely to engage in aggressive behaviors (attacking the screen, etc.). Davitz (1952) trained his subjects to be differentially aggressive, but found that when at tractive consununatory responses were interrupted (watching cartoons while eating candy), the expected differences in aggression did not occur. He suggests that his subjects had learned nonaggressive responses to frustration prior to his experiment, and that these responses were not affected by his training (Davitz, 1952). As mentioned before, Buss (1961) feels that noxious variables have a much greater likelihood of instigating aggression than pure frustrations, especially those noxious variables falling under the heading of attacks. Actually, this same position is taken from a slightly different standpoint by F-A theorists (Dollard,et al.. 1939; Berkowitz, 1962; Sears et al.. 1957) when they speak of punishment serving as an instigation to aggression, or aggression as provoking punishment. In either case they are describing attack behavior motivating an aggressive response. The response can be one of a number of alterna tives (e.g., fight, flight, inhibition, displacement, sublimation, reaction formation, etc.) but tends fairly often to be of an aggressive nature. One of the main 15 variables affecting the tendency of an attack to provoke aggression is its intensity. This is demonstrated in a study in which degree of insult and derogation was varied (McClelland and Apicella, 1945). More intense attacks (including cursing) were followed by increased aggression in the subjects. The relative cue values in terms of sex- typed appropriate behaviors could be considered to be the relevant variable in the above study. Being cursed proba bly activates a highly restricted response hierarchy in which some form of aggressive reaction is called fort in male subjects. Replication of this study with female sub jects might result in more responses of moralistic indig nation and attempts to leave the scene (escape behavior). Buss (1961> proposes differentiation of aggression* anger and hostility as responses to antecedents of frus tration and noxious stimuli. He defines aggression ast " . . . a response that delivers noxious stimuli to another organism ..." (Buss, 1961, p. 1). The concept of "intent," as in "intent to do harm" is purposely excluded on grounds that this makes the definition teleological. He discriminates between instrumental aggression, which is re inforced by extrinsic rewards, and aggression that is re inforced bjr "the stimulus of the victim suffering injury or being in pain ..." (Buss, 1961, p. 2). This latter classification is usually associated with antecedents of anger, and is the sole class of aggressive response 16 considered by F-A theorists (Berkowitz, 1962). Buss (1961) finds it convenient to define certain I logical parameters of the aggressive response, such as: Physical-Verbal, Active-Passive, and Direct-Indirect. Verbal aggression is further divided into classes of re jection and threat. In discussing anger and hostility, Buss (1961) defines anger as an autonomic, facial-skeletal response with drive properties. The drive properties stem from the internal stimuli associated with anger, or labelled as anger. When these stimuli are mild, they can serve as positive reinforcing stimuli. When they become intense, their sudden decrease, as in expressive outbursts of motoric behavior during rage, seems to be reinforcing. Berkowitz (1962) deals with anger as a feeling state that motivates aggressive behavior. It seems clear that both points of view acknowledge the existence of angry aggres sion" that inflicts pain on the subject, thereby reducing angry behavior and feelings of anger. Buss is the only theorist to single out hostility as a separate response. He views hostility as, ". . . an implicit verbal response i involving negative feelings (ill will) and negative evalu ations of people and events." (Buss, 1961, p. 12). The j term, "Observing-labe11ing response" is used to characterize! ! the effectless nature of hostility. Association of hostile i labels with anger reactions to stimuli establishes a hostile 17 response that is tine binding and persists after the feeling of anger subsides. He hypothesises this as the mediating mechanism of revenge (delayed angry aggression). When the long-absent stimuli reappear, they are observed i and evaluated, anger responses are reelicited, and approprir ate aggressive actions are undertaken. Feshbach1s Theory of Alternate Mediating Responses Feshbach (1964) proposes refinements of the dis tinction of classes of aggression made by Buss (1961). Incidental aggressive acts that inflict pain or injury without intent (with no contingency between the inflicted pain and increased response probability) are distinguished from instrumental and hostile acts of aggression. Instru mental aggression is defined by elimination as goal di rected aggression other than that having the infliction of pain or injury as its goal. The latter form is labeled hostile aggression, and the development of hostile aggres sive habits, along with their reduction, is the interest of Feshbach (1964). These conditions leading to reduction are labeled "Catharsis," but it is not necessary in this paper to consider the controversy surrounding this term (e.g., Buss, 1961; Berkowitz, 1962). Feshbach is satisfied with the developmental theories used to explain the learning of instrumental aggression (e.g., Buss, 1961; Sears et al.. 1957) in which children's aggressive behavior is largely shaped by parental reinforcements. He focusses on the means by which the goal of angry aggressive behavior shifts from hitting (an expressive action) to hurting (a hostile action) the object of aggression. He hypothesizes that the child learns by imitation and observation that social values call for retaliation in certain circumstances, and that the appropriate form of retaliation is the infliction j of pain, by whatever means, on the attacker. He follows the position of Sears et al.. (1957), that a relationship exists between punishment history and hostile aggressive tendencies. Following Worchel (1960), attacks on self esteem are seen as the primary, if not sole, instigation to hostile aggression. Alternative mediating responses are presented as a means of explaining how threats to status and frustration of self esteem enhancing behaviors are actually handled. The first response involves a mediating emotional response of anger giving rise to responses anticipated to arouse pain in the object. The second response also involves anger as a mediating response, but instead of this leading to pain inflicting behavior, there is ". . . modification of the initial eliciting stimulus condition either through removal of the stimulus or through a change in the meaning of the stimulus.” (Feshbach, 196k, pp. 266-267). Examples of removal of the stimulus are participating in social movements for the eradication of injustices, or fund raising; 19 for private charities to help eliminate disease and pov- i erty. Examples of changing the meaning of the stimulus are seeing that a slight was unintentional, or the act of an irresponsible agent. The third class of response con sists of positive feelings, such as love and compassion, incompatible with angrp feelings, which are associated with verbal evaluative mediating responses (e.g., "Love thine enemy," "To err is human, to forgive divine"). This leads to behavioral responses of a distinctly nonaggressive, pacifying character (e.g., "Turning the other cheek"). A Behavioral Theory of Sex Differences Tn Aggression Berkowitz (1962) notes the existence of a relation ship between aggressiveness and sex: Men are more aggres sive than women. Studies of nursery school children (Jersild and Markey, 1935), juveniles (Sears, 1961), and teenage youths (Lansky, Crandall, Kagan, and Baker, 1961) are cited which report either sex differences in amount of overt aggressive behavior or sex differences in degree of aggressiveness claimed in self-ratings. It is noted that sex differences in observer rated aggression occurs both in natural settings, such as nursery school, and in more arti ficial, gamelike settings such as doll play and doll inter views (Gordon and Cohn, 1961; Levin and Sears, 1956; Sears, 1951; Yarrow, 1948). Berkowitz (1962) postulates specific relationships 20 between eex role end aggressive behavior, with justifying arguments. These postulates are listed below. 1. Any given obstruction to goal-directed activity may actually mean more of a deprivation to one sex than to the other (Berkowitz, 1962, p. 269). McClelland, Atkinson, Clark, and Lowell (1953) are cited in their finding that men have more need for achieve ment than women. It is hypothesized that derogation of leadership or intellectual capacities will be more frus trating, hence more aggression provoking, for men than for women. Similarly, findings of Lansky et al.. (1961) are 4 mentioned that indicate a greater need for affiliation and social success in women than in men. It is hypothesized that attacks on popularity will be more frustrating for women with consequent greater instigation to aggressive behavior. 2. Men probably have stronger aggressive habits than do women and thus generally make stronger hostile responses to the evoking cue. (Holding anger arousal state constant) (Berkowitz, 1962, p. 269). | Berkowitz (1962) attributes this stronger aggres siveness habit to greater parental reinforcement of aggres- | i sion by boys, citing Sears, et al. (1957) who found that boys were encouraged to fight back more than girls, when 1 ! they are the target of peer aggression. He also argues by analogy from animal studies (Scott, 1958) demonstrating that mala hormones increase aggressive behavior, although he gives greatest weight to social factors in human aggres sion. 3. Women often have stronger inhibitions against direct aggression than do men (Berkowitz, 1962, p. 269). j i This is Berkowitzr (1962) explanation of his i j finding (Berkowitz, Green, and Macauley, 1962) that women show a greater decrement in aggressive behavior following highly instigated prior aggression. It is asserted that this greater guilt in women is due to greater concern on their part with social propriety. This leads to the corollary: (a) The average woman has a^ fairly strong wish to conform to the cultural definition of her sex role and in our society this means she is to avoid behaving aggressively (Berkowitz, 1962, p. 271). Evidence of this greater degree of conformity is derived by Berkowitz (1962) from studies by Janis and Field (1959) and Sears et al. (1957). A second corollary is: (b) Aggression-inhibiting women should be more feminine than their more aggressive sisters ^ (Berkowitz, 1962, p. 272). Berkowitz (1962) cites findings of Lansky et al. (1961) in studying attitudes, values and personality traits of adolescents to suggest that aggression in girls is related to suisculinity of interests, behaviors and needs. Girls who were highly critical of their mothers tended to be higher on scales of such masculine traits as need for achievement, and lower on feminine traits such as need for affiliation. The negative evidence of greater aggression in girls than in boys (Johnson, cited in Sears et al.. 1957) is explained by Berkowitz by reference to Sears' concept of "Prosocial aggression" (Sears, 1961). This view is that propriety is of more value to women than men. Women will therefore show more aggression in its defense, following Postulate 1. Prosocial aggression is seen as arising from, * * . . . strong internal restraints against direct aggression ..." (Berkowitz, 1962), and evidence in support of this position is cited (Sears, 1961). Studies of Sex Differences Terman and Miles (1936), in reporting on the devel opment of an exhaustive test of sex differences, refer to popular notions of differences in emotional style. Perhaps no other differences . . . have been so much stressed as those in the field of emotions .... Women are characterized as having a higher degree of affectibility than men .... They respond to a greater variety of stimuli and experience su»re intense feelings .... It is commonly believed that there are . . . sex differences in the relative contribution of the specific emotions to the tokkl affective life; 23 man, for example, is supposed to be more given to anger and woman to the sympathetic emotions . . . sex differences are alleged with respect to the stimuli that are effective in calling forth a given emotion; men and women do not fear the same things, are not angered by the same things, do not pity the same people or withdraw in disgust from the same objects (Terman and Miles, 1936, p. 35). Descriptions of emotion-provoking situations that it was believed would discriminate between sexes were tested on samples of eighth grade, high school, and college students, and 34 anger items selected by item analysis as eliciting differential responses. The items that charac terize anger responses of women are described as evoking sympathy, while men's reactions of anger are characteristi cally to business items. An interesting observation is that 8 of the 10 most feminine items refer to witnessing a social injustice, (e.g., seeing boys make fun of old people) and the other two items refer to attacks on personal repu tation (being called lazy, and being socially slighted). Other emotions studied by Terman and Miles (1936) were pity and disgust. A qualitative analysis such as that just applied to the anger scale indicates that the pity scale is composed of items on the same dimension of sympathy or social concern. Men score lower and could be characterized as more callous or tough, or as caring less about other people’s troubles. Similarly the disgust scale is composed of items relating to social propriety, with women being disgusted more than men at things that offend 24 the social sensibilities, (e.g., a drunken man, words like "belly" or "guts"). Of the two remaining scales, "Fear” and "Ethical Values," we may omit consideration of the former as