V ‘ ‘Tfivg— — THE EFFECT OF VIEWING _ PHYSICAL CONSEQUENCES OF VIOLENCE 0N PERCEPTIONS AND AGGRESSIVENESS Thesis for the Degree of Ph. D. 7 MICHIGAN STATE UNIVERSITY THOMAS F. GORDON ' 1 9 7 3 . LIBRARY L Mishigan Sm: QWZIZZ/M/QILMMI/IW/ W 4 THE EFFECT OF VIEWING PHYSICAL CONSEQUENCES OF VIOLENCE 0N PERCEPTIONS AND AGGRESSIVENESS presented by Thomas Frank Gordon has been accepted towards fulfillment of the requirements for Ph . D. degree in Comuni cation 0-7639 {v ‘1 luubl‘hhdlln " x 4 N Oxford O5 1 WWW ELL: 08 C“ 1 no A 16 10:51 0110:9321" r "r-qlu “"v‘\ m, 95 1731)”; .M-. .- oéPRGNQUUZ A”: \ ‘1’) 09! ff @38'8533; ’ use V'J ABSTRACT THE EFFECT OF VIEWING PHYSICAL CONSE— QUENCES OF VIOLENCE ON PERCEPTIONS AND AGGRESSIVENESS by Thomas Frank Gordon (This research focused on the effects of viewing television violence. Three television manipulations were compared: (1) Violence with_physical conse— quences, (2) violence without consequences, and (3) nonviolence. The study compared levels of: (1) Social class, (2) fantasy aggression, and (3) aggression anxiety to determine the effect of these factors on perceptions of and aggressive reactions to violence with and without consequences2> The subjects were 283 seventh, eighth, and ninth grade boys from middle- and lower—class junior high schools. METHODS: The two violent sequences were edited versions of the same fist fight, taken from the film The Chase. In this scene, a sheriff is being severely beaten. The physical consequences evident include excessive beatingb blood, swollen face, uncoordinate movement, and sslurred speech. The no consequences version used 1 2 filters to remove the blood, other physical conse- quences were edited out. The nonviolent version used scenes from the same film, thus holding the major characters constant. A consistent length and story line were maintained. Aggressive reactions to the scenes were assessed through a set of situation items presenting potentially aggression provoking encounters. Perceptions of the scenes were assessed on the dimensions of perceived violence, acceptability of the behavior, and professed liking for the content. The following controls were used: The variables of race, sex, age, and social class were controlled in the study design. The physical consequences of the action were controlled in the stimulus materials. Assign- ment of subjects to treatments and time of presentation were controlled in the execution of the study. In the analyses, the effect of the boys' overt aggression level and the effects of differential perceptions of the scenes were accounted for. RESULTS: ‘Perceptions.-—The violence with and without con- sequences, as compared to the nonviolence, was liked more, seen as more violent and less acceptable. The 3 consequences vs. no consequences comparison produced the same trends but only the perceived violence dimen- sion was statistically significant. Social Class Differences.--The major social class interaction hypotheses received partial support. For the consequences vs. nonviolence comparison, there was a marginally significant three—way interaction in- volving social class, the boys normal overt aggression level, and the TV treatments. Boys high in overt aggression were more aggressive than boys low in overt aggression and this effect was greater for lower-class boys than for middle—class boys. In the no consequences vs. nonviolence compari- son, a similar three—way interaction was evident. Here, however, the pattern was broken by the lower—class boys high in overt aggression; they were more aggressive in the nonviolent condition than in the violent condition. Fantasy Aggression.--As expected, the fantasy aggression level of the child did not relate to his aggressive reactions to the scenes. Specifically, no aggression differences were noted for boys high in overt aggression and low in fantasy aggression. Aggression Anxiety.--The predicted interaction effects involving aggression anxiety and social class did not develop. However, boys high in aggression A anxiety did evidence less aggressiveness, liked the violent scenes less and indicated that the violent scenes were more violent than did the boys low in aggression anxiety. CONCLUSIONS: The overall findings of the study fail to demon— strate that violence with consequences differs from violence without consequences, as related to aggressive tendencies. With the exception of perceived violence, the same was true of the perception dimensions. Suggestions for extended research are offered. THE EFFECT OF VIEWING PHYSICAL CONSEQUENCES OF VIOLENCE ON PERCEPTIONS AND AGGRESSIVENESS by Thomas Frank Gordon A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Communication 1973 Accepted by the faculty of the Department of Communication, College of Communication Arts, Michigan State University, in partial fulfillment of the requirements for the Doctor of New. SCEW‘ 1HectorgFThes1fl Philosophy degree. Guidance Committee: ‘ Chairman ACKNOWLEDGMENTS To my committee, I owe a sincere thanks: Dr. Bradley S. Greenberg, Dr. Verling Troldahl, Dr. Thomas Baldwin, and Dr. Andrew Barclay. I am deeply indebted to Dr. Bradley S. Greenberg. Throughout my doctoral study he provided me with research privileges that few graduate students have the opportunity to experience. It is impossible to express the personal and academic effects of this mutual con- fidence. As a close friend, Dr. Edward Wotring provided assistance at all stages of the project, a reflection of my respect for his initiative and competence. Dr. John Michelson also provided the assistance which could only be asked of a friend and colleague. My wife Nancy provided the stability for our journey through undergraduate and graduate school. She knows my full gratitude. iii TABLE OF CONTENTS ACKNOWLEDGMENTS LIST OF TABLES LIST OF FIGURES . . . . . . . . LIST OF APPENDIXES. Chapter I. INTRODUCTION The Research Focus. . The Effect of TV Violence Content Differences Social Class Differences Fantasy Aggression as a Mediator Aggression Anxiety II. METHODS The Sample. . Variables and Operationalizations Television Treatment Antecedent Variables. Social Class. Fantasy Aggression Aggression Anxiety Perceptions Control Variables Overt Aggressivenesse_fl Parental Attitudes Exposure to TV Violence.l iv Page iii vi oviii ix 2O 2O 2O 21 “£33 \ ‘\\j Chapter Basic Study Design General Procedures Analytic Procedures. III. RESULTS. Television Treatment Differences Situational Aggressiveness Perception Differences Classification Differences by Dependent Measure . . . Situational Aggressiveness Perception Differences Professed Liking Perceived Acceptability Perceived Violence~V// Summary of Findings. Situational Aggressiveness Differences Professed Liking Perceived Acceptability Perceived Violence\v/ IV. DISCUSSION Television Treatment Differences Social Class Differences Fantasy Aggression Aggression Anxiety Summary Research Extensions. BIBLIOGRAPHY APPENDIX Page 39 110 L13 145 “5 50 87 88 98 102 108 Table 10. ll. l2. 13. LIST OF TABLES Summary of Hypotheses Aggression Anxiety Items and Intercor— relations . . . . . . . . . . . . Situational Aggression Items and Inter— correlations. . . . . . . . . . . Factor Analysis of Perception Items Overt Aggression Items and Intercorrela— tions Items for Parental Attitudes Toward Aggression and Intercorrelations. Overall Television Treatment Effects by Dependent Variables Simple (r 2) and Partial (rl.2) Correla- tions 0% Variables with Dependent Variables Median Values Analyses of Variance for Social Class (SES), TV Treatment, and Overt Aggressiveness (OA) . Analyses of Variance for Aggression Anxiety (AA), TV Treatment and Overt Aggressiveness (OA) . . . . Analyses of Variance of Fantasy Aggres- sion (FA), TV Treatment, and Perceived Acceptability (PA) Analyses of Variance for Aggression Anxiety (AA), TV Treatment and Per- ceived Acceptability (PA) vi Page 19 25 26 31 35 37 A9 52 55 58 62 6A 67 Table 14. 15. l6. 1?. Analyses of Variance for Fantasy Aggres— sion (FA), TV Treatment, and Professed Liking (PL) Analyses of Variance for Aggression Anxiety (AA), TV Treatment, and Pro- fessed Liking (PL). . . . . . Analyses of Variance for Fantasy Aggres- sion (FA), TV Treatment and Perceived Acceptability (PA). Analyses of Variance for Aggression Anxiety (AA), TV Treatment and Per— deived Acceptability (PA) . vii Page 71 7A 77 8O LIST OF FIGURES Page Figure 22 l. TAT Stimulus Pictures 2. Study Design . . . . . . . . . . . . . . Al viii LIST OF APPENDIXES Page APPENDIX: INSTRUMENT . . . . . . . . . . . . . . 108 ix CHAPTER I INTRODUCTION The American conscience has been stirred by recent dramatic acts of violence. Assassinations of political leaders, ghetto riots, mass murders, the Vietnam war, and violent demonstrations against the war have all combined to focus public attention on the problem of violence in American society. The social scientist has been challenged to examine relevent aspects of aggressive behavior in an effort to under— stand and prevent the occurrence of such tragedies. The ultimate goal is to control or channel aggressive behavior. Television entertainment has been accused as a potential contributor to aggressive behavior. The per- vasiveness of the medium is unquestioned. (Census figures place the number of U. S. households with at least one TV set at more than 95 percent.) As Singer (1971) points out, most Americans, including about 90 percent of the U. S. armed forces, probably have never personally seen anyone wounded or killed. However, incidents of such violence are commonplace in television 1 OK I '3 . 