not bearing on this study. The latter consists of a scale of infractions of moral principles of varying severity. Women tend to be much harsher than men in punishing small infractions, but are less severe than men in responding to the more serious items. Sex-related differences in interests interact with emotional patterns and values as in the male preference for adventure, travel, science, tools, and physical activity, all of which express a manipulative, instrumental, control ling approach to the environment, and a self-oriented set of values. Women's interests tend to consist of passive appreciation for beautiful things, developstent of social relations through affiliative, dependent, ingratiating behaviors, and preserving self esteem by maintaining custody of social values. Instrumental action is limited mainly to homemaking and caretaking in the widest sense, and, most pertinent to the present study, inculcating and enforcing social mores. Terman and Tyler (1954) note that these sex differences in patterns of interest and degrees of aggres siveness are established at about the same age (3 to 5 years). i The findings of Terman and Tyler (1954) in refer ence to development of aggression, social attitudes and 25 interests may be summarized as follows: Boys are, at all ages, more aggressive than girls. Boys are more likely to quarrel over possessions, while girls are more inclined to quarrel over inter ference with organized activity (this applies to nur sery school children). The antisocial nature of ag gression in boys is evident at an early age. Girls fantasy more social action than they engage in (introvertive tendencies and inferiority feelings inhibit social participation, in the opinion of the authors). The only area where girls manifest more anger than men is that of attack on affiliation status. Jealousy is more prevalent in girls (degree of jealousy can be conceived as a function of value placed on social relationships). Girls are more coop erative and earn better scores on tests of character. Studies of reputation indicate that girls are given more credit for moral superiority than they actually deserve, however, they do exceed boys in behaviors such as self control, persistence, cooperativeness, moral knowledge, and have more moral opinions. Before turning to areas of cultural factors and details of studies relating to this one, it should be noted that much of the literature discussed and evaluated by Terman and Tyler (195U) is also reviewed in another, more recent source (Mussen, Conger, and Kagan, 1963) with essentially similar findings. Mussen et al. (1963) place somewhat more emphasis on imitative learning of sex roles as an antecedent of sex differences than do Terman and Tyler (1954). Terman and Tyler (1954) are strongly in fluenced in their hypothesis that aggression is sex typed by the study of Sears, Pintler, and Sears (1946), who found that boys without fathers available to serve as imitative models still developed role appropriate aggressiveness with only minor developmental delay. There was evidence that the mothers of these father-absent boys taught their sons to be aggressive by other means. Cultural Factors in Sex Differences Rosenblith and Allinsmith (1962) note three factors that interact in the determination of sex role differences. 1. Certain behaviors may have been reinforced more in one sex than in the other. 2. Interpersonal experiences may be different according to one's sex. Or, 3. Certain groups have as members persons of only one sex, and these groups may then mold behavior (Rosenblith and Allinsmith, 1962,p.25D The effect of the last factor mentioned, grouping along sex lines, is noticeable in a study of 110 cultures made from the Area Files (Barry, Bacon, and Child, 1957). The most general finding was that degree of sex-role dif ferentiation seemed to bo greater as a function of tho need for physical strength in the economy of the society, and tha degree of social isolation of the nuclear family. Isolation imposes a practical necessity of cross-sex role playing (e.g., if the woman becomes ill, the husband must do household duties and tend the children). In less isolating cultures there are same-sexed substitutes to fill most roles in the event of an actor's disability.. Similarly, need for physical strength implies differentia tion of roles, as in a hunting economy where men form groups to drive game. As men band together to multiply their strength, and as women do likewise, there is much more opportunity for sex differences to become institu tionalized in these fraternal and sororal groupings. Seward (I9h6), in surveying the historical, anthro pological, and sociological data on differences in sex roles, both in modern western civilized cultures, preliter ate cultures and past civilizations presents comparisons designed to show a positive relationship between differ ential status of the sexes and personality traits such as aggressiveness, dominance and submissive, affiliative, dependent behavior. Sex status seems to be closely related ! to economic productivity, or control of wealth. The example is used of a tribe in New Guinea (the Tchambulis) in which women became the mainstay of the economy. The sen have since manifested many characteristics highly associated 28 with femininity in our culture( such as greater emotionality; sensitivity to slights, and preoccupation with artistic interests. Another example of preliterate cultural social differences of sex roles relating to the hypothesis of the present study is the twofold differentiation among Piegan Blackfoot indian women. The dimension of goodness or i virtue is separated from the dimension of effectiveness or competence. Four extremes of adjustment at the poles of these dimensions are: 1. "Manly-hearted” women, characterized by highly successful, instrumental behavior, which is rewarded by social sanction of masculine hx- jpressive behavior (e.g., taking initiative in love-making). 2. "Feminine" woman, characterized by extreme docility, compliance, and good nature, and labelled "The favorite wife." 3. "Crazy" women, who are not effective, like "Manly-hearted" women, and have socially deplorable attitudes. 4. "Sun Dance" women, who are models of virtue, and thereby gain status and are given the task | of performing the sacred dances. This is a i position akin to that held by the Vestal virgins; in the Roman civilization. I 29 Sex roles in contemporary American culture, accord ing to the hypothesis of the present study, would consist of fusions of these two dimensions. Ddcility and virtue j are combined in an amalgam of ambiguous reactions in women, who are placed in a passive role in the economic and politi cal areas of the society, but control or determine family behavior by means of moral codes, the nature and application of which is a female monopoly. Men, by contrast, are cast in a role combining the quality of competence with the less socially desirable quality of antisocial aggressiveness. Men are tough, self-seeking, callous, etc., all undesirable attributes in terms of avowed virtues of middle-class Western society. Seward (1964), in a recent reassessment of the literature pertaining to sex identity and socio-cultural factors, stresses the wide range of social class variability; in values, norms, attitudes, and experiences entering into sex role learning in our society. The confusions of role identity predicted earlier (Seward, 1946), are seen as con firmed. Preverbal establishment of basic identity in the i j sphere of sexuality is seen as providing an underpinning upon which a later variety of soeiomodes" of sex role behavior patterns may be superimposed, satisfying the changing social pressures of modern western middle-class culture. A sociological theory that also stresses the sex typing hypothesis of acquisition of differences to a greater extent than the sex role imitation hypothesis is that of Parsons (1954a, 1954b, 1954c). Utilizing a psycho analytic personality model, he traces the vicissitudes of sex differentiated aggressive behavior patterns through a series of steps. Predicating instigation to aggression to be a function of insecurity, he examines the mother-child relationship in preschool family life for sources of in security and deliberate sex role indoctrination. Depriva tion of affection and failure to conform to behavioral standards are seen as important antecedents of insecurity. The preschool boy is subjected to both antecedents by his mother, while inculcating in her son the normal sex role stereotypes. He is subtly reinforced for bad behavior, especially in fighting male peers, but aggressive behavior is labelled as unequivocally bad in other contexts, such as aggression against the parents. If aggression training is not carried out in the home, the lack is made up during school years by the peer culture and extrafamilial authori ties such as teachers. Positive and negative social re inforcers shape the appropriate aggressive behavior: Interest in instrumentally aggressive acts of an asocial or social nature, such as science, adventure, patriotic war, etc., are positively reinforced, while lack of aggressive ness on moral grounds incur negative peer pressure (e.g., 31 taunts of being "momma's boy," etc.). Supporting the peer culture derogation of virtuous behavior is the prejudice of social authorities, as seen in studies of teacher's atti tudes and classroom interactions (Goodenough, 1931; Meyer and Thompson, 1956; Wickman, 1938). Lending support to the hypothesis that insecurity leads to aggressive behavior are the findings of Whiting and Child (1953), that gentle, security-breeding child rearing atmospheres are associated with gentleness as a character trait, and that this trait extends across age and sex differences (e.g., Hopi vs. Comanche, or Mundugamoor vs. Arapesh). The writer of the present study feels that more important for development of male aggression patterns than feelings of insecurity, are the positive reinforcements of the male role. That the male role is clearly the more desirable one in our society can be deduced from a variety of evidence. Studies of sex role preference have consist ently indicated a cross-sex preference for the male role (Terman and Miles, 1936; Brown, 1958). Analysis of sex roles in children's readers also indicates that the male role is superior: Female characters tend to be neglected or subordinated, and gain little in the way of rewards, while male characters typically are central, powerful, achieve desirable goals by their own efforts, and are more frequently rewarded by others (Child, Potter, and Levine, 1946). Treatment accorded preschool girls also involves a 32 good deal of consciously diractad rainforcamant, only tha amphasis is on "goodness" and passivity. Varbal instrumen talities are cultivated in lieu of physical aggression, and imitation learning is more important, due to the presence of tha mother. Parsons (195Ua) feels this leads to earlier psychological maturing in girls. Tha transition to school ; age, juvenile pear culture is smoother for girls, with the same social reinforcements of approval from authorities following good behavior as occurred at home. However, difficulties arise in adolescence, when it becomes apparent that 'being good" is not sufficient, or necessarily even desirable, in gaining security as an adult. This is be cause security for a woman is found in marriage, which involves securing masculine preferment, with the penalty for failure being condemnation to the role of "Old Maid." No equally attractive alternative to marriage is provided, mainly due to covert discrimination against equality of career opportunities. The equivalent of "Manly-hearted" women in our society encounter the same obstacles to a career as a would-be negro professional student (e.g., Seward, 19h6). Studies of Acquisition of Aggression Patterns Sears et al., (1957) interviewed a fairly represent-* ative samplesof mothers of 5 year olds about their child rearing practices and experiences. In reference to aggres sion, Sears et al. (1957) £ound that the usual pattern. o£ parental attitudes was discouragement o£ aggression directed at parents and other author ities, somewhat greater permissiveness o£ aggression between siblings, with great est freedom of expression of aggression permitted in peer situations. Reported degree of aggressiveness was found to be related to amount of permissiveness of parents, es pecially toward authority figures, amount of physical punishment used, and lack of affection (as inferred from mothersT reports, by the investigators). Mothers claimed they punished aggressiveness in the same way for both boys and girls, when they punished at all. They did admit to greater encouragement of peer aggression by sons. Keeping in mind the demonstrated poor validity of mothers* recall of their own child rearing acts (Goodenough, 1931? Mussen, Conger, and Kagan, 1963), it seems possible to interpret these findings as supporting the view that male aggressive ness is largely due to social pressures toward conformity to the sex stereotype. In a follow-up study of some of these same children at age 12 (Sears, 1961), most of the earlier correlations between punishment, permissiveness, and aggression measures had fallen away to insignificance. Using objective self ratings of a number of aggression parameters, (pro- and anti-social aggression, self aggression, aggression anxiety, and projected aggression) sex differences were found in 34 antisocial and prosocial aggression. Boys scores higher than girls on the antisocial aggression scale, while girls scored higher than boys on the prosocial aggression scale. After a lengthy theoretical analysis of his results, Sears (1961) concludes that sex typing is the chief determi nant of the difference. The difficulty of attempting to relate sex differences in aggression to child rearing practices may partly be the result of changing pressures from the larger social milieu on the relatively invariant parental attitudes. Seward (1954, 1956) describes how such shifting