2 entertainment (Gerbner, 1972). (Emma, the primary source of exposure to violence and aggression in American society is television entertainment and news) fiiThe Research Focus Building on that assumption, this experimental study was designed to examine factors which may affect adolescents' perceptions of and reactions to television violence. Characteristics of the violent content, namely physical consequences of aggressive behavior, were manipulated. Primary viewer characteristics of concern were (1) social class, (2) fantasy aggression level, and (3) aggression anxiety level. The study examined: 1. The effect of explicit physical consequences on adolescent's perceptions of and reactions to scenes of violence. 2. Social class differences in adolescents' perceptions of and reactions to televised violence with and without physical consequences. 3. The relationship between the adolescent's fantasy aggression level and his perceptions of and reactions to violence with and without consequences. A. The effects of the adolescent's aggression anxiety level on his perceptions of and reactions to violence with and without physical consequences. 3 Measures of the youngster's overt aggressive- ness, his general exposure to TV violence, and his parent's attitudes toward aggression were assessed as possible control variables. Research shows that these variables correlate with aggressive reactions to tele- vised violence (Dominick & Greenberg, l972; Feshbach & Singer, 1971; Siegel, l969). The study foci will be elaborated in this chapter and study hypotheses developed. The Effect of TV Violence Content Differences The bulk of research evidence supports the proposition that television violence has a facilitative or stimulative effect (Berkowitz, 1962; Baker & Ball, 1969; Goranson, 1969; Weiss, 1969). This proposition has been qualified, however, for certain content condi- tions. For example, it has been demonstrated consis- tently that the more the aggressor is punished, and/or the aggressive behavior is unjustified, the less likely it is the viewer will behave aggressively in a post- viewing situation (Berkowitz, 1962, 1964, 1965; Berko— witz & Rawlings, 1963; Berkowitz, Corwin & Heironymous, 1963; Bandura, 1965; Bandura, Ross, & Ross, 1963; Walters, at 31., 1965)- A More recently, it has been suggested that the depiction of the physical consequences of violence (e.g., blood and suffering) will reduce aggressive tendenciesf\\This proposition has received less em— pirical support. Two studies, one unpublished, have demonstrated thatiif the negative consequences of k/ violence are shown} subsequent aggressive behavior will be diminished>‘ Bramel, Taub, and Blum (1968) demonstrated that when a subject observed the suffering of a confederate who had earlier insulted him, he was less punitive than if he had not seen him suffer. The second study (Tannenbaum & Goranson, unpublished), as reported by Goranson (1969a), angered subjects, then had them view a prize fight and listen to a tape recorded ending. Subjects were less punitive when the consequences were negative (suffering and death) than when the outcome was neutral (reviewing the fight) or positive (success and fame). The rationale for the findings of these two studies posits that negative consequences served to evoke aggression anxiety in the subject (i.e., anxiety about committing acts of aggression), thus inhibiting his aggressive behavior. This inhibitory reaction Should hold equally well whether the negative conse- Quences are applied to the "good guy" or the "bad guy," 5 although justification for the action should mediate the effect. If the action were highly justified, the aggression anxiety arousal would be somewhat diminished. Goranson (1969b) cites a study by Ihrtman (1969) which manipulated pain-cues. Three experimental films were constructed. The control version showed a group of boys playing a vigorous but nonviolent basket— ball game. In the experimental versions, the game was interrupted by an argument between two boys which resulted in a fist fight. In one version, the film focused on the attacker's responses; his punching and kicking, angry facial expressions, and aggressive verbalizations. In the "pain-cues" version, the camera focused on the victim with closeups of his face as he is knocked down, his groans, cries, and other expressions of distress. A pseudo learning task, shock procedure, was used to assess the subjects' aggressiveness. Con- currently, varying degrees of pre—film aggressive arousal were generated by having a confederate make either insulting or neutral comments about the subject. The results showed that in comparison to the control group, the effect of both aggressive film versions was to increase aggressiveness. This effect was different for aroused and nonaroused subjects. 'lp 6 The aroused subjects were more aggressive after the "pain-cues" version than were the nonaroused subjects. For the nonaroused subjects, aggressiveness was lower after the "pain-cues" version than after the version without such cues. Although the lower aggressive behavior of the nonaroused subjects in this study was not significant, the direction of the effect was con- sistent with the studies cited. The prior arousal appears to have overcome or suppressed any aggression anxiety raised by the pain—cues. In general then, with the tentative exception of prior aggressive arousal, filmed aggressive action which presents negative physical consequences is expected to diminish aggressive tendencies compared to violent action without visual consequences. The process by which this effect functions is not well understood. The proposition that aggression anxiety serves to inhibit aggressive behavior is tenable (Staub, 1971; Megargee, 1971; Goranson, 1969b). The present re- search examined the effect of portraying physical consequences on adolescent boys, through the following hypotheses: i? H : Boys exposed to violence with consequences 1 will exhibit less aggression than boys ex— posed to nonviolence. 7 H : Boys exposed to violence without conse- quences will exhibit more aggression than boys exposed to nonviolence. The question of catharsis is of immediate concern here. [The classic catharsis formulation would contend that watching violence serves to "drain" ag- gressive tendencies, leaving the viewer less likely to This L- _../ perform acts of aggression (Feshbach, 1961); should hold for both the violence with and without consequences. With the aggression anxiety formulation, on the other hand,[the subject may still be aggressively aroused but his overt behavior is inhibited or sup- pressed?) Thus, when aggressive action is witnessed and this action is paired with aggression anxiety arousing cues (e.g., blood and suffering), the anxiety functions to suppress subsequent aggressive tendencies (Aron- freed, 1968). Further, it is expected that the adolescents' perceptions of or attitudes toward scenes with and without consequences will vary. The social norm would contend, in general, that physical aggression is not an acceptable means of solving problems. Also, given that most TV violence is relatively free of blood and suffer— ing, presenting consequences should result in differ- ences on the dimensions of perceived violence and acceptability. The following is predicted: 8 H3: Violence with consequences will be: a. liked less b. seen as less acceptable, and c. seen as more violent than violence without consequences. H“: Violence without consequences will be: a. liked more b. seen as less acceptable, and c. seen as more violent than nonviolence. The rationale for predicting more liking of the no consequences violence than the nonviolence, even though it is seen as less acceptable and more violent is based on an action—involvement conception. Given that the sequences the boys were to watch are relatively short, the child may not be highly involved in the film story. When this kind of involvement is lacking, the principal means of holding the viewer's attention is through high action. Thus, since the violent sequences are somewhat higher in action, they will hold the boys' attention better, and they will thus indicate that they liked watching them more than the nonviolence. The implications of presenting or not presenting physical consequences of violence should be considered. (If television entertainment presents only glorified or / "clean" violence, the effect may be to increase the P probability of stimulating aggressive behavior. It has been suggested that control of aggression might better 9 be implemented by showing the consequences of violencer} (Goranson, 1969b). Perhaps more important, however, is consideration of the effect of presenting consequences over time. It could be argued that to do so would eventually result in satiation such that the end result ivould be to raise the threshold of what is perceived as 'violent and thus anxiety arousing (Wolpe, 1958). This ssatiation notion has interesting parallels in the re— seearch involving social class differences in perceptions of“ television violence. Efigggial Class Differences The earlier proposition that TV violence is the .DI?jgmary source of exposure to aggressive behavior must b6? qualified for the low-income or ghetto dweller. Rfissearch indicates that these individuals are exposed tCD actual physical aggression to a greater extent than al?ee middle— or upper—income individuals (Gans, 1962; <3l£ark, 1965; Chilman, 1965; Moles, 1965; U. S. Govt., 319658). CAt the same time, the low—income child spends mCU?e time with TV than does his middle-class counterpart (ereenberg & Dervin, 1970). Thus, the lower-class child is eexposed to more real-life as well as vicarious vio— leruge in his environment. The effect that the low-income child's