about of sex role appropriateness with changing age can necessitate adjustments of identification through narrowing or widening the range of models. Per haps Sears' (1961) differences between reactions of 5-year old boys to punishment (increased aggressiveness) and 12-year old boys' reactions (decreased aggressiveness) are explainable on the basis of a broadened range of models to imitate. When the 12-year old is punished, the appropriate % reaction may be Spartan inhibition of all visible affect, rather than visible angry aggression such as the 5-year old might show. In delineating a syndrome of authoritarianism, Adorno et al. (1950) found that one of the important com ponents was authoritarian aggression, which is equivalent to the concept of prosocial aggression set forth by Sears (1961). They theorized (Adorno et al.. 1950) that, 35 The individual who has been forced to give up basic pleasures and to live under a system of rigid restraints . . . is likely , . . to be particularly annoyed at the idea that another person is "getting away with something." (Adorno et al.. 1950, p. 232) And, It is to be expected . . . that the conventionalist who cannot bring himself to utter any real criticism of accepted authority will have a desire to condemn, reject, and punish those who violate these values. (Adorno et al.. 1950, p. 232) Studies of Imitative Learning and Identification Although greatest emphasis so far has been placed on more or less deliberate manipulation of reinforcements by authority figures as the major antecedent of sex dif ferences in patterns of aggressive behavior, there is no intent to deny the importance of imitative learning as an associated causal factor. Consideration of three theories of imitative learning follows. Studies of doll play aggression indicate a reliable relationship between presence of a father figure in the home and sex role appropriate degrees of aggressiveness in boys. Boys with fathers show average sex differentiation in this respect by age three. Boys from father-absent homes make up the difference in development of aggression by age five (Bach, 1946; Sears et al.. 1946; Sears, 1951). This catching up in aggressiveness on the part of boys with no father is probably related to extrafamilial social pressures and imitation of extrafamilial models, such as peers and other children's fathers. Different theories of the way in which this nodeled behavior is mediated have been proposed. Three views are the "Identification with the Aggressor," or "Status Envy" theory (Freud, 1937); the "Social Power" theory, occurring in two forms (Sears et al.. 1957; Brim, 1958) , one stressing emulation of a warm, affectionate father, the other, a father with positive and negative reinforcing potential; and the third, "Secondary Rein- forcement" theory in which pairing of positive reinforce ments with the identity of the reinforcing figure leads to greater identification with that figure (Mowrer, 1950). These theories can be thought of as asking who is imitated, the person receiving the rewards (Status Envy theory), the person owning the rewards (Social Power theory), or the person giving the rewards (Secondary Reinforcement theory)? The evidence seems to favor the Social Power theory (Bandura, Ross, and Ross, 1963b; Brim, 1958; Mussen and Distler, 1959; Rosenblith, 1961). Mussen and Distler (1959) selected boys of high and low sex role identification, as determined by adminis tration of a test of sex role preference (Brown, 1956). They then completed a set of family situation stories pre sented through the medium of doll play. The stories were to elicit nurturant and punishment themes in the projected behavior of parent figures. Thus, a measure of subjectively 37 perceived perentel qualities was obtained. "High masculine* identity subjects perceived the father as significantly more nurturant and punitive than did "low masculine" iden tifiers. The authors concluded that the more "salient" the father (combined positive and negative reinforcing power), the more the son will identify with him. Brim (1958) compared teachers' ratings of children from two-sibling families on 31 traits associated with sex role by a set of judges. He found that in pairs of sib lings of opposite sex, the younger of the pair would adopt more of the opposite sex characteristics than the older sibling would adopt. He offers this finding as support of the "Social Power" theory of identification. Rosenblith (1961) studied the effect of variations in amount of nurturant behavior displayed by models on the subsequent imitative behavior of child subjects. Male models elicited more imitative behavior than female models. This finding may be explained by the "Social Power" theory, and sociological evidence of the greater status of men than women in our society (e.g., Seward, 1946). Bandura et al.. (1963) experimentally duplicated a two parent, one child triad, and varied the sex and family position of the rewarding, rewarded, and ignored roles, in a situation involving the dispensation of highly attractive | toys and refreshments. They found that the figure control- j ling dispensation of rewards was imitated more in a sub- sequent learning situation, than either the non-consuming, ignored adult, or the rival adult consumer. This finding militates against the "Status Envy" and "Secondary Rein forcement" theories and lends support to the "Social Power" theory. In addition, there was evidence that male figures were invested with controlling power in excess of that provided by the design. Non-consuming, ignored male adults were believed in a number of cases to be the real owner of the desirable materials. This belief was verbalised by some children as having been taught them by their mothers. Studies of Imitation of Aggressive Behavior Bandura et al., (1961, 1963a, 1963c), in a series of studies utilizing live and film-mediated aggressive models, have studied the effects on nursery school children of anti-social aggression, rewarded and punished. When the violence is directed at toy surrogate human figures (Bobo dolls) , the results duplicate the findings of doll play aggression studies: Incidence of aggression increases, with boys displaying more aggression than girls, as time passes. The results are unchanged when the modeled be havior is depicted on film. When consequences are added to j t the design, by having instrumental antisocial aggression go ; unpunished in one condition and punished in the other, both j boys and girls imitate the successful aggressor more than the unsuccessful one, although overall sex differences in 39 incidence of aggression (imitative and nonimitative com bined) persist. Girls and boys verbally deplore the anti social character of the victorious aggressor, and give rationalizations to justify the outcome, all of which are in terras of the vanquished defender being morally wrong for being ineffective. CHAPTER III HYPOTHESES The hypotheses fall into three groups: Those dealing with attitude toward aggressive behavior, those dealing with patterns of aggressive behavior, and those dealing with total amount of aggressive behavior. Com parability of age and education of subjects are assumed in all cases. Group 1 Hypothesis One: Men will be more accepting of antisocial aggressive attitudes than will women. Hypothesis Two: Subjects with less religious emphasis will be more accepting of antisocial aggres sive attitudes than will subjects with more religious emphasis. Hypothesis Three: Sex differences in amount of acceptance of antisocial aggressive attitudes will be smaller between subjects with more religious emphasis than between subjects with less religious emphasis. 40 Group 2 Hypothesis Pour: Women will manifest more pro- soeial aggressive behavior than men. Hypothesis Five: Subjects with more religious emphasis will manifest more prosdcial aggression than will subjects with less religious emphasis. Hypothesis Six: Subjects exposed to greater instigation to aggression will manifest more pro social aggression than will subjects exposed to lesser instigation to aggression. Hypothesis Seven: Sex differences in amount of prosocial aggression will be smaller between subjects with more religious emphasis than between subjects with less religious emphasis. Hypothesis Eight: Sex differences in amount of prosocial aggression will be greater between subjects exposed to greater instigation to aggression than between subjects exposed to lesser instigation to aggression. Hypothesis Nine: Men will oumifest more antisocial aggression than women. Hypothesis Ten: Subjects with less religious emphasis will manifest more antisocial aggression than will subjects with more religious emphasis. Hypothesis Eleven: Subjects exposed to greater instigation to aggression will manifest more antisocial k2 aggression than will subjects exposed to lesser instigation to aggression. Hypothesis Twelve: Sex differences in amount of antisocial aggression will be greater between subjects exposed to greater instigation to aggression than between subjects exposed to lesser instigation to aggression. Hypothesis Thirteen: Sex differences in amount of antisocial aggression will be smaller between subjects with more religious emphasis than between subjects with less religious emphasis. Group 3 Hypothesis Fourteen: Men will manifest more total aggression than will women. Hypothesis Fifteen: Subjects with less religious emphasis will manifest more total aggression than will subjects with more religious emphasis. Hypothesis Sixteen: Subjects exposed to greater instigation to aggression will manifest more total aggression than will subjects exposed to lesser instigation to aggression. Hypothesis Seventeen: Sex differences in amount of total aggression will be greater between subjects exposed to greater instigation to aggression than between subjects exposed to lesser instigation to aggression. Hypothesis Eighteen; Sex differences in amount of total aggression will be smaller between subjects with more religious emphasis than between subjects with less religious emphasis. Hypothesis Nineteen: Amount of acceptance of anti social aggressive attitudes will be directly related to amount of antisocial and total aggression, and inversely related to amount of prosocial aggression. CHAPTER IV METHOD Subjects Two samples of subjects were used to test the variable of religious emphasis. One sample of 70 men and 53 women consisted of all students present on the day of the experiment in an undergraduate Psychology class at the University of Southern California (USC), and is the sample with less religious emphasis. The other sample of 4-8 men and 66 women consisted of all the students present on the day of the experiment in an undergraduate Psychology class at Westmont College, and is the sample with more religious emphasis. "The studied and deliberate position of Westmont College is that all areas of true knowledge are ultimately compatible with each other and with Christian revelation." (Westmont College Catalogue, 196k, p. k). The objectives of Westmont College are "A Christ-honoring scholarship * . . i A Christ-pervaded atmosphere . . . A Christ-centered ; i curriculum . . . (and) a Christ-serving graduate." (West mont College Catalogue, 196k, p. 5). Further evidence of 45 the religious emphasis of Westmont College is the admission requirement of "a personal recommendation by the applicant's! i pastor (on form provided)." (Westmont College Catalogue, 1964, p. 25). The closest equivalent by USC to the above state ments of religious intent is the statement that "the program . . . is designed to give the student a better understanding . . . of the meaning and significance of i values and the criteria of evaluation; of moral, religious, and aeathetic experiences and the role they play in the life of an individual and a culture." (Bulletin of the University of Southern California, College of Letters, Arts, and Sciences, 1964, p. 10). Evidence that both schools dram their students from j the same social class is the fact that both charge equiva lent tuition fees. USC charges $1,200 per school year, blanket rate, while Westmont College charges $1,100 per school year, also blanket rate. (Bulletin of the University of Southern California, College of Letters, Arts, and Sciences, 1964; Westmont College Catalogue, 1964). ! Means and standard deviations of age and educational! i level are ahewn in Table 1. j Students in both samples were assigned to experi mental and control conditions by systesiatically distributing) materials booklets first to women and then to men, the i booklets having been previously arranged in alternating 46 TABLE X MEADS AND STANDARD DEVIATIONS OF AGE AND EDUCATION OF SUBJECT GROUPS (MORE RELIGIOUS-LESS RELIGIOUS , MALE-FEMALE) Age (years) Years College Group Mean Standard Deviation Mean Standard Deviation More Religious-men (n=48) 20.31 .99 3.01 .91 Less Religious-men (n=70) 20.54 1.02 3.06 .81 More Religious-women (n=66) 20.02 1.00 2.96 .79 Less Religious-women (n=53) 20.10 .95 2.99 .85 h7 order. This resulted in the following group compositions: More religious, high anger (experimental) males, N=25; less religious, low anger (control) males, N=23; more religious, high anger females, N=3k; more religious, low anger females, N=32; less religious, high anger males, N=36; less religious, low anger males, N=34; less religious, high anger females, N=27; and less religious, low anger females, N=26. Materials and Administration Two sets of ten (10) statements, of the kind found on personality tests such as the MMP1, but answered on three or four point rating scales were administered. These sets of items differed in anger arousing potential. In structions were: "These are some items being tried out in the development of a new personality test. Please complete the items and wait for further instructions.1 1 The purpose of giving these items was to produce differential instiga tion to total aggression, and feelings of anger, in the two groups. The high anger group received the more aggression instigating and angering set of