environ— lO ment has on his perceptions of the world provides the parallels to the satiation question. The basic notion posits that an individual's environment, a contributor to the process of socialization, functions to influence his behaviors and perceptions. Socialization here is considered to be the process whereby the child acquires ‘the norms and values of his immediate society, culture, (Jr reference group (Elkin, 1960). Recent studies of children's perceptions of TV vniolence reveal social class differences. Low—income bcxys, 10-11 years of age, perceive scenes of TV violence 518 less violent, the behavior more acceptable, they like VVEltSChing them more, and see them as more real than do rrliddle-class boys (Greenberg & Gordon, 1972a). By age 9114 , the perception difference for amount of violence <3i_ssappears though the other differences hold (Greenberg & (Eordon, 1972b). Given these perception differences, tklei crucial remaining question is, what are the overt aaggressive tendencies? This research tested the follow- irug hypotheses concerning class differences and the intueraction of class with the content manipulation of physsical consequences: H : \Boys who view violence with consequences 5 {Will exhibit less aggression than boys who view nonviolence; this treatment effect will be greater for middle-class boys than for lower—class boys. 11 H6: Boys who View violence without consequences will exhibit more aggression than boys who view nonviolence; this treatment effect will be greater for lower—class boys than for middle—class boys. The class differences predicted here are con- sistent with the evidence cited. However, these hypotheses run directly counter to p2§t_hgg hypotheses Iaut forth by Feshbach and Singer (1971) following a ssix week field experiment. Their findings focused on tide role of the child's socioeconomic level and his fkintasy aggression level. Their field experiment tried to control televi- ESian entertainment programming to groups of middle— Eirid.lower—class boys in boys' homes, residential private So u <0 Azoa a emaev q a 2 2pmflxe< .mma u <2 pomeem seasoweoucH u \ Azoa w amfisv q a m .ww< 2mmpcmm u eoz u >2 mmmaoueozoq u A moocmzdomcoo oz n 02 mmmmmm «<2 <22\02 o "swam mocmH0H> .emm .2 aaq aam\02 o "swam «<4 <<2\>2 o2 ”0HH2 spfiaflpmpamooa .emm .m aaa <<2\02 o ”swam aaa <<2\>2 02 "MHH2 mcflxflq emmmmcoem .N «am <<2\02 o H OH: <22 <2 02 ” Ham .wma Hmeoapmspflm .H "muoax:< coemmopww< 22 2S\02 0 Home aem <22\>2 02 "oem moemH0H> .emm .2 22 2H\02 0 ”9mm <22 <22\>2 02 ”ps2 spaflfinmpamooa .emm .m 22 e \02 o ”emu «ma 2 02 ”mam mefixflq emmmmeoem .m >2 02 Naom+<2q “m2 .mm< HmeoapmSCHm .H ”coammopww< 2mmpcmm mucoH0H> .Lom .: >paaanmpamoo< .emm .m mcfixfiq commemopm .m a 2\>2 0 ”mm a 2\>2 02 ”22 .mma HHQOHpmspHm .H uwmmHO Hmfioom 02 0 ”0mm >2 02 ”022 moemHoa> .emm .2 02 0 ”pm: >2 02 ”2:2 >pfififlnmpamoo< .emm .m 02 o Heme >2 oz ”mam mcfixfiq emmmmeoem .m >2 0 ”H2 >2 02 "m2 .mwa Hmeoapmspflm .H ”mucoEpmopB >9 moocosvomcoo oz ooCma0fi>soz mosoHOH>coz moocozvomcoo moocozvomcoo \moocozwomcoo oz AmoppHEo ohm mmmonp0d2: psoUQOQoUQchozv WMmmmeomwm mo wm¢222m H mqm<8 CHAPTER II METHODS The Sample Data were collected from 286 seventh, eighth, and ninth grade boys on June 13 and 1A, 1971. There were 283 usable questionnaires. Blacks were excluded to control for potential racial differences. Given the emphasis on aggressive behavior, girls were ex— cluded to control for sex differences. The boys were samples from two junior high schools in Jackson, Michigan. One school was in a low—income neighborhood in the inner city. The second was a middle-class school in a residential area. Variables and Operationalizations Television Treatment Three video tapes of approximately four minutes and thirty seconds each were constructed. The physical consequences of violence were controlled. One version, nonviolence, contained no aggressive behavior. The other two versions depicted a fist fight, with the physical consequences of the fight and without those 20 21 consequences. The basic violent scene was taken from the film, The Chase. In this scene, the sheriff of a small town is being severely beaten by three local men. The physical consequences evident included excessive beating, blood, swollen face, uncoordinated movement, and slurred speech. In the consequences version, the scene was presented with very little editing. For the no con- sequences version, the scene was recorded on video— tape using a filter to remove the reds, thus wiping out the blood. Other consequences were edited out. The nonviolent version used scenes from the film which included the major characters of the fight scene. The three versions were edited to maintain a consistent story line. This story line was fostered to give the action in each version a consistent context. Antecedent Variables Social Class Social class was determined primarily on the basis of parental occupation. A combination of the Troldahl Occupational Prestige Scale (Troldah, 1967) and the North-Hatt Occupational Ratings (1947) was used as aids in the assignment of the occupational 22 information provided by the child to either a middle- or lower—class categorization. Where occupational information was uncodable (9 percent of the cases), the social class of the school being sampled was used for classification. Since the two junior high schools sampled were selected because their student bodies most typified the middle—class and lower-class subjects de- sired, this procedure was adopted. Fantasy Aggression The adolescents' fantasy aggression level was assessed through a Thematic Aperception Test procedure (TAT). Each child was asked to write a short story in response to each of three stimulus pictures (see Figure l). Fig. l. TAT Stimulus Pictures The degree of aggressiveness evidenced in each story was rated using a modified version of Barclay's (1970) aggressive scoring procedure. Each story was given a 23 score ranging from one (n9_expression of aggression) to four (expressed strong aggression). Two independent judges rated all stories. Percent of overall agreement to a point estimate was 8A percent. Calculation of agreement allowing for differences of plus or minus one scale unit yielded an agreement figure of 93 percent. When independent ratings were completed and differences in ratings were found,the judges discussed the ratings to decide upon a final assigned score. For the analysis, the fantasy aggression story scores were intercorrelated and the two stories with the highest overall intercorrelation (.25) were summed as an index of fantasy aggression for each adolescent. Scores ranged from 2—8. The intercorrelations of the three stories were as follows: 1 and 2 = .1“; l and 3 = —.002; 2 and 3 = .25. .Three percent of the stories (17 of 533) were either uncodable or not completed. These stories were assigned the grand mean for that story as a missing data score. The scores ranged from 2-8 with a mean of 5.1 and a standard deviation of 1.A5. Aggression Anxiety A combination of eight items selected from the Sears (1961) Aggression Anxiety Scales, items devised by Buss and Durkee (1957), and specially constructed items were used to assess the aggression anxiety level rr. (“8 on iv p . .\ 2A of the boys. These eight items were submitted to a principle axis factor analysis with varimax rotation in an attempt to assess the best items to combine as indices. However, since only two of the eight items loaded to— gether in the factor analysis, the six items designed to tap anxiety which evidenced positive intercorrela— tions were summed as the index. These items and their intercorrelations are presented in Table 2. Two percent of the items (18 of 1698) were either uncodable or un— answered. These items were assigned the grand mean for that item as a missing data score. The scores ranged from 6-18 with a mean of 11.1 and a standard deviation of 2.5 These aggression anxiety items were designed to elicit general reactive style as opposed to an assessment of immediate feelings. However, since these items were assessed following the television treatment, a check was made to determine whether the treatment was having a significant influence on responses to the items. The mean aggression anxiety ratings were: (1) after nonviolence 11.03; (2) after violence without con- sequences 11.01; and (3) after violence with con- sequences 11.22. The E-test comparisons of mean values showed: nonviolence vs. violence without con- sequences t = 0.06, df = 191. For nonviolence vs. consequences t = 0.A9, df = 175. Thus, there was no 25 apparent bias due to the treatment. TABLE AGGRESSION ANXIETY ITEMS 2 AND INTERCORRELATIONS Intercorrelations 2 3 4 5 I usually don't hit back, even if someone hits me first. I can't think of a good reason for ever hitting anyone. It makes me nervous to see two of my friends fighting. There's too much blood and gore in the news these days. Fighting is not a good way to get even with someone you hate. I'd rather not have my own way if I have to punch a guy to get it. .25 .11 .13 .21 .20 .16 .18 .15 .03 .18 .10 .15 .14 .17 .30 26 Dependent Variables Aggressiveness The adolescents' aggressive tendencies follow- ing the television treatment were assessed through a set of six situation items which presented potentially aggression provoking encounters. A list of behavioral alternatives was offered for each situation. The boys were asked to imagine themselves in that situation and to honestly indicate what they would do. A complete listing of the items and the item intercorrelations are in Table 3. TABLE 3 SITUATIONAL AGGRESSION ITEMS AND INTERCORRELATIONS 1. Since early in the school year you have been