items, while the low anger group received the less instigating, less angering items. Common practice in studies of aggression using college students is to anger one class or group by in sulting them in some way, such as derogating their ability on some purportedly simple but actually difficult task, as we have seen in reviewing the literature, above. A control group receivee identical treatment, save for the substi tution of neutral comments for insults. Such a procedure requires two separate groups, introducing sizeable inter group variability. To reduce this variability, two written forms of comparable items were developed, the reading of which forms would induce differential degrees of anger and instigation to aggression. This allowed simultaneous administration of both conditions to one group of subjects. Ten items were constructed in such a manner that their completion would be incidentally anger arousing. An additional ten items were selected from the Minnesota Multiphasic Personality Inventory in the following manner: Item one, and every tenth item, up to and including item 90,— was taken. The list of twenty items, ten items of presuma bly high anger arousal potential and ten items of presuma bly no more anger arousal potential than personality test items in general, were presented to a class of undergraduate students in a Psychology course at Los Angeles State Colley, consisting of 19 men and 19 women. Instructions were: "Please read the following personality test items with the idea of seeing how mad or angry they make you feel. Ex press the degree of anger on an imaginary scale of 0 to 100. with 0 equalling no anger at all. and 100 equalling the most anger possible to experience. Write the number 49 just to the left of the item, in the space provided." Means and standard deviations of the ratings for each item were computed and the variances tested for homogeneity, using Bartlett's test. Variances were comparable between sexes. Items had been arranged in odd-even order, with anger arousing items even, MMP1 items odd. See Appendix A for the list of twenty items, with mean ratings of both sexes. In addition to having subjects rate the two lists of items for anger arousal potential, the lists were ad ministered as personality tests to another, independent sample of comparable students from the same College. Nine teen men and 11 women were divided into experimental (10 men, 6 women) and control (9 men, 5 women) groups. They were given the appropriate set of items with the instruc tions, "Please complete the following.V This was im mediately followed by administration of a criterion measure of degree of anger arousal, the Emotional Situations Test, described below. Both the male and female experimental groups, which had been administered the high instigating items, had higher mean scores on the criterion measure, the Emotional Situations Test, than did the male and female control groups, which had been administered the low insti gating items. The difference between means of the female groups was significant (p less than .05, two tailed test), while the mean difference of the male groups approached 50 significance Cp less than .10, more than .05). Since the largest mean anger arousal potential possible for any instigating item is 100, and since there are 10 items in each list, with the items presumably having an additive effect, dividing the sum of the mean ratings for the ten items in each list would yield the percentage of the possible anger arousal potential that the list as a whole had for the 38 raters. Table 2 presents these percentages of possible anger arousal for the two lists of items (high anger arousal items and low anger arousal items) In order to increase the difference in anger arousal potential between high anger arousing lists and low anger arousing lists, the items on both lists were rearranged. Some items were transferred from one list to the other, two items were discarded, and two new items were added. The added items were assigned mean ratings equal to the mean of the mean ratings of each sex, of the other eight items in each list. (See Appendix B for the revised lists of high and low anger arousing iter's). Table 3 presents the percentages of possible anger arousal for the two lists of items, as revised to increase the difference in anger arousal potential between them. Ratings made by men do not differ significantly from ratings made by women for either high or low anger arousing lists of items. Ratings of the high anger arousal list are significantly greater than 51 TABLE 2 PERCENT OF POSSIBLE ANGER AROUSAL POTENTIAL FOR HIGH AND LOW ANGER AROUSING LISTS OF ITEMS, RATED FOR AMOUNT OF ANGER AROUSAL POTENTIAL Sex o£ Rater Male (N=19) Female (N=19) High anger items 36.64% 36.12% Low anger items 29.51% 30.75% TABLE 3 PERCENT OF POSSIBLE ANGER AROUSAL POTENTIAL FOR HIGH AND LOW ANGER AROUSING LISTS OF ITEMS, REVISED, RATED FOR AMOUNT OF ANGER AROUSAL POTENTIAL Sex of Rater Male (N*19) Female (N-19) High anger items 40.26% 41.60% Low Anger items 22.40% 18.25% 52 ratings of the low anger arousal list, for both men and z women raters (X = 5.09, p less than .05, men's ratings; X = 9.10, p less than .01, women's ratings). The revised i lists of high anger arousing and low anger arousing items were those used in the present study to produce conditions of high and low instigation to aggression. To test hypotheses about patterns of aggression, it was necessary to measure prosocial and antisocial aggression. Buss (1961) reviews the literature pertinent to measurement of aggression, and cites a number of rele vant studies making use of thematic instruments (e.g., Beliak, 19*44; Mussen and Naylor, 1954; Purcell, 1956). Beliak (1944) was able to measure changes in aggressiveness in college students under high and low instigation (criti- | cism of performance vs. no criticism) by the simple expedi-: ent of counting the number of hostile, aggressive words used by the subjects in responding. It was decided to use j i this word-count method for assessing total aggression level in this study. Mussen and Naylor (1954) and Purcell (1956) found inverse relationships between antisocial aggression themes and both fear and guilt themes. Feshbach, Singer, and Feshbach (1963), who developed a set of 18 thematic slides i suitable for investigating aggressive behavior, were in terested in the relative effectiveness of age similarities j in changes of attribution of aggression induced by insulting 53 experimental subjects. A set of these slides was obtained for use in the present study. Buss (1961), reviewing the use of projective devices, concluded that thematic materials are very suit able for use with college students. Such materials have been demonstrated to be sensitive to transient induced changes in emotional reaction tendencies. After completing the high anger or low anger instigating items, all subjects were shown the set of 18 slides depicting pairs of people interacting in different situations. Subjects were instructed to "write down in £ word or two, how the persons in the slides are feeling." The above procedure was designed to detect and measure the extent of prosocial and antisocial angry feelings aroused by the foregoing instigating items. The slides consist of photographs of live models of college-age men and women, and juvenile feoys, interacting with im portant figures, such as mothers, teachers, fathers, etc. Each figure is labelled "AM or "B", and there are only two figures on each slide. (See Appendix C for detailed des criptions of the slides.) Slide responses were recorded on a form in the materials booklet having 18 numbered, blank spaces. Subjects were told to label each response with the appropriate letter, A or B, to identify the figure to which the response was attributed. Anger responses attributed to A figures were considered to indicate 5 4 - pro social aggression, since the A figures are authority figures and are initiating the action, in each case. Anger responses attributed to B figures were considered to indi cate antisocial aggression, since such responses involved resistance to, or action against authority. Administration of slides was followed by adminis tration of the "Emotional Situations Test" (Emo. sit.), which was designed to measure differences in degree of total anger arousal. The test consists of 20 printed items on a sheet in the materials booklet, with space before each item for recording the response. (See Appendix D for a copy of this form.) Each item describes the subject in some anger provoking situation. The subject is instructed to rate the items on a 0 to 100 scale of increasing anger arousal. Test items were developed in an earlier study as described below (Thompson, Jacobs, and Meehan, 1962). Over 300 male and female undergraduate college students from an undergraduate course in Psychology were asked to "list five situations that would arouse a given feeling." Sixteen feelings were listed, including anger. The situations were also rated by each subject as to in tensity of feeling evoked, on a scale of 0 to 100 of feeling intensity. The 20 situations making up the Emotion al Situations Test used in this study are a compilation of the top 20 discriminable situations, given with some degree of reliability (recurring at least four times for each aex in the sample), and rated with appreciable mean intensity (mean ratings of 25 or more). The reliability of the com piler's judgments in categorizing the responses was checked by having an independent rater categorize a random sample of protocols. An agreement of 90% was obtained between raters. The twenty items were selected as much as possible from categories occurring in samples of both sexes, in order to eliminate sex bias in response. Elimination of this potential bias was checked by selecting a random sample of protocols from each of the low anger-high anger, male-female conditions of the low religious emphasis group, and testing the significance of mean differences between sexes for each item. These means and mean differences are shown in Table None of the differences were significant. Finally, administration of the Emotional Situations Test was followed by administration of a set of 16 MHPI items, constituting a scale of items found by previous in vestigators to be factorially related, and correlated with sex of subject (Mees, 1959; Welsh, 1956). This scale, the Antisocial Acceptance Scale, has been found to measure differences in attitude toward antisocial aggressive behavior, as defined in this study (Mees, 1959). (See Appendix E for a copy of this test.) Response measures 1* Slid* responses. Responses for each figure, 56 TABLE 4 MEANS, MEAN DIFFERENCES, AND STANDARD ERRORS OF MEAN DIFFERENCES, RATINGS OF ANGER, EMOTIONAL SITUATION TEST ITEMS Item Mean Men(N=10) Women(N=10) Mean Difference Standard Error of Mean Difference I 71.20 52.50 18.70 10.10 2 61.70 56.50 5.20 11.70 3 61.50 63.00 1.50 12.40 4 55.00 44.50 10.50 13.00 5 72.00 85.00 13.00 13.70 6 52.50 65.00 12.50 8.90 7 61.00 56.50 4.50 12.20 8 75.50 69.50 6.00 13.50 9 34.50 65.50 9.00 13.00 10 70.50 72.00 1.50 10.30 11 62.00 71.50 9.50 7.80 12 50.50 62.00 11.50 9.30 13 56.50 77.00 20.50 11.10 14 51.50 45.00 6.50 12.00 15 61.00 54.50 6.50 12.80 16 70.50 57.50 13.00 10.10 17 48.50 58.50 10.00 8.80 18 63.00 71.00 8.00 10.00 19 59.00 52.00 7.00 10.10 20 76.50 89.00 12.50 6.20 57 A and B, on each alide were checked against a liat (described below) for occurrence of words that denoted angry feelings. (See Appendix F for the list of words used.) The list was adapted from a list developed in an earlier study of classifi cation of emotional adjectives (Jacobs, Capek, And Meehan, 1959), in which subjects selected adjectives denoting broad feeling categories, such as anger. The sun of anger responses attributed to A figures was used to measure prosocial aggression. The sum of anger responses attributed to B figures was used to measure antisocial aggression. The sum of all anger responses was used to measure total ag gression. 2. Emotional Situations Test ratings. The obtained ratings on the 20 items were summed and divided by 100, to yield a total Emotional Situation rating, with possible range of 0 to 20. 3. Antisocial Acceptance scale scores. The sum of keyed items on the Antisocial Acceptance scale was used as a measure of acceptance of antisocial aggressive attitudes. The range of possible scores was from 0 to 16. Design Five 2x2x2 complete factorial analyaes of variance were performed, one for each dependent variable meaaured: Proaocial, antiaocial and total alide aggres- aion, total Emotional Situatione teat aggreaaion, and Antiaocial Acceptance scale acorea (Edwards, 1950, p. 175; Lindquiat, 1953, p. 220). In order to equate the eight treatment conditiona, to allow application of the fac torial deaign, it waa neceaaary to discard protocols from some of the cells. This waa done randomly, and the eight cells were equated with N=23, in eachnease. This did not materially affect the means and variabilities computed from the unequated groups. CHAPTER V RESULTS Antiaocial Acceptance scale comparisons Mean scores and standard deviations of men and women, under high and low conditions of instigation to aggression, from groups of high and low religious emphasis on the Antisocial Acceptance scale, are shown in Table 5. TABLE 5 MEAN SCORES AND STANDARD DEVIATIONS FOR MEN AND WOMEN UNDER HIGH AND LOW CONDITIONS OF INSTIGATION TO AGGRESSION, AND MORE AND LESS RELIGIOUS EMPHASIS Group N Mean Standard Deviation More religious-high anger- men 23 5.61 2.06 More religious-low anger-men 23 6.26 2.63 Less religious-high anger- men 23 6.65 2.55 Less religious-low anger-men 23 6.00 2.83 More religious-high anger- women 23 ■F • O o 2.09 More religious-low anger- women 23 3.7 4 2.14 Less religious-high anger- women 23 5.00 2.13 Less religious-low anger- women 23 5.04 2.99 59 60 Results of analysis of variance of Antisocial Acceptance scale scores for these groups are given in Table 6. TABLE 6 ANALYSIS OF VARIANCE OF ANTISOCIAL ACCEPTANCE SCALE SCORES Source df Sum of Squares ms F A (Sex of subjects) 1 130.56 130.56 21.83* B (Religious emphasis) 1 27.39 27.39 4.58*' C (Instigation level) 1 .13 A x B 1 6.67 6.67 1.12 A x C B x C A x B x C 1 1 1 10.46 (sum of above three Within cells 177 1,058.52 sources not sig nificant if appor tioned to any one source) 5.98 Total * .. ___ 184 1,233.73 p less than .001. **p less than .05. The very significant main effect of sex, as seen in the fact that all men's mean scores exceed all women's, supports the prediction of Hypothesis 1, that men will be more accepting of antisocial aggressive attitudes than * 61 women. The significant main effect of high religious emphasis supports Hypothesis 2, which predicted the finding that high religious emphasis group means would be lower on this scale than low religious emphasis group means. The predicted interaction of these two main effects (Hypothesis 3) was not significant. Prosocial aggression score comparisons Mean scores and standard deviations of men and women, under high and low conditions of instigation to aggression, from groups of high and low conventional religious emphasis, on the measure of prosocial aggression (responses to A figures on slides) are shown in Table 7. Results of analysis of variance of prosocial aggression for these groups are given in Table 8. The predicted sex difference in prosocial aggression (Hypothesis * » • ) , was not confirmed. In addition, Hypothesis 5, predicting increased prosocial aggression in high religious emphasis subjects was contradicted by a signifi cant main effect in the opposite direction. Hypotheses 6, 7, and 8, pertainint to, respectively, main effect of instigation level, interaction between sex and religious emphasis, and interaction between sex and instigation level, were not confirmed. 62 TABLE 7 MEAN FROSOCIAL AGGRESSION SLIDE SCORES AND STANDARD DEVIATIONS FOR MEN AND WOMEN UNDER HIGH AND LOW CONDITIONS OF INSTIGATION TO AGGRESSION, AND MORE AND LESS RELIGIOUS EMPHASIS Group N Mean Standard Deviation More religious-high anger- men 23 2.61 2.44 More religious-low anger- men 23 2.61 1.75 Less religious-high anger- men 23 3.09 2.23 Less religious-low anger- men 23 4.44 3.09 More religious-high anger- women 23 3.87 1.89 More religious-low anger- women 23 3.09 2.19 Less religious-high anger- women 23 4.13 2.49 Less religious-low anger- women 23 3.83 3.36 63 TABLE 8 ANALYSIS OF VARIANCE OF FROSOCIAL AGGRESSION SCORES Source df Sum of Squares ms F A (Sex of subjects) 1 13.59 13.59 2.21 B (Religious emphasis) 1 31.39 31.39 5.11 (p<.05) C (Instigation level) 1 .20 A x B 1 4.89 A x C 1 17.04 17.04 2.78 B x C l} 11.75* A x B x C lj Within Cells 177 1,086.80 6.14 Total 184 1,165.66 ♦The sum of squares of both interactions (B x C and A x B x C) combined is not significant if totaLLy apportioned to either source. 64 Antiaocial aggression acorc comparisons Mean acorea and atandard deviationa of men and women, under highjmd low conditiona of inatigation to aggression, from groupa of high and low religioua emphasis, on the meaaure of antiaocial aggreaaion (responses to B V- figures on slides) are shown in Table 9. Results of analysis of variance of antiaocial aggression measures for these groups are given in Table 10. The predicted main effect of sex, with greater scores for men across religious emphasis and instigation level waa not confirmed (Hypothesis 9). The predicted main effect of religious emphasis, with subjects with less religious emphasis showing more antisocial aggression, was strongly confirmed. The predicted main effect of insti gation level (Hypothesis 11) was not confirmed. The pre dicted interaction between sex and degree of religious emphasis, with smaller differences between sexes hypothe sized for the high'' religious emphasis group, was not supported (Hypothesis 13). The predicted interaction between sex and level of instigation, with greater differ ence predicted for higher instigation level (Hypothesis 12), is supported, but requires qualification, since there was a reversal of rank order between the two sexes in the two conditions. Instead of antisocial aggression scores for men increasing in the high instigation condition, they decrease. while the scores of women increase. and exceed 65 TABLE 9 MEAN ANTISOCIAL AGGRESSION SLIDE SCORES AND STANDARD DEVIATIONS FOR MEN AND WOMEN UNDER HIGH AND LOW CONDITIONS OF INSTIGATION TO AGGRESSION, AND MORE AND LESS RELIGIOUS EMPHASIS Group N Mean Standard Deviation More religious-high anger- men 23 2.09 1.47 More religious-low anger- men 23 1.83 1.30 Less religious-high anger- men 23 3.22 2.33 Less religious-low anger- men 23 4.52 2.71 More religious-high anger- women 23 2.09 1.41 More religious-low anger- women 23 2.09 1.59 Less religious-high anger- women 23 4.96 2.53 Less religious-low anger- women 23 3.04 1.89 i r 66 TABLE 10 ANALYSIS OF VARIANCE OF ANTISOCIAL AGGRESSION SCORES Source d£ Sum of Squares ms F A (Sex of subjects) 1 .77 B (Religious emphasis) C (Instigation level) 1 1 168.34 2.17 168.34 43.39 (p<.001) A x B 1 .01 A x C B x C 1 1 25.14 .35 25.14 6.48 (p<.05) A x B x C Within Cells 1 177 34.77 686.36 34.77 3.88 8.96 (p<.01) Total 184 917.91 67 the scores of men in the high instigation condition. The difference between scores increases in the high instigation condition, producing a significant interaction term in the analysis of variance. Total aggression score comparisons Mean scores and standard deviations of men and women, under high and low conditions of instigation to aggression, from groups of high and low religious emphasis, on measures of total aggression are shown in Table 11 (total slide responses) and Table 12 (Emo. sit. test scores). Results of analysis of variance of these two measures of total aggression are given in Table 13 (total slide responses) and Table Ik (Emo. sit. test scores). The predicted main effect of religious emphasis, greater emphasis being associated with lesser aggression, (Hypothesis 15), is confirmed on both measures (total slides and Emo. sit.). Predicted main effects due to sex and instigation level (Hypotheses 1U and 16) and predicted interactions between sex and religious emphasis (Hypothesis 17), and sex and instigation level (Hypothesis 18), are not confirmed, with the exception that the interaction of sex and instigation level on the slides measure is significant. However, this is due to the same artifact of reversal of rank order in the low religious emphasis group, discussed 68 TABLE 11 MEAN TOTAL AGGRESSION SLIDE RESPONSES AND STANDARD DEVIATIONS FOR MEN AND WOMEN UNDER HIGH AND LOW CONDITIONS OF INSTIGATION TO AGGRESSION, AND MORE AND LESS RELIGIOUS EMPHASIS Group N Mean Standard Deviation More religious-high anger- men 23 4.70 3.27 More religious-low anger- men 23 4.43 2.46 Less religious-high anger- nen 23 6.30 3.89 Less religious-low anger- men 23 8.96 4.89 More religious-high anger- women 23 5.96 2.62 More religious-low anger- women 23 5.17 3.42 Less religious-high anger- women 23 9.09 3.78 Less religious-low anger- women 23 7.30 4.07 69 TABLE 12 ANALYSIS OF VARIANCE OF TOTAL AGGRESSION SCORES Source df Sum of Squares ms F A (Sex of subjects) 28.17 28.17 2.15 B (Religious emphasis) 1 373.06 373.06 28.52 (p<.001) C (Instigation level) 1 .08 A x B 1 2.18 A x C 1 70.6* 70.6* 5.*0 (p<.05) B x C 1 10.53 A x B x C 1 *3.99 *3.99 3.36 Within Cells 177 2,315.33 13.08 Total 18* 2,8*3.98 70 TABLE 13 MEAN EMOTIONAL SITUATIONS TEST SCORES AND STANDARD DEVIATIONS FOR MEN AND WOMEN UNDER HIGH AND LOW CONDITIONS OF INSTIGATION TO AGGRESSION, AND HIGH AND LOW RELIGIOUS EMPHASIS Group N Mean Standard Deviation More religious-high anger- men 23 11.38 3.13 More religious-low anger- men 23 10.19 2.24 Lesa religious-high anger- men 23 11.77 2.25 Leas religious-low anger- men 23 12.49 2.37 More religious-high anger- women 23 11.73 2.72 More religious-low anger- women 23 11.51 3.02 Lesa religious-high anger- women 23 11.99 2.55 Leas religious-low anger- women 23 12.92 3.50 71 TABLE 14 ANALYSIS OF VARIANCE OF EMOTIONAL SITUATIONS TEST SCORES Source df Sun of Squares ms F A (Sex of subjects) 1 15.44 15.44 2.05 B (Religious emphasis) C (Instigation level) 1 1 54.83 .16 54.83 7.26 (p<01) A x B 1 2.97 A x C 1 4.00 B x C 1 27.01 27.01 3.58 A x B x C 1 1.78 Within Cells 177 1,336.22 7.55 Total 184 1,442.41 1% above, and to be considered in the discussion. Relations between measures of attitude and behavior Correlations between acceptance of antisocial aggressive attitudes and measures of prosocial, antisocial, and total aggression are presented in Table 15. It is apparent that there is no consistent relationship between attitude toward antisocial acts and different forms of aggressive behavior. A relationship was predicted by Hypothesis 19. TABLE 15 CORRELATIONS (PEARSON'S r) BETWEEN ANTISOCIAL ACCEPTANCE SCALE SCORES, AND SLIDE-AGGRESSION SCORES AND EMO. SIT. SCORES, FOR MEN AND WOMEN UNDER HIGH AND LOW CONDITIONS OF INSTIGATION, AND MORE AND LESS RELIGIOUS EMPHASIS i Group N Prosocial Aggression Antisocial Aggression Total Slide Aggression Total Aggression Emotional Situ ations Test Less religious-high anger men 35 .24 - .11 .02 .00 Less religious-low anger men 32 - .06 .31 .13 .56* Less religious-high anger women 26 .09 .15 .38* .14 Less religious-low anger women 26 .33 - .13 .16 .26 More religious-high anger men 25 .22 .32 .29 .00 More religious-low anger men 23 .29 .32 .28 .31 More religious-high anger women 34 - .30 .19 - .11 - .21 More religious-low anger women 32 .19 - .09 - .16 .23 l*p less than .05, two tailed t test. « sj U CHAPTER VI DISCUSSION There are two conclusions that seem warranted by the results of this study. One is that men subscribe to a greater number of antisocial aggressive statements (antisocial acceptance scale items) , than do women. This tendency.is more powerful than the effects of levels of instigation and levels of religious emphasis, as used in the present experiment. The other conclusion is that degree of religious emphasis in the person's history and his current environment is strongly related to inhibition of a variety of forms of aggressive verbal behavior. The question of the relative power of these two effects of sex and religious emphasis was raised in con sidering the findings of Sears (1961) and Adorno et al. (1950). Sears found 12 year old girls more prosocially aggressive than boys. Adorno et al., did not find con clusive evidence of a similar sex difference. Instead, they found a difference in background, similar to that used in the present study, in degree of conventional 74 75 religious emphasis of thair adult subjects. The results of the present experiment support the results of Adorno et al. (1950). Apparently, sex differences in prosociality of aggression, so evident in Sears' (1961) study of juve niles, become less important with increasing age, and ideo logical factors, such as those studied by Adorno et al. (1950), increase in importance. However, the sex differ ences in attitude toward antisocial acts found in the present study are an important exception to the rule enunciated above. Even though a traditional, conservative upbringing, with a high degree of religious emphasis produces a significant lowering of all types of verbal aggressiveness, still, the sex role related greater ten dency for men to espouse, if not act on, antisocial aggres sive sentiments prevails. Sex typing of men's attitudes toward aggression, may be more powerful than imitative learning from power figures in the home or school (e.g., Bandura et al., 1963b; Brim, 1958), supporting the position taken by Parsons (195<m) that a societal expectation exists that males shatl be ag gressive, and that their aggression shall have negative connotations. It may be speculated that the antisocial, negatively valued aspect of masculine aggressiveness is maintained through a process of avoidance learning. If the rewards of filling the more desirable masculine role such as > 76 greater power, freedom and mobility (e.g., Brown, 1958; Parsons, 1954c; Seward, 1946) exceed the negative consequen ces of masculinity, such as having aggressive behavior punished and devalued (e.g., Meyer and Thompson, 1956; Wickman, 1936), and if the same aggressive behavior is considered by society as a sign of masculinity (Parsons, 1954c), then the negative valuation of such behavior by others may acquire reinforcing properties. For example, college men raised by conservative Christian parents and espousing the ideals of these parents, in terms of high value being placed on charity, meekness, etc., as responses, still acquiesced to statements of a clearly antisocial character significantly more often than college women raised in similar environments. The conclusions seems evident that the antisocial nature of these items served the function of a cue, evoking appropriate masculine behavior (acquiescence to antisocial aggression statements). Further, the evocative tendency of these items had to be capable of overriding the negatively reinforcing, or in hibiting, effects of religious indoctrination. Another result of the present experiment requiring discussion was the change induced in slide responses by increased instigation in the low religious emphasis group. Men's and women's scores were reversed in order of rank, with initially more aggressive men's scores, in the low instigation condition, decreasing significantly, while women'a acorea increased in the high instigation condition. Since these were independent aamplea, rather than repeated meaaurementa of the same subjects, it is possible that these unexpected results are due to sampling error; this seems unlikely, however, due to the adequate size of the sample (N=46). A possible explanation could be made by combining the concepts of aggression anxiety (Sears, 1961), and culture conflicts (Seward, 1945; Seward, 1956). It ia necessary to assume that the more modern, low religious emphasis group of men and women are subject to conflicting attitudes toward aggressiveness to a much greater degree than those subjects reared in more traditional religious backgrounds. The nature of