bugged by a bully who really thinks he's tough. This guy has not only lipped—off to you but has given your friends a bad time too. Imagine that both of you are on the football team. On one particular play during practice you have a wide open shot at him from the side. How hard do you hit him? I would try to hit him hard enough to: stop the play knock the wind out of him put him out of practice for the day put him out of practice for a week wrack him up for a good long time put him in the hospital for a long time 27 TABLE 3 (Continued) Your friend John is in a bind and needs some help to move some things this Saturday. You would like to help him but you can't spare much time because you have to help your dad fix the garage roof. How much time would you help John? I would help him: about one hour 2—3 hours 4—5 hours 6-7 hours more than 7 hours Your class has gone to a larger classroom to watch a movie. In the desk drawer where you're sitting there are rubberbands, paper, and a box of straight- pins. Although the room is dark, you notice that the guy who tripped you in the hall yesterday is sitting a couple of seats ahead of you. You really don't like him very much and would like to get even with him for tripping you. What would you do? I honestly would: do nothing throw paper balls to bother him shoot him with some rubber bands shoot him with some spit wads bend a straightpin and shoot him with it The guys are playing a game of baseball over the noon hour. Imagine that Bill, your friend, is pitch- ing and a guy he doesn't like comes up to bat. This guy has been giving Bill's girlfriend a bad time. Yesterday he pulled her books out of her hands which almost knocked her down the stairs. He is a real creep. How should Bill pitch to him? Bill should: pitch as usual try very hard to strike him out throw the ball fast and close to scare him try to hit him with the ball throw the ball fast and hard to hit him in the head 28 TABLE 3 (Continued) While dressing in the locker room, the local joker snaps you hard with a towel. You really don't like him and none of the other guys do either. When they see him do it, they grab him and hold him so that you can snap him back. It really hurt when he snapped you. What would you do? I honestly would: tell them to let him go snap him once snap him two or three times snap him four or five times snap him as hard as I could until he cried A student in your class has become very ill and needs special treatment. His family doesn't have much money so your class voted to give them some of the money that was earned in the car wash last Saturday. 0f the $50 that was earned, how much should the class give? The class should give: less than $10 $20 ”$30 $40 $50 While playing volleyball, your friend Jim was lined up near the net waiting for the game to get started. Before the game starts and without good reason, the guy across from him spikes the ball hard and hits Jim in the face. Jim dislikes the guy anyway; what should he do? Jim honestly should: do nothing and ignore it tell him to watch what he is doing grab the ball and throw it back at him charge under the net and slug him one charge under the net and beat him up wait to catch him alone after school and really beat him up good 29 TABLE 3 (Continued) 8. You have just bought a new transistor radio. On your way home a guy you don't like asks to look at it. He takes off running and will not give the radio back to you. What would you try to do? I would really try to: ignore him, hoping he would give it back later keep asking him for it until he gave it to me catch him and take it back even if I had to shove him around catch him, take the radio and slug him a couple of times catch him, take the radio and punch him five or six times catch him, take the radio and beat his head in good Intercorrelations Item # 1 3 4 5 7 8 1 _ 3 31 - 4 .33 .38 - 5 .33 .42 .40 — 7 .29 .39 .39 .43 — 8 .34 .37 .36 .50 .44 — 30 The sum of the six items was used to index the aggressive tendencies of each youth. The scale range was 6-33. Only four of the 1698 total items were blank or uncodable. Each of these items was recoded to the grand mean for that item. The scores ranged from 6-33 with a mean of 19.6 and a standard deviation of 6.5. Perceptions The boy's perceptions of the scenes were measure using items from two recent studies by Greenberg and Gordon (1972a, 1972b). Ten items were used to tap the dimensions of (l) perceived violence, (2) perceived acceptability of the behavior observed, and (3) pro- fessed liking for the content of the scene. The perception items were submitted to a gxrinciple axis factor analysis with varimax rotation tc> verify that they were tapping the intended dimensions. At: the three factor solution, all items were clustered ajs expected and explained 66 percent of the total vaitiance. Table 4 presents the items and factor load- ings. The scale range and response range of the four items for Perceived Violence was 4—16, with a mean of 11-3 and a standard deviation of 3.3. The three items for Perceived Acceptability had a scale and response range of 3-12 with a mean of 9.3 and a standard 31 TABLE 4 FACTOR ANALYSIS OF PERCEPTION ITEMS % Total Variance Factor 1. Perceived Violence ................. 25% Item ITEMS: Loadings Were the people ....... Not Very Angry A Little Angry ... .75 Very Angry Extremely Angry Was what you saw ...... Not Very Violent Pretty Violent Very Violent ... .78 Extremely Violent Was what you saw ...... Not Very Serious A Little Serious Pretty Serious ... .70 Very Serious ‘Was what you saw ...... Not Very Cruel A Little Cruel Pretty Cruel ... .74 Very Cruel Fkictor 2. Perceived Acceptability .............. 20% Item ITENB: Loadings Is it ...... Very Right For People To Be This Way A Little Right Not Very Right ... .83 Not Right At All For People To Be This Way Was What you saw ..... A Very Good Thing to Do A Pretty Good Thing A Pretty Bad Thing ... .75 A Very Bad Thing To Do 32 TABLE 4 (Continued) Is it ...... Very Nice For People To Act _—_Like This Pretty Nice _——Not Very Nice ... .60 ———Not Nice At All For People To —_—Act Like This Factor 3. Professed Liking ........................ 21% ITEMS: Item Loadings What you saw was ...... A Very Good Thing To Watch A Pretty Good Thing A Pretty Bad Thing ... .73 A Very Bad Thing To Watch Was it ................ A Wonderful Show A Pretty Good Show A Pretty Bad Show ... .87 A Terrible Show Vkis what you saw a show Llike .................. ___You Really Like to See ___You Sometimes Like To See ___You Don't Like To See Very Much ... .80 _*_You Don't Like To See At All ’1 '1 \ ’1 . ,ing 15 13ercent of the total variance. These four items 33 deviation of 2.3. The three items for Professed Liking had a scale and response range of 3-12 with a mean of 7.1 and a standard deviation of 2.1. In all three cases, the items were summed as the index for those dimensions of perception. Only nine items in total were blank or uncodable, these were recoded to the grand mean of the item. Control Variables Information was collected on three variables which have shown a positive relationship to aggressive behavior in prior research. The intent was to account for their influence, should a treatment effect be demon— strated. OveITt Aggressiveness The youth's normal aggressiveness was assessed thrmbugh two independent measures. The first, was a set .\of exight items designed to tap general overt aggression ,tendenicies (Buss & Durkee, 1957). These items were I factcn? analyzed (principle axis with varimax rotation) Vfilfii‘the eight aggression anxiety items to determine the'best;items for index construction. Four overt \ aggressxion items held together as one factor explain— were SLnnmed as index #1 of overt aggression. The scale 34 range and response range was 4—12 with a mean of 7.7 and a standard deviation of 2.1. The 19 missing data responses were recoded to the grand mean of the item. The items used in this index, factor loadings and item intercorrelations are in Table 5. These overt aggression items were designed to elicit general overt reactive style as opposed to an assessment of immediate feelings. However, since these items were administered following the TV treatment, the same t-test procedure used for the aggression anxiety items was used to test for a treatment effect. The mean overt aggression ratings were: (1) after non- violence 7.86; (2) after violence without consequences 7.60; and (3) after violence with consequences 7.79. The t-test comparisons of mean values showed: Non- violence vs. violence without consequences t = 0.86, df 191. For nonviolence vs. consequences t = 0.49, di'= 175. Thus, there was no apparent effect due to 'the treatment. The second measure was supplied by the boys' Iphysical education instructor. The instructor rated teach boy on a five point scale ranging from "one of ‘the least aggressive in the class" to "one of the most aggressive in the class." The actual rating served as overt aggression index #2 for each child. Scores 35 TABLE 5 OVERT AGGRESSION ITEMS AND INTERCORRELATIONS Intercorrelations Factor 1 2 3 4 Loadings 1. I think a fight is a good way to settle an argument. - .61 2. Sometimes I get the urge to pound someone just for fun. .24 - .65 3. Anyone who says bad things about me is asking for a punch in the nose. .50 .36 — .75 4. Sometimes you have to get tough with other kids to get your way. .28 .30 .36 — .66 Total Variance 15% 36 ranged from 1—5 with a mean of 2.8 and a standard devia— tion of 1.0. This measure of overt aggressiveness cor- related .15 with the overt aggression index #1. The instructor rating produced a .12 correlation with the major dependent measure of aggressiveness, the situation items. Index #1 showed a .52 correlation with the situation items and was thus used as the measure of overt aggression for the analyses. Parental Attitudes The attitudes of the youths parents toward aggression were assessed through eight items selected from Sears (1961), fimml Dominick zuuiGreenberg (1972) and from specially constructed items. The items and intercorrelations are presented in Table 6. Aggressive items scored 1—3 with 3 being the score for a positive attitude toward aggressive behavior. Thus, the index for this variable was the sum of the eight items, with a.sca1e range of 8—24. The four missing data items were :recoded to the grand mean of the item. The scores :ranged from 11-22 with a mean of 16.1 and a standard (deviation of 2.2. 37 TABLE 6 ITEMS FOR PARENTAL ATTITUDES TOWARD AGGRESSION AND INTERCORRELATIONS 1. "My parents think hobbies are a good thing." (Check only one) My parents think that way I'm not sure what they'd say My parents don't think that way 2. "If I was rude to someone, my parents would want me to apologize." 3. "My parents think that a guy is a sissy if he's afraid to play rough games." 4. "My parents think it's important to finish school." 5. "When killing is shown on TV news, my parents always dislike it." 6. "If I happen to get into a fight now and then, my parents don't really mind." 7. "My parents think that pets are good for children." 8. "My parents say that campus protesters deserve to have their heads beat in.” 9. "My parents think that most children should try to play a musical instrument." 10. "My parents think it's perfectly ok for boys to fight sometimes." 11. "My parents think it's ok for a policeman to shoot someone who is robbing a store." 12. "My parents will punish me if I get into fights." 13. "My parents think it is wise for children to learn to swim." 38 TABLE 6 (Continued) Intercorrelations 5 6 8 10 11 12 2 _ 3 .08 5 .06 .06 —— 6 .10 .14 -.01 -— 8 .08 .11 .14 .11 -- 10 .02 .01 .07 .30 .30 —— 11 .04 .23 .01 .15 .22 .03 -- 12 .06 .08 —.001 -.25 .11 —.25 04 — 39 Exposure to TV Violence The youth's general exposure to television violence was determined through a checklist of 24 violent and nonviolent TV shows offered in the Jackson area. The boys were asked to check the programs they watched every week or almost every week. Seventeen of the shows were classified as violent on the basis of ratings provided by Greenberg and Gordon (1971). Two new shows, "Men at Law" and "Strange Report” were also classified as violent because of their similarity to "Perry Mason" and "The FBI" respectively. The scale range was 0—19. These scores were used as the index of exposure to TV violence. Scores ranged from 0—19 with a mean of 7.7 and a standard deviation of 3.8. The total listing of TV shows is in Appendix G. Basic Study Design The basic study design involved two levels of the classification variables—-(l) Social Class, (2) Fan- tasy Aggression, and (3) Aggression Anxiety. The classification variable was compared across three ‘treatment conditions, (1) violence with physical con- sequences, (2) nonviolence, and (3) violence without consequences. The dependent measures were (1) aggres- siveness, (2) perceived violence, (3) perceived 40 acceptability, and (4) professed liking. Control variables were (1) overt aggression level, (2) parental attitudes toward aggression, and (3) amount of exposure to TV violence. The study design is summarized below (p. 41). General Procedures Each school provided two viewing rooms for the experiment. Each room was equipped with two video monitors and a video tape machine. Viewing in each room was done in two groups of 7—10 boys. In both schools, the boys were randomly assigned to a treatment condition and escorted to the viewing rooms during what was normally the physical education class period. The entire testing procedure took one class session of 50 minutes. Upon entering the testing room the boys were given a general introduction to the study which posed that "We are interested in what people your age think about a lot of different things. Some of the things ‘we are interested in will be presented to you on questionnaires, others will be on television." 'Throughout the testing procedure and within the questionnaire design, the attempt was to reduce the emphasis on the aggressive focus of the study. Before 41 emewea 2e3em .m .mem _ a oASmomxm _ wcflxflq commemoem . muofixcm ooQoH0H> >8 _ coammopwmm _2pHHHomemoo< Uo>flooeom_ mopzpflpp< commmopwwm Hmpcopmm " ooQoHOH> Uo>aoohom " 2mmpcmm Coemmopww< _ mmoco>fimmopww< _ um>m mmmao peo>o _ _ Hmfloom _ p moocodvomcoo mocoH0H>Qoz moocosvomcoo 0\3 oocmH0H> Suez mosoH0H> moHQmHmm> moanmfihm> Hoppcoo COHpGOfiMHmmmHo 42 starting, the boys were reminded that their responses were private and would not be shown to their principal, teachers, or parents. They were also requested not to sign their names. The first instrument administered was the Thematic Aperception Test (TAT), introduced as "we want to see if you can make up stories for the three pictures in the booklet." The boys completed three short stories before introduction of the next instrument. The parental attitudes instrument followed the TAT. For masking, items such as "My parents think it's important to finish school," were interspersed among the eight aggression items. The television treatment was presented follow— ing completion of the parental attitudes measure. The order of presentation for the treatments was system- atically varied in each room to control for time of day. Following the television exposure, a set of eight situational aggression items were administered to assess the effect of the television treatment on the boys' aggressive tendencies. As the second major dependent measure, the ten perception items followed the situa— tional items. These were followed by the overt aggression and aggression anxiety items. Finally, the mass media exposure items were presented. 43 Some means of identifying questionnaires was necessary for correlation with the instructor's ratings of each child's aggressiveness. When all materials were completed, a blank sheet of paper was passed out and each boy was asked to . . ."put your name and the grade you are in on the paper. We want to know who has helped us out today." They then left all materials in their seats. After the boys left, the materials for each boy were stapled together. Before leaving the room, the boys were asked not to talk about what they had seen or done. Analytic Procedures The initial analysis involved determining the necessity for the control variables. A multiple regres- sion routine provided this information for each dependent measure. Running all variables with each dependent variable demonstrated which variables were having the greatest influence, i.e., were the best predictors of, the dependent measure. As a final defining criterion, a partial correlation of .30 or better with the dependent variable was considered sub- stantial enough to warrant control. Where substantial correlations existed and the hypotheses called for 2 x 2 analyses of variance, the con- trol variable was included in the basis analysis, making 44 it a 2 x 2 x 2 design. Thus, if the variable in question was having a significant effect on the dependent variable, it would be detected. Since this procedure resulted in disproportionate cell frequencies,an analysis of variance procedure was used which corrected for this problem through a least squares procedure. Hypotheses predicting group differences which did not require an analysis of variance procedure were analyzed by t—test. CHAPTER III RESULTS The results will be presented first for the overall television treatment effects. Secondly, the evidence will be presented for each dependent variable: (1) Situational aggressiveness, (2) perceived violence, (3) perceived acceptability, and (4) professed liking. Under each dependent variable, findings will be presented for the classification variables: (A) Social class, (B) fantasy aggression, and (C) aggression anxiety. Television Treatment Differences Mean differences between treatments were examined by t-test to determine whether overall dif- ferences existed on the dependent measures. The differences are presented by dependent variable. Situational Aggressiveness It was hypothesized that: H : Boys exposed to violence with consequences 1 would exhibit more aggression than boys exposed to nonviolence. H : Boys exposed to violence without conse— quences would exhibit more aggression than boys exposed to nonviolence. 45 46 The overall mean differences for situational aggressiveness across the hypothesized treatment com- parisons showed no differences. On this gross comparison, then, this dependent variable did not detect differential effects of the video manipulation on the adolescents' aggressive tendencies. The summary data for the t—test comparisons on situational aggressiveness are presented in Table 7. Perception Differences Given the manipulation of physical consequences in the TV content, differences were expected to be evident in the adolescents' perceptions of the scenes. These hypothesized perception differences are presented below. The findings for each dimension will be dis- cussed separately. H3: Violence with_consequences would be: a. liked less b. seen as less acceptable, and c. seen as more violent than violence without consequences. H“: Violence without consequences would be: a. liked more b. seen as less acceptable, and c. seen as more violent than nonviolence. Professed Liking.