these conflicting attitudes could be described as follows: The men are motivated to present themselves as aggressive and masculine, while the women are motivated to present themselves as desirable _ companions (Parsons, 1954c). Thus, in a state of low instigation, the men earn higher scores on the slide measures of aggression, than do women. When instigation level is increased, men react with aggression anxiety, inhibiting all forms of aggression, while women, unlike their less emancipated sisters in the high religious emphasis group, react by sumifesting significantly in creased antisocial aggression. It is as if the modern woman, seeing man's role as preferable, feels free to adopt it in this limited area of reacting to noxious stimuli, 78 while modern men, in a like situation, suddenly loses confidence in traditional masculine response styles and engages in avoidance behavior. An additional failure of prediction needing dis cussion is the failure of women's prosocial aggression scores to increase significantly under the high instigation condition. This failure, in the low religious emphasis group, could be explained in terms of women adopting the masculine role, with subsequent increase in antisocial aggression, rather than prosocial aggression tendencies. However, this explanation cannot be applied to the high religious group. An explanation may have been provided by Feshbach (1964), who stressed non-angry, incompatible responses, and also angry, but non-aggressive, modifying responses, as alternatives to hostile, angry responses, in situations where aggression instigating stimuli are present. Non-angry, incompatible responses could be deemed particularly appropriate alternatives for the high religious emphasis group in question (e.g., "turning the other cheek"). Limitations of this study were the relative weak ness of the instigating stimuli to arouse clearly differ entiating levels of anger, the high degree of intersubject variability, and possible confounding of subject variables such as social class and religious emphasis. It is ques tionable whether the high instigation group was as effectively aroueed as they might have been* had the more traditional insulting techniques been used instead. The high degree of intersubject variability in studies such as this has been a source of difficulty to other investi gators (e.g., Feshbach et al.. 1963), who have had re course to repeated measurements designs as the only practi cal way of getting meaningful results. The writer of the present study had hoped to avoid the problems of repeated measurements designs, especially in terms of rapid dis sipation of artificially induced states, such as anger, by using sufficiently large numbers of subjects in an inde pendent groups design that could be treated factorially (Lindquist, 1952). More serious than the above criticisms is the possibility of confounding of subject variables of religious; emphasis with social class. Although evidence was presented! of differences in religious emphasis between the schools from which subjects were drawn, and evidence was also presented of similarities of social class of students from both schools, direct measurement of these variables in each I i subject would have eliminated the possibility that the sub- j jects used were actually deviant from school norms, or the j possible entry of systematic biases into the distribution of these variables. The writer would make the following modifications in any future study of aggression: (1) The attitudinal measures (e.g., antisocial j acceptance scale, prosocial aggression scale, etc.) ; should be administered prior to manipulation of j anger instigation level. Although in this study the manipulation of this independent variable had no untoward effect on antisocial acceptance scale j scores, it seems pointless to take the chance of confounding these measures of attitude in this way. (2) A repeated measurements design would be used instead of a factorial design, to eliminate the effects of excessive intersubject variability, and j confounding of subject variables. It is possible that the predicted prosocial aggression changes, which unfortunately failed to achieve acceptable levels of significance in this study, might be better demonstrated in a repeated measurements design. (3) The method of arousing anger used in this study would be discarded in favor of face-to-face insult, which has been repeatedly demonstrated to be an effective anger arousing technique (e.g., Feshbach et al.. 1963). ~<4) The Emotional Situationa Test could be dis pensed with, since it adds little to the definition j of total aggression, which is adequately measured I by total slide aggression responses. CHAPTER VII SUMMARY A behavioral theory of sex role related differences in aggression habit patterns was derived from existing behavioral theories of aggression, theories of child development, and a sociological theory of sex role differ ences. It was hypothesized that men's aggression patterns would be characteristically antisocial, while women's would tend to be prosocial. It was also hypothesized that an important modifier of aggression habits is degree of religious emphasis, with high emphasis resulting in relatively more prosocial *h4n antisocial aggressiveness. Forty-eight men and 66 women undergraduate college students from a college with strong fundamentalist Christian orientation and 70 men and S3 women of comparable age and educational level from a college with no particular religious emphasis were randomly assigned to two conditions. Those subjects assigned to the higheanger instigation condition answered a set of 10 personality test items that differed significantly in anger arousing potential from 81 another set of 10 comparable items administered to sub jects in the low anger instigation condition. All subjects were then shown a set of 18 slides depicting figures of varying degrees of authority (girlfriend, teacher, mother, etc.) interacting with figures of varying degree of age similarity to the subjects (college men vs. juvenile boys). The depicted situations could readily be interpreted as anger arousing. Subjects were told to record briefly how they thought the slide figures felt, and these responses were scored for presence of words with hostile, angry, aggressive connotations. Three measures were taken: Angry responses attributed to authority figures (prosocial aggression), non-authority figures (antisocial aggression), and the sum of these. In addition, subjects completed an Emotional Situations test, designed to measure total aggression, which consisted of 20 empirically derived descriptions of anger arousing situations to be rated for degree of arousal on a scale of 0 to 100 points. Finally, subjects completed a factored scale of attitude toward antisocial aggressive behavior consisting of 16 personality test items. It was found that attitude toward antisocial aggressive behavior was related to sex, men manifesting significantly higher scores at both levels of religious emphasis. Attitude toward antisocial aggressive behavior scores of high religious emphasis subjects were signifi- 83 cantly lower than scores of low religious emphasis sub jects. There was no significant interaction between sex and religious emphasis of subjects. LIST 0 F REFERENCES LIST OF REFERENCES Adorno, Allison AmseL, Bach, 6 Bandura Bandura Bandura Bandura Barker, T. W., Frenkel-Brunswik, Else, Levinson, D. J., and Sanford, R. N. The authoritarian personality. New York: Harper & BrosTJ 1$56. , J., and Hunt, D, E. Social desirability and expression of aggression under varying conditions of frustration. J. consult. Psychol.. 1959, 23, 528-532. ~ I. The role of frustrative nonreward in noncontin- uous reward situations. Psychol. Bull.. 1958. 55. 102-119. . R. Father-fantasies and father-typing in father separated children. Child Develpm., 1946. 17. 63-80. , A., Ross, Dorothea, and Ross, Sheila A. Trans mission of aggression through imitation of aggres sive models. J. abnorm. soc. Psychol.. 1961, 63, 575-582. , A., Ross, Dorothea, and Ross, Sheila A. Imitation of film mediated aggressive models. J. abnorm. soc. Psychol.. 1963, 66, 3-11. (a) , A., Ross, Dorothea, and Ross, Sheila A. A comparative test of the status envy, social power, and secondary reinforcement theories of identificatory learning. J. abnorm. soc. Psychol.. 1963, 67, 527-534. (b) , A., Ross, Dorothea, and Ross, Sheila A. Vicarious reinforcement and imitative learning. J. abnorm. soc. Psychol.. 1963, 67, 601-607. (c) R. 6., Dembo, Tamara, and Lewin. K. Studies in topological and vector psychology. II. Frus tration and regression: An experiment with young children. Univer. Iowa stud, child welfare. 1941, 18, (Whole NoTTT. 85 86 Barry, H. Ill, Bacon, Margaret K., and Child, I. L. A croaa-cultural aurvay of some sex differences in aocialization. J. abnorm. aoc. Pavchol.. 1957, 55, 327-332. ” Beach, F. A. The deacent of inatinct. Psychol. Rev.. 1955, 62, 401-410. Beliak, L. The concept of projection. Psychiatry. 1944, 2, 353-370. Berkowitz, L. The expression and reduction of hostility. Psychol. Bull.. 1958, 55, 257-283. Berkowitz, L. Aggression: A social psychological analysis. New York: McGraw-Hill, 1962. Berkowitz, L., Green, J. A., and Macauley, Jacqueline R. Hostility catharsis as the reduction of emotional tension. Psychiatry. 1962, 2j>, 23-31. Berkowitz, L., and Rawlings, Edna. Effects of film violence on inhibitions against subsequent ag gression. J. abnorm. soc. Psychol.. 1963. 66. 405-412. Block, Jeanne and Martin B. Predicting the behavior of children under frustration. J. abnorm. soc. Psychol.. 1955, 51, 281-285. Brim, O. G., Jr. Family structure and sexrole learning by children: A further analysis of Helen Koch's data. Sociometry. 1958, 21^ 1-16. Brown, D. G. Sex-role preference in young children. Psychol. Monogr.. 1956, 70, (14, Whole No. 421). Brown, D. G. Sex-role development in a changing culture. Psychol. Bull.. 1958, 55, 232-242. Brown, J. S., and Farber, I. E. Emotions conceptualized as intervening variables with suggestions toward a theory of frustration. Psychol. Bull.. 1951, 48, 465-495. Buss, A. Jhe psychology of aggression. New York: Wiley, 1961. Child, I. L., Potter, E. H., and Levine, Estelle M. Children's textbooks and personality development: An exploration in the social psychology of edu- cation. Psychol. Monogr.. 1946, 60, No. 3, entire. 87 Cohen, A. R. Social norma, arbitrariness of frustration and status of the agent of frustration in the frustration-aggression hypothesis. J. abnorm. soc. Psvchol.. 1955, 51, 222-226. Davitz, J. R. The effects of previous training on post frustration behavior. J. abnorm. soc. Psychol., 1952, 47, 309-315. Dollard, J., Miller, N. E., Doob, L. W., Mowser, 0. H. , I and Sears, R. R. Frustration and aggression. New Haven: Yale University Press, 1939. Feshbach, S. The drive-reducing function of fantasy behavior. J _. abnorm. soc. Psychol.. 1955, 50, 3-11. Feshbach, S. The function of aggression and the regulation of aggressive drive. Psychol. Rev.. 1964, 71, 257-272. Feshbach, S., Singer, R. D., and Feshbach, Norma. Effects of anger arousal and similarity upon the attri bution of hostility to pictorial stimuli. J. consult. Psychol., 1962, 27, 248-252. Freud, Anna. The ego and the mechanisms of defense. London: Hogarth, 1937. Gillespie, J. Aggression in relation to frustration, attack and inhibition. Unpublished doctor's dissertation, University of Pittsburg, 1961. Cited in A. Buss, The psychology of aggression. New York: Wiley, 19<>1, pp. 51-32. Goodenough, Florence L. Anger in young children. Child Welfare Monogr. No. 9> , Minneapolis: University of Minnesota Press, 1931. Gordon, J. E., and Cohn, Fay. The effects of affiliation drive arousal on aggression in doll interviews. Unpublished experiment, 1961. Reported in L. Berkowitz, Aggression: A social psychological analysis. New York: McGraw-Hill, 1962. Grayson, H. M., and Tolman, Ruth S. A semantic study of concepts of clinical psychologists and psychia trists. J. abnorm. soc. Psvchol.. 1950. 45. 216-231. “ — 88 Haner, C. F., and Brown, Patricia A. Clarification of the instigation to action concept in the frustration-aggression hypothesis. J. abnora. soc. Psvchol.. 1955, 51, 204-206. Jacobs, A., Capek, L., and Meehan, J. P. The development of an adjective checklist to measure affective states. Psvchol. Newsletter, 1959, 10, 115-118. Janis, 1. L., and Field, P. B. Sex differences and per sonality factors related to persuasibility. In I. L. Janis, C. 1. Hovland, et al., Personality and persuasibility. New Haven: Yale University Press, 1^5$. Jeraild, A. T., and Markey, F. V. Conflicts between preschool children. Child develpm. Monogr.. 1935, No. 21. Lansky, L. M., Crandall, V. 0., Kagan, J., and Baker, C. T., Sex differences in aggression and its correlates in middle-class adolescents. Child Develpm.. 1961, 32, 45-48. Levin, H., and Sears, R. R. Identification with parents as a determinant of doll play aggression. Child Develpm.. 1956, 27, 135-153. Levy, D. M. The hostile act. Psychol. Rev.. 1941. 48. 356-361. — Lindquist, E F. Peaign and analysis of experiments in psychology and education. Boston: Houghton- McClelland, D. C., and Apieella, F. S. A functional classification of verbal reactions to experiment ally induced failure. J. abnorm. soc. Psvchol.. 1945, 40, 376-390. ~ McClelland, D., Atkinson, J W., Clark, R. A., and Lowell, i E. L. The achievement motive. New York: Apple ton-Century-Crofts, 1953. Maier, N. R F. Frustration. New York: McGraw-Hill, 1949: Maslow, A. H. Deprivation, threat, and frustration. Psvchol. Rev.. 1941, 48, 364-366. 89 Maslow, A. H. Self-esteem (dominance-feeling) and sexuality in women. J. Soc. Psvchol.. 1942, 16, 259-294. Mees, H. L. Preliminary steps in the construction of factor scales for the MMP1. Unpublished doctor's dissertation, University of Washington, 1959. Meyer, W. J., and Thompson, G. G. Sex differences in the distribution of teacher approval and disapproval among sixth grade children. J. Ed. Psvchol.. 1956, 42, 385-396. Miller, N E. The frustration-aggression hypothesis. Psvchol. Rev.. 1941, 48, 337-342. Milner, E. Effects of sex role and social status on the early adolescent personality. Genet. Psvchol. Monogr.. 1949, 40, 231-325. Mowrer, 0. H. Learning theory and personality dynamics. Mew York: Ronald Press, 1950, pp. 69-94. Mussen, P. H., Conger, J. J., and Kagan, J. Child dev< Row development and personality. New York: Harper & , 1963. Mussen, P. H., and Distler, L. Masculinity, identification, and father-son relationships. J. abnorm. soc. Psvchol.. 1959, 59, 350-356. ~ Mussen, P. H., and Naylor, H K. The relationship between overt and fantasy aggression. J. abnorm. soc. Psvchol.. 1954, 49, 235-240. ~ Otis, Nancy B., and McCandless, B. Responses to repeated frustrations of young children differentiated according to need area. J. abnorm. soc. Psychol.. 1955, 50, 349-353. Parsons, T. Certain primary sources and patterns of aggression in the social structure of the western world. In T. Parsons, Essays in sociological theory. Glencoe, Illinois: Free Press, 1954. (a) Parsons, T. The kinship system of the contemporary United States. In T. Parsons, Essays in sociological theory. Glencoe, Illinois: Free Press, 1954. (b) Parsons, T. Age and sex in the social structure of the United States. In T. Parsons, Essays in sociolo gical theory. Glencoe, 111. : Free Press, 1954. (c) 90 Paatorc, N. The role of arbitrariness in the frustration- aggression hypothesis. J. abnorm. soc. Psvchol.. 1952, 47, 728-731. ” Rabban, M. Sex-role identification in young children in two diverse social groups. Genet. Psvchol. Monogr.. 1950, 42, 81-158. Reiser, M. F., Reeves, R. B., and Armington, J. Effects of variations in laboratory procedure and experi ments upon ballistocardiogram, blood pressure, and heart rate in healthy young men. Psychosom. Med.. 1955, 17, 185-199. Rosenblith, Judy F. Imitative color choices in kinder garten children. Child Develpm., 1961, 32, 211-223. Rosenblith, Judy F., and Allinsmith, W. (Eds.) The causes of behavior: readings in child development and e5ucatxonal psychology. Boston: Allyn and Bacon, 1962. Scott, J. P. Aggression. Chicago: University of Chicago Press, 1958. Sears, Pauline S. Doll play aggression in normal young children: influence of sex, age, sibling status, father's absence. Psvchol. Monogr.. 1951, 65. Mo. 42. Sears, R. R. Relation of early socialization experiences to aggression in middle childhood. J. abnorm. soc. Psvchol.. 1961, 63, 461-465. Sears, R R., Maccoby, Eleanor E., and Levin, H. Patterns of child rearing. Evanston, 111.: Row, Peterson, 1957: Sears, R. R., Pintler, M H., and Sears, Pauline. Effect of father separation on preschool children's doll play aggression. Child Develpm.. 1946, 17, 219-243. Seward, Georgene H. Culture conflict and the feminine role: an experimental study. J. soc. Psvchol.. 1945, 22, 177-194. “ Seward, Georgene H. Sex and the social order. Mew York: McGraw-Hill, 91 Seward, Georgene H. Learning theory and identification: V. some cultural aspects of identification. J. genet. Psvchol.. 195**, 84, 229-236. Seward, Georgene H. Psychotherapy and culture conflict. New York: Ronald Press,1956. Seward, Georgene H. Sex identity and the social order. iL* ncrv» ment. Disease. 1964, 139, 126-136. Terman, L. M., and Miles, Catherine L. Sex and person ality. New York: McGraw-Hill, 19^6. Terman, L. M., and Tyler, Leona E. Psychological sex differences. In L. Carmichael (Ed.) Manual of child psychology (2nd ed.) New York: Wiley, 1954. Thompson, E. G., Jacobs, A., and Meehan, J. P. The effect of emotional stimulus words on response measures. Paper presented at Western Psychol. Assn., San Francisco, April, 1962. Underwood, B. J. Psychological research. New York: Applet on-Century-Crofts, 1937. University of Southern California, Bulletin of the College of Letters, Arts, and Sciences. Los Angeles: 1964-1965. Welsh, G. S. Factor dimensions A and R. In G. S. Welsh and W. G. Dahlstrom (Eds.) Basic readings in the MMPI in psychology and medicine. Minneapolis’ : University of Minnesota Press, 1956. Westmont College Catalogue, Santa Barbara, California: 1965-1965. Whiting, J. W. M., and Child, I. L. Child training and personality: a cross-cultural study. New Haven: Yale University Press, 1953. . Wickman, E. K. Teachers and behavior problems. New York: Commonwealth Fund, 1938. Worchel, P. Catharsis and the relief of hostility. J. abnorm. soc. Psvchol.. 1957, .55, 238-243. "" Worchel, P. Status restoration and the reduction of hostility. J. abnorm. soc. Psvchol.. 1961. 63. 443-445. ”----------------- ---- 92 Yarrow, L. J. Tha affect of antecedent frustration on projective play. Psychol. Monogr.. 1948, 62. No. 293. APPENDICES APPENDIX A ITEMS USED IN THE DEVELOPMENT OF THE HIGH ANGER AND LOW ANGER INSTIGATING LISTS (Items are- separated into tilgh anger and low anger categories for convenience, and rating scales omitted) Anger Arousal Potential Mean Rating Male Female Item High Anger Items 2. I have been drunk . . . 27.23 4. In my lifetime I have tried narcotics . . . 44.58 6. In the past I have seen my mother drunk . . . 74.55 8. I have cheated on tests . . . 35.91 10. I have had more intelligent friends help me do written assignments . . . 32.50 12. I tend to lie about why I am late . . . 32.82 14. I pretend to be sick to get out of going to school or work . . . 42.60 16. I pretend to understand discussions about people or things I really have little knowledge about . . . 18.02 18. I have tried to give up smoking and failed . . . 29.71 20. I lose my temper and raise my voice at others . . . 28.48 42.00 39.00 50.00 38.49 15.10 36.99 29.60 49.01 20.00 41.00 Total 366.40 361.19 Percent of possible anger arousal potential 36.64 36.12 95 Item Low Anger Items Anger Arousal Potential Mean Rating Male Female 1. I like mechanics magazines ... 17.02 3. There seems to be a lump in my throat . . . 35.48 5. My sex life is satisfactory . . . 49.22 7. I feel like swearing ... 24.02 9. I would rather sit and daydream than do anything else . .. 38.70 11. My soul leaves my body ... 39.01 13. I do not read every editorial in the newspaper . . . 15.23 15. I used to like drop-the- handkerchief . .. 22.60 17. I tease animals ... 31.21 19. I put off until tomorrow what I ought to do today . . . 22.60 1.00 22.49 54.50 39.00 33.50 43.51 6.00 11.00 61.50 34.99 Total 295.09 307.49 96 APPENDIX B COMPOSITION OF HIGH ANGER AND LOW ANGER LISTS OF INSTIGATING ITEMS (REVISED) (Item number refer* to piece on reting list, ratings identical to Appendix A, where available) Item High Anger Items Anger Arousal Potential Mean Rating Male Female 6. In the past I have seen my mother drunk ... 74.55 50.00 5. My sex life is satisfactory . . .49.22 54.50 4. In my lifetime I have tried narcotics . . . 44.58 39.00 5. My soul leaves my body ... 39.01 43.51 8. I have cheated on tests . . . 35.91 38.49 12. I tend to lie about why I am late .. . 32.82 36.99 16. I pretend to understand dis cussions about people or things I really have little knowledge about ... 18.02 49.01 2. I have been drunk . . . 27.23 42.00 14. I pretend to be sick to get out of going to school or work .. 42.60 29.60 9. I would rather sit and daydream than do anything else . .. 38.70 33.50 Total 402.64 416.00 Percent of possible anger arousal potential 40.26 41.60 98 Item Low Anger Items Anger Arousal Potential Mean Rating Male Female I. I like mechanics magazines . . . 17.02 I.00 13. I do not read every editorial in the paper . . . 15.23 6.00 15. I used to like drop-the- handkerchief . . . 22.60 11.00 19. I put off until tomorrow what I ought to do today . . . 22.60 34-.99 18. I have tried to give up smoking and failed . . . 29.71 20.00 20. I tend to lie about why I am late . . . 32.82 36.99 10. I have had more intelligent friends help me do written assignments . . . 32.50 15.10 3. There seems to be a lump in my throat . . . 35.48 22.49 Added: I like to take walks in the country . . . (est.>26.00 18.45 Added: I like to hear good music . . . Cest.)26.00 18.45 Total 207.96 147.57 99 APPENDIX C DESCRIPTIONS OF SLIDES USED IN OBTAINING SLIDES-ANGER MEASURES Slide Number Description 1 Mother figure seated at kitchem table, cup at elbow, college-age boy standing looking at her, with cup in hand. She says, "Why do you always ask me so late?" 2 Mother figure seated, using telephone. College-age boy standing, coat on, hands on hips, looking at her. She says, "Can you wait until I make just one more phone call before I drive you to the station?" 3 College-age sales girl standing at a toy counter, hand on a stuffed animal, looking at another college-age girl, who is watching the animal, and is obviously a customer. Salesgirl says, "We only have the floor model left and we can't sell it." < 4 - Juvenile age boy sitting on floor in front of TV set, looking up, mouth open, at mother figure in apron, hands on hips, who says, "Isn’t that program over yet?" 5 Two college-age men facing each other, one has arm out, hand on other's shoulder. No caption. 6 College-age coy, arm out and hand on shoulder of juvenile age boy. Older boy says, "I can't take you to the game today but I will next week." 7 College-age boy looking at college-age girl, who has coat on, is standing at door, hand on knob, and is looking back at boy. No caption. 8 Juvenile age boy sitting on couch facing sideways to father figure sitting on other end, facing boy. Father says, "I hear your friend Bobby was elected to the honor society." 9 College boy sitting in restaurant. Portly waiter, dressed in formal outfit says, "If you don't have a jacket with you we can't serve you." 101 Slide Number Description 10 Juvenile boy, jacket on, cap in hand, hand on door knob, looking back at mother figure who is looking at boy. No caption. 11. Mother figure at sink, in apron, looking over shoulder at juvenile boy, who stands facing camera, back to mother, holding broom or mop handle. No caption. 12 Classroom scene, with juvenile boy standing beside desk, looking at woman teacher, who is standing in front of class looking at him. No caption. 13 Father figure, hat on, golf bag over shoulder, at door. Juvenile boy, back to camera, looking at father, left hand raised to father. No caption. 1M- Male teacher, cigar in hand, seated at desk, behind him a clock mounted on wall, sur mounted by a rifle. College boy standing beside him, holding out paper, looking at it, teacher, gesturing at paper ask, "Do you think this is an acceptable paper?" 15 Same as slide 10, but different boy and mother. 16 Two college boys, one holding necktie, who asks other, "You don*t mind if 1 borrow your tie?" 17 Same as slide 7, but girl has dark, rather than blond, hair. No caption. 18 Older man behind desk holding sheaf of papers, examining them, while younger man leans over desk attentively. No caption. 102 APPENDIX D COPY OF THE EMOTIONAL SITUATIONS TEST USED TO MEASURE AGGRESSIVE RESPONSES This is a test to see how much anger different kindi of situations arouse. Imagine a scale running from O to 100 with 0 standing for no feeling or complete indiffer ence and 100 standing for the most anger possible to ex perience. Read each item, imagining yourself in the sit uation described, and decide how much feeling on this imaginary scale you have. Write the number just to the left of the item in the space provided. 1. You find you just can’t reason with some person. 2. Someone lies to you. 3. Someone treats you unjustly. 4. Someone becomes a nuisance. 5. You hear that your fiancee has been going out with someone else. 6. Your parents Cor boss) interfere with your plans. 7. You make a stupid mistake. 8. Someone attacks and hits you. 9. Someone antagonizes you by always disagreeing and arguing. 10. Someone treats you with lack of consideration and rudeness. 11. Someone forces you to do something you don’t want to do. 12. You do something you vowed not to do. 13. Someone treats you with disrespect and makes derogatory remarks. 14. Others expect too much of you. 15. You discover you have been left out of the group’s plans. 16. Your car won't start when you are in a hurry. 17. Your wishes are disregarded and you are denied something you wanted. 18. Someone lets you down by not doing what they said they would do. 104 19. 20. COPY OF THE EMOTIONAL SITUATIONS TEST (Cont.) Someone is critical and angry at you. Someone unjustly and falsely accuses you of something. 105 APPENDIX E COPY OF ATTITUDE TOWARD ANTISOCIAL ACTS SCALE (AA) , USED TO MEASURE ATTITUDE TOWARD ANTISOCIAL AGGRESSIVENESS. (KEYED ANSWERS SHOWN) Please complete the following: (True or False) 1. I have often had to take orders from someone who did not know as much as I did. (True) 2. I emjoy a race or game better when I bet on it.(True) 3. At times I feel like picking a fist fight with someone. (True) It is not hard for me to ask help from my friends even though I cannot return the favor. (True) 5. I don't blame anyone for trying to grab everything he can get in this world. (True) 6. If several people find themselves in trouble, the best thing for them to do is to agree on a story and stick to it. (True) 7. I like to read newspaper articles on crime. (True) 8. I feel that it is certainly best to keep my mouth shut when I'm in trouble. (True) 9. In school I was sometimes sent to the principal for cutting up. (True) 10. It wouldn't make me nervous if any members of my family got into trouble with the law. (True) 11. When I get bored I like to stir up some excitement. (True) 12. At times I feel that I can make up my mind with unusually great ease. (True) 13. I have been inspired to a program of life based on duty which I have since carefully followed. (True) 14. I am entirely self-confident. (True) 15. I am embarrassed by dirty stories. (False) 16. I enjoy gambling for small stakes. (True) 107 APPENDIX F CHECKLIST OF WORDS USED TO SCORE SLIDES FOR SLIDES-ANGER SCORES Annoyed Angry Mad Hostile Indignant Disgusted Jealous Resentful Reproachful Envious Mean Irritated Bitter Rebellious Sarcastic Antagonistic Defiant Exasperated Aggravated Peeved Scornful Piqued Contemptuous Cross Provoked Animosity Rage Belligerent 109

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Sex-Role Preferences Of Early Adolescents In Relation To Adjustment

PDF

An Investigation Into The Communication Style Of Suicidal Individuals

PDF

First Steps In An Attempt To Construct An Objective Test Of Character Structure

PDF

Behavioral Seriousness And Impulse-Control Balance In Delinquency

PDF

Role Playing Ability And Social Adjustment In Children

PDF

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

PDF

Prediction Of Therapeutic And Intellectual Potential In Mentally Retardedchildren

PDF

Self-Worth, Future Goals, Mood And Time Perception

PDF

Improvement Rate In Tuberculosis As A Function Of Accessibility To Repressed Emotional Disturbance

PDF

Perceptual Defense And Somatization: A Comparison Of The Perceptual Thresholds Of Obese And Peptic Ulcer Patients

PDF

Delinquency As A Function Of Intrafamily Relationships

PDF

Attitudes Of Ministers And Lay Leaders Of The American Baptist Conventionof The State Of Washington On Selected Social Issues

PDF

Acculturation And Value Change

PDF

Empathy And Social Perception

PDF

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

PDF

A Temporal Approach-Avoidance Conflict In An Academic Test Situation

PDF

Measuring Thought Process As An Ego Function In Schizophrenic, Mentally Retarded And Normal Adolescents By Means Of The Rorschach

PDF

Predicting Success In The Study Of Descriptive Linguistics

PDF

Conflict And Generalized Conflict In Verbal And Motor Responses

PDF

The Use Of Multiple Objective Measures In Survey Research As A Means For Predicting Respondent Future Behavior

Asset Metadata

Creator Thompson, Eric Guy (author)

Core Title An Experimental Study Of Aggression Habit Patterns And Sex Role

Degree Doctor of Philosophy

Degree Program Psychology

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

Tag OAI-PMH Harvest,psychology, general

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Jacobs, Alfred (committee chair), Arnold, Aerol (committee member), Seward, Georgene H. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-184228

Unique identifier UC11359820

Identifier 6600583.pdf (filename),usctheses-c18-184228 (legacy record id)

Legacy Identifier 6600583.pdf

Dmrecord 184228

Document Type Dissertation

Rights Thompson, Eric Guy

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