-—It was hypothesized that the violence with consequences would be liked less than the violence without consequences (H3a)° Although the overall 47 mean difference between these conditions was in that direction, the effect did not approach significant (t = 0.72, df = 175). The liking comparison for violence without con- sequences and nonviolence (H4a) was marginally signifi— cant in the predicted direction at the .10 level of significance (t = 1.77, df = 191). The summary data for this comparison are presented in Table 7. Perceived Acceptability.—-In H3b’ it was pre- dicted that the violence wigh consequences would be seen as less acceptable than the violence without con— sequences due to the added blood and suffering. The mean difference between these conditions was in the predicted direction but did not reach significance (t = 1.08, df = 194). The comparison of the violence without conse- quences and the nonviolence on perceived acceptability (H4b) was significant at the .001 level in the predicted direction (t = 5.27, df = 191). Thus, the behavior in the nonviolent version was seen as more acceptable than the aggressive behavior in the no consequences violence. Again, the t-test data for these comparisons are in Table 7. Perceived Violence.-—It was expected that the presentation of physical consequences in violence would 48 make the action of the scene appear even more violent (H ). The mean comparison between the violence with 3c consequences and that without verified this expectation. The consequences version was seen as significantly more violent than the no consequences version (p<:.01, t = 3.03, df = 194). The comparison of the nonviolence and the violence without consequences (H4c) further substantiated that the scenes, as constructed, differed significantly in degre of apparent violence (t = 12.50, df = 191, p HHm mom* So. Va So. Va 3. Va .92 1 a doe u we Hod u we Hma u we Hma H mm om.mA u e em.m u e AA.A u p m.o H e >2 1 0 >2 1 e >2 1 a: o 1 m o O\3 > A 32V QO\2 > A :22 o 0\z > A 2V >2 1 o 0\z > A 2v Ho. uva .m.n O Q .m.c O Q . . nae u as ema n he 20H n he m e u a mo.m u p wo.a u p mm.o n p mmsm m mm m 0\3 cm W 0\3 pm 0 0\3 mm 1 H > O\> A 2V > o\> A 2v > . o\> A 2V >2 1 0\> A 2v Ao 0\3 >V m.m o.m H.m m.m .U.m moocosvmmcoo AeoAv Aeoav Aeoev Aeoev e escapee mm.mH mm.m mH.m mm.mH M mocoH0H> m.a H.m o.m m.m .m.m Aemv Apmv Aemv Anmv e A>2v Ho.m mm.m mm.~ om.mH m oo:oaoe>coz A0\>V m.m m.m m.m 0.0 .c.m moocoswomcoo Aomv Aomv Aomv Aomv c Suez mm.mH mm.: mm.> mm.ma M ooQOH0H> Aeaue Amenm Ameum Ammum ”emcee mamomv "owcmm mamom~ ”omcmm mamomv ”emcee mamomv mocmfiow> muflaeompmoco< mcfixflq mmocm>fimmmemw< Uo>floomom Uo>Hooeom commoeomm HmQOHpmszm *mmqm¢Hm<> Bzszmmma Mm meommmm BZMEBmQMB A4¢mm>o w mqmde 50 Classification Differences byiDependent Measure To examine appropriately the hypothesized differences for the classification variables, it was necessary to determine which of the potential control variables should serve as "controls." A multiple regression routine was used to make this judgment. All variables were run against each dependent measure to determine which variables were having the greatest in— fluence on each dependent variable. Those variables having high partial correlations with dependent measures were deemed worthy of control. A partial correlation value of .30 or greater with the dependent measure was established as an arbitrary criterion for control. This would indicate that the variable in question could be contributing as much as 9% of the variance displayed by the dependent variable. Most of the hypotheses involving classification variables originally called for 2 x 2 analyses of variance. Where significant partial correlations were evident, however, the variable in question was divided by a median split procedure and the basic design was expanded to make it a 2 x 2 x 2. Thus, the influence of the "control" variable on the dependent measure could be determined. "Control" here is not used 51 in the sense of holding the influence of the variable constant. Rather, the variable is allowed to interact and its contribution to the interaction is examined. Dependent variables which showed high partial correla- tions with other dependent variables were handled in the same way. Table 8 presents the simple and partial correla- tions of all variables with the dependent variables. By dependent variable, the variables showing a .30 or better correlation were: 1. Situational aggressiveness-- Overt aggressiveness 2. Professed liking-- Perceived acceptability 3. Perceived acceptability-— Professed liking and perceived violence 4. Perceived violence—— Perceived acceptability The variables with partial correlations of .30 or more with dependent measures are starred in Table 8. Because the "control" variable was selected for its high correlation with the dependent measure, it could be expected that a median split on that variable would produce a significant main effect. This was the case for all analyses to be reported. In all cases, the "control" variable demonstrated at least a .001 level of significance. Thus, the main effect 52 Ho.1 No.1 mo.| om.1 no.1 >H.1 mo. 2m. ooCoH0H> >8 on oLSmooxm mo. mo. Ao.1 2A. AA.1 om. mA.1 mm.- Aemexea scammosmmd :oo. ma.1 oH.1 mm.1 oH.1 mm.1 *o:. mm. mmoco>fimmmpm 1w< peo>o II 11 +mm. um. om.l mo. mo. mo.l ooCmHOH> oo>Hooeom *mm. em. 1- 1. *mm. mm. 00.1 om.n Apeaeeeeaee 10¢ oo>Hooeom oa.1 mo. *mm. mm. 11 11 No.1 mm.1 mcfixfiq powwowopm mo. mo.| oo.1 mm.1 No.1 mw.1 11 1| mmoco>fimmommw< Heeoeemseem mo. mo.1 mH.I ma.1 No.1 mH.1 mo. ma. mOUSqu Ip< fiducmpmm Ho.l mo. mo. Ho.l 50.1 mo.1 Ho. :0. QOfimmmkm Iw< hmmpcmm Ho. mo. no. NH. mo.l mo. No.1 mH.I mmmao amaoom moanmflem> HH< m.Ae map m.Ae mes m.Ae mAs m.Ae mAe cocoH0H> mpfiaflompmooo< mcflxaq mmoco>ammopwm< ”moaomflmm> oo>Hoomom Hoo>floomom Hommmomomm Hmcoflpmzpfim pcoocoooa mmqm Bzmozmmma 39H: mmqm mo monEHpmmoc mm emoaom Hafiz moaomfism> omonp noaomp wasp mom oopooamoh p0: ohms moeoom mocfim .mHOpomm mozpo one Co ommm>oe Ca oomoom meme zpflaflompmooo< Uo>floohom cam mqflxflq pommoAOLmH .QOHpmcmHon moo pxop mow moaomwhm> :Hoepcoo: mm oom: poz+ .oaomamm> :Hoepcoo: mm oomox Aomzcflpcoov m mqmde 54 difference on the "control" variable will be reported in the analysis of variance tables but will not be discussed as a significant main effect in itself. The "control" variable becomes important when it is involved in a significant interaction. Its presence is par- ticularly important when the analysis reveals a third order interaction and no other interaction effects are evident. In such a case, the two way analyses of variance would have indicated no treatment effects where effects were present but taking a more complex form. Median splits were used in several analyses to examine differences within a classification variable dimension, as well as to split the "control" variable in the three-way analyses of variance. In all cases, the split was calculated using the entire sample pool. Relevant information about these dimensions is presented in Table 9. In the single case where perceived acceptability was the dependent measure this procedure of generating median splits from the entire sample produced a problem. When acceptability is the dependent measure, as can be noted in Table 8, the variables of professed liking and perceived violence are significant correlates. However, since the amount of perceived violence in the 55 TABLE 9 MEDIAN VALUES Median Scale* Median Split for Variable Range Value Analyses Fantasy Aggression 2—8 5.69 5 = low (n=l60) 6 = high (n=l23) Aggression Anxiety 6-18 11.46 10 = low (n=l23) 11 = high (n=160) Overt Aggression 4-12 8.19 7 = low (n-l33) 8 = high (n=150 Perceived Acceptability 3—12 4.94 5 = low (n=l39) 6 = high (n=l44) Professed Liking 3—12 7.69 8 = low (n=l25) 9 = high (n=158) 8 In all cases, the response range was equal to the scale range. For all variables, the higher the score, the more the attribute, e.g., more fantasy aggression, more liking, etc. 56 nonviolent sequence was very low, the high segment of the overall median split produced exceptionally small cell frequences for the nonviolent-~high perceived violence cells. Thus, only the variable of professed liking was used as a "control" factor when perceived acceptability was the dependent variable. Although the analyses to be presented examined primarily classification variable differences, the results will be presented by dependent variable. Clas- sification variables will be discussed under each dependent variable. Situational Aggressiveness Social Class Differences.--It was theorized that social class environment would contribute to the social- ization of a child's behavioral responses. As related to the video materials constructed, the following was predicted: H5: Boys who View violence with consequences will exhibit less aggression than boys who View nonviolence; this treatment effect will be greater for middle-class boys than for lower—class boys. H6: Boys who View violence without consequences will exhibit more aggression than boys who View nonviolence; this treatment effect will be greater for lower—class boys than for middle-class boys. The relevant'bontrol" variable here was the boys' 57 normal overt aggression level. This variable was in- cluded with the social class and TV treatment factors to make a 2 x 2 x 2 design. A three—way analysis of variance permitting unequal cell frequences (Snedecor, 1956) was used to test H5 and H6' Tables 10A and B present the means and sources of variance for this analysis. Analysis of variance Table 10B for the con- sequences-—nonviolence comparison showed that the only effect, other than the "control" variable, was a marginal two—way interaction between class and the "control" variable of overt aggression (p<:.10). In general, the pattern of the means for this interaction showed that boys high in overt aggression demonstrated more aggressiveness than boys who were normally low in overt aggression and that this was even more the case for lower-class boys than for middle-class boys. Thus, H5 was not confirmed. For the violence without consequences-—nonviolence comparison (H6), analysis of variance Table 10B indi- cated a significant three—way interaction (p<:.03). This third order interaction effect demonstrated that when the adolescent's overt aggression level, his social class, and the TV treatment were considered to- gether, the boys' aggressive reactions to the television 58 TABLE 10 ANALYSES OF VARIANCE FOR SOCIAL CLASS (SES), TV TREATMENT, AND OVERT AGGRESSIVENESS (OA) Table 10A: Cell Means (The higher the mean, the more situational aggressiveness) Middle Class Lower Class High Low High Low Overt Overt Overt Overt Aggres. Aggres. Aggres. Aggres. With Consequences 20.5 16.5 22.7 18.6 (n=23) (n=27) (n=24) (=16) Nonviolence 20.5 16.9 24.3 14.1 (n=l8) (n=27) (n=30) (n=12) Without Consequences 22.2 17.0 21.2 17.4 (n=30) (n=30) (n=25) (n=21) Table 10B: Variance Tables WITH CONSEQUENCES VS. NONVIOLENCE Source of Variance MS df F SES 67.9 1 2 TV 17.6 1 - OA 1206.2 1 38 SES x TV 25.7 1 - SES X 0A 114.6 1 3 TV x OA 87.1 1 2. SESxTV X 0A 107.0 1 3 ERROR 31.8 169 TOTAL 176 EN0\ 0 }_J ns ns .0005 ns .10 ns ns 59 TABLE 10 (Continued) WITHOUT CONSEQUENCES VS. NONVIOLENCE Source of Variance MS df F P SES 0.5 l - nS TV 11.5 1 - ns OA 1449.2 1 39.6 .0005 SES x TV 6.8 l - ns SES X OA 74.3 1 2.0 ns TV X OA 60.8 1 1.7 ns SES x TV X 0A 176.6 1 4.8 .03 ERROR 36.6 FJH mum Mkfl TOTAL 6O violence would be more easily detected than if all these factors were not taken into account. Looking at the means (Table 10A), it appeared that the differences were consistent in that boys high in overt aggression displayed more aggression than those low in overt aggression. This pattern was exag- gerated for the lower-class boys. Also, the lower— class boys who are high in overt aggression broke the pattern by showing more aggression in the nonviolent television treatment while all other groups showed less, therefore, the third order interaction. Thus, partial support was found for H6 when the third factor of overt aggression was considered. Three of the four groupings of youngsters were more aggressive following exposure to violence as opposed to nonviolence. The exception was the lower-class boys who were high in overt aggression. Fantasy Aggression and Overt Aggression.-—The Feshbach and Singer (1971) hypothesis (Hg) was tested by t—test: H9: Boys low in fantasy aggression and high in overt aggression will exhibit less aggression after exposure to violence without consequences than after exposure to nonviolence. It was expected that H9 would not be confirmed. The mean aggression value for the low fantasy--high 61 overt aggression boys after exposure to violence without consequences was 21.1. For those exposed to nonviolence, the mean was 21.9. The t—test was not significant (t = 0.43, df = 51). Thus, the hypothesis was not supported. Aggression Anxiety Differences.--The aggression anxiety level of the adolescent was expected to influence his behavior as follows: H10: Boys who view violence with consequences will exhibit less aggression than boys who view violence without consequences; this effect will be greater for boys high in aggression anxiety than for boys low in aggression anxiety. The "control" variable of overt aggressiveness was again relevant here. The same three—way AOV pro- cedure was applied to test the television treatment differences of violence with consequences and violence without consequences as posited in H10. Table 11 presents the AOV information for this analysis. Since no interaction effect appears in Table 11, hypothesis 10 was not confirmed. There was, however, a main effect evident in the table which lends some support to the reasoning behind the interaction hypothesis proposed. This main effect difference showed that boys who were high in aggression anxiety were less aggressive on the situation items than were boys low in aggression 62 TABLE 11 ANALYSES OF VARIANCE FOR AGGRESSION ANXIETY (AA), TV TREATMENT AND OVERT AGGRESSIVENESS (OA) ‘ Table 11A: Cell Means (The higher the mean, the more situational aggressiveness) High Agg. Anx. Low Agg. Anx. High Low High Low Overt Overt Overt Overt Aggres. Aggres. Aggres. Aggres. With Consequences 21.5 16.9 21.9 18.9 (n=26) (n=27) (n=21) (n=20) Nonviolence —- —- -- —— Without Consequences 19.3 15.7 24.7 19.3 (n=30) (n=3l) (n=25) (n=l6) Table 11B: Variance Tables WITH CONSEQUENCES VS. WITHOUT CONSEQUENCES Source of Variance MS df F P AA 397.8 1 10.9 .001 TV .0 l -— ns OA 858.5 1 23.5 .0005 AA x TV 132.9 1 3 6 ns AA X 0A .1 1 -- ns TV X 0A 5.8 l -- ns AA x TV X OA 36.7 1 1.0 ns ERROR 36.6 200 TOTAL 207 63 anxiety (p<(.001) Perception Differences Professed Liking Perception differences will be presented in the same format, that is, by dependent variable with clas- sification variables following. Since no social class differences were examined concerning perceptions, the classification variables for the remaining perception items are fantasy aggression and aggression anxiety. Fantasy Aggression Differences.--The fantasy aggression level of the adolescent was expected to affect his liking for the TV content. These hypotheses were posed: H7a: Violence without consequences will be liked more than nonviolence; this treat- ment effect will be greater for boys high in fantasy aggression than for boys low in fantasy aggression. Violence with consequences will be liked less than violence without consequences; this treatment effect will be greater for boys low in fantasy aggression than for boys high in fantasy aggression. a. The "control” variable for these analyses was perceived acceptability. The three—way AOV comparisons for these hypotheses are presented in Tables 12A and B. 64 TABLE 12 ANALYSES OF VARIANCE OF FANTASY AGGRESSION (FA), TV TREATMENT, AND PERCEIVED ACCEPTABILITY (PA) Table 12A: Cell Means (The higher the mean, the more professed liking) High Fan. Agg. Low Fan. Agg. High Low High Low Per. Per. Per. Per. Accept. Accept. Accept. Accept. With Consequences 6.7 9.8 7.5 8.9 (n=23) (n=15) (n=37) (n=15) Nonviolence 7.6 8.0 6.3 8.0 (n=7) (n=3l) (n=l7) (n=32) Without Consequences 7.7 8.9 7.4 9.0 (n=29) (n=19) (n=3l) (n=27) Table 12B: Variance Tables WITHOUT CONSEQUENCES VS. NONVIOLENCE Source of Variance MS df F P FA 5.2 l 1.3 ns TV 22.8 1 5.9 .02 PA 56.6 1 14.6 .0005 FA x TV 3.4 l - ns FA x PA 5.2 l 1.4 ns TV x PA 0.9 1 - ns FA x TV x PA 2.1 l — ns ERROR 3.9 185 TOTAL 192 65 TABLE 12 (Continued) WITH CONSEQUENCES VS. WITHOUT CONSEQUENCES Source of Variance MS df F P FA 0.1 l — ns TV 0.0 l - ns PA 129.1 1 32.9 .0005 FA x TV 0.0 l - ns FA x PA 4.8 l 1.2 ns TV x PA 7.5 l 1.9 ns FA x TV x PA 9.2 l 2.3 ns ERROR 3.9 176 TOTAL 183 66 The analyses in Tables 12A and B failed to con- firm the liking interaction hypotheses in either treat- ment comparison. No interaction effects were present. The AOV table for the violence without vs. nonviolence comparison (H73) did indicate a significant main effect for the TV treatment (p<:.02). As expected, the violence without consequences was likaimore than the nonviolence. Aggression Anxiety Differences.-—The adolescents' aggression anxiety level was expected to interact with the TV treatments such that: Hll : Violence without consequences will be a liked more than nonviolence; this treatment effect will be greater for boys low in aggression anxiety than for boys high in aggression anxiety. H123: Violence with consequences will be liked less than violence without consequences; this treatment effect will be greater for boys high in aggression anxiety than for boys low in aggression anxiety. The "control" variable for the three way AOVs for these hypotheses was again perceived acceptability. Table 13A and B present the AOV results. The analysis for the treatment combination of violence without con- Sequences and nonviolence (Hlla) produced a three—way iIl‘teraction (pnfreed, J. Conduct and Conscience: The Socializa- tion of Internalized Control Over Behavior. New York: Academic Press, 1968. EBailcer, R., & Ball, S. Mass Media and Violence, a staff report to the National Commission on the Causes and Prevention of Violence. Government Printing Office, November, 1969. EBairdfiiura, A. "Influence of Models' Reinforcement Contingencies on the Acquisition of Imitative Responses." Journal of Personality and Social Psychology, 1965, 1, 589—595. EBailcczlay, A. Manual for Scoring Sexual and Aggressive Imagery, Defensiveness, and Distortion on the TAT. Psychology Department, Michigan State University, December, 1970. (Mimeographed) iBeex°Lcowitz, L. Aggression: A Social—Psychological Analysis. New York: McGraw-Hill, 1962. "The Effects of Observing Violence." Scientific American, 1964, 210, 35-41. _L, . "Some Aspects of Observed Aggression." Journal of Personality and Social Psychology, 1965, 2, 359-369. __________5 Corwin, R.; & Heironimus, M. "Film Violence and Subsequent Aggressive Tendencies." Public Opinion Quarterly, 1963, 27, 217-229. ._________, & Rawlings, E. "Effects of Film Violence on Inhibitions Against Subsequent Aggression." Journal of Abnormal and Social Psychology, 1963, 66, 405-412. Biblwi, EL "The Role of Fantasy in the Reduction of Aggression.” Unpublished Ph.D. dissertation, City University of New York, 1970. 103 104 IBrwimel, B.; Taub, B.; & Blum, B. "An Observer's Reaction to the Suffering of his Enemy." Journal of Personality and Social Psychology, 1968, 8, 384-392. 1311553, A. The Psychology of Aggression. New York: Wiley, 1961. , & Durkee, A. "Assessing Different Kinds of Hostility." Journal of Consulting Psychology, 19573 21: 3&3-3u8- C2r1j_lman, C. "Child Rearing and Family Relationship Patterns of the Very Poor." Welfare in Review, 1965) 33 9'19. (31L6111k, K. Dark Ghetto: Dilemmas of Social Power. New York: Harper & Row, 1965. IDCDIrLinick, J., & Greenberg, B. "Attitudes Toward Violence: The Interaction of TV Exposure, Family Attitudes and Social Class." In Television and Social Behavior: Television and Adolescent Aggressiveness (Vol. 3). A technical report to the Surgeon General's Scientific Advisory Committee on Television and Social Behavior. Edited by E. Rubinstein, G. Comstock, and J. Murray. Government Printing Office, 1972. IZlLCiJfl, F. The Child and Society: The Process of Socialization. New York: Random House, 1960. E‘GBShbach, S. "The Stimulating Versus Cathartic Effects of a Vicarious Aggressive Activity." Journal of Abnormal and Social Psychology, 1961, 63, 381—385. __________J & Singer, J. Television and Aggression. San Francisco: Jossey—Bass, 1971. Gans , H. The Urban Villagers: Group and Class in the Life of Italian Americans. Glencoe, 111.: Free Press, 1962. 105 (Gearbner, G. "Violence in Television Drama: Trends and Symbolic Functions.” In Television and Social Behavior: Media Content and Control (Vol. 1). A technical report to the Surgeon General's Scientific Advisory Committee on Television and Social Behavior. Edited by G. Comstock and E. Rubinstein. Government Printing Office, 1972. C3c>1ranson, R. "A Review of Recent Literature on Psycholog- ical Effects of Media Portrayals of Violence." In Mass Media and Violence. Edited by R. Baker and S. Ball. Government Printing Office, 1969, Pp. 395-413 (a). . "The Catharsis Effect: Two Opposing Views." In Mass Media and Violence. Edited by R. Baker and S. Ball. Government Printing Office, 1969, pp. 452—465 (b). (Eireeeenberg, B., & Dervin, B. Use of the Mass Media by the Urban Poor. New York: Praeger, 1970. (Exceeeanberg, B., & Gordon, T. "Social Class and Racial Differences in Children's Perceptions of Tele- vision Violence." In Television and Social Behavior: Television's Effects-—Further Explorations (Vol. 5). A technical report to the Surgeon General's Scientific Advisory Committee on Television and Social Behavior. Edited by G. Comstock, E. Rubinstein and J. Murray. Government Printing Office, 1972. "Children's Perceptions of Television Violence: A Replication." In Television and Social Be- havior: Television's Effects—~Further Explorations (Vol. 5). A technical report to the Surgeon General's Scientific Advisory Committee on Television and Social Behavior. Edited by G. Comstock, E. Rubinstein, and J. Murray. Government Printing Office, 1972. __~‘_____, "Critics' and Public Perceptions of Violence in Television Programs." Journal of Broad- casting, 15, 1, Winter, 1970-71, 29—43. Hartnuan, D. "The Influence of Symbolically Modeled Instrumental Aggression and Pain Cues on the Disinhibition of Aggressive Behavior." Un- published Ph.D. dissertation, Stanford Univer- sity, 1965. 106 ffiatt, P., & North, C. "Jobs and Occupations: A Popular Evaluation." Opinion News, September, 1947, pp- 3-15. Igezsser, G. "Application of Guttman's Scaling Method to Aggressive Fantasy in Children." Educational and Psychological Measurement, 1958, 18, 543- 551. Ddeeggargee, E. "The Role of Inhibition in the Assessment of Understanding of Violence." In The Control of Aggression and Violence. Edited by J. Singer. New York: Academic Press, 1971, pp. 125-147. DGCDILes, 0. "Child Training Practices Among Low-Income Families: An Empirical Study." Welfare in Review, 1965, 3, 12, 1—19. ESGEEiJTS, R. "The Relation of Early Socialization Exper— iences to Aggression in Middle Childhood." Journal of Abnormal Psychology, 1961, 63, 466- 492. Ekieaggel, A. "The Effects of Media Violence on Social Learning." In Mass Media and Violence. Edited by R. Baker and S. Ball. Government Printing Office, 1972, pp. 261—283. fSiiugger, J. Daydreaming, New York: Random House, 1966. "Research Applications of the Projective Techniques." In Projective Techniques in Personality Assessment. Edited by A. Rabin. New York: Springer, 1968. _______i__. "The Influence of Violence Portrayed in Television or Motion Pictures upon Overt Aggressive Behavior." In The Control of Aggression and Violence. Edited by J. Singer. New York: Academic Press, 1971, pp. 19-59. Sneekecor, G. Statistical Methods. Ames, Iowa: Iowa State University Press, 1956. Sprack, G. "Some Cognitive Deficiencies in Poorly Self-Controlled Adolescents." Paper presented at the meeting of the American Psychological Association, 1964. lx 107 Eitaub, E. "The Learning and Unlearning of Aggression: The Role of Anxiety, Empathy, Efficacy, and Prosocial Values." In The Control of Aggres— sion and Violence. Edited by J. Singer. New York: Academic Press, 1971, pp. 94—124. Eirnsldahl, V. ”Occupational Prestige Scale." Depart- ment of Communication, Michigan State Univer- sity, 1967. (Mimeographed.) [J - S. Government. Report of the National Advisory Commission on Civil Disorders. New York: Bantam, 1968. VJEijlters, R.; Park, R.; & Cane, V. "Timing of Punish— ment and the Observation of Consequences to Others as Determinants of Response Inhibition." Journal of Experimental Child Psychology, 1965, 2, 10-30. 1M63:i:ss, W. "Effects of the Mass Media of Communication." In The Handbook of Social Psychology (Vol. 5). Edited by G. Lindzey and E. Aronson. New York: Addison Westley, 1969, pp. 77-195. Mhsilzoe, J. Psychotherapy By Reciprocal Inhibition. Stanford, Calif.: Stanford University Press, 1958. APPENDIX INSTRUMENT 109 MAKING-UP STORIES This booklet will be used to see how easily you can make—up a story. Look at each picture for a few seconds and try to see what is happening. Use the page after the picture to write a story about what is happening in the picture. Use the questions on 'the sheets to guide you through the story. Feel free ‘to write anything you want to. Try to make the story liave a beginning and an ending. Since these are for fun, there are no right car wrong stories, so relax and have a good time. 111 What is happening? Who are the people? lflhat has led up to the situation? That is, what has liappened before? Dfrlat is being thought? What is wanted? By whom? VJPlat will happen? What will be done? 112 113 What is happening? Who are the people? What has led up to the situation? That is, what has happened before? What is being thought? What is wanted? By whom? Mnjat will happen? What will be done? 115 What is happening? Who are the people? What has led up to the situation? That is, what has happened before? What is thought? What is wanted? By whom? VWhat will happen? What will be done? 116 HERE ARE SOME STATEMENTS KIDS HAVE MADE ABOUT WHAT THEIR PARENTS HAVE SAID OR WHAT THEY THINK. UNDER EACH STATEMENT PUT AN 'X' ON THE LINE WHICH BEST DESCRIBES WHAT YOUR PARENTS HAVE SAID OR WHAT THEY THINK. "My parents think hobbies are a good thing." (Check only one) My parents think that way I'm not sure what they'd say My parents don't think that way "If I was rude to someone, my parents would want me to apologize." (Check only one) My parents think that way I'm not sure what they'd say My parents don't think that way "My parents think that a guy is a sissy if he's afriad to play rough games." (Check only one) My parents think that way I'm not sure what they'd say My parents don't think that way "My parents think it's important to finish school." (Check only one) My parents think that way I'm not sure what they'd say My parents don't think that way "When killing is shown on TV news, my parents always dislike it." (Check only one) My parents think that way I'm not sure what they'd say My parents don't think that way :10. lfil. 122. 117 "If I happen to get into a fight now and then, my parents don't really mind." (Check only one) My parents think that way I'm not sure what they'd say My parents don't think that way "My parents think that pets are good for children." My parents think that way I'm not sure what they'd say My parents don't think that way "My parents say that campus protesters deserve to have their heads beat in." My parents think that way I'm not sure what they'd say My parents don't think that way "My parents think that most children should try to play a musical instrument." My parents think that way I'm not sure what they'd say My parents don't think that way "My parents think it's perfectly ok for boys to fight sometimes." My parents think that way I'm not sure what they'd say My parents don't think that way "My parents think it's ok for a policeman to shoot someone who is robbing a store." My parents think that way I'm not sure what they'd say My parents don't think that way "My parents will punish me if I get into fights." My parents think that way I'm not sure what they'd say My parents don't think that way 118 13. "My parents think it is wise for children to learn to swim." My parents think that way I'm not sure what they'd say My parents don't think that way 119 BELOW ARE SOME THINGS THAT HAVE HAPPENED TO KIDS YOUR AGE. LISTED BELOW EACH ARE SOME OF THE THINGS THEY DID IN THAT SITUATION. READ EACH THING CAREFULLY AND PUT AN 'X' ON THE LINE WHICH YOU THINK IS THE BEST THING TO DO. Since early in the school year you have been bugged by a bully who really thinks he's tough. This guy has not only lipped-off to you but has given your friends a bad time too. Imagine that both of you are on the football team. On one particular play during practice you have a wide open shot at him from the side. How hard to you hit him? I would try to hit him hard enough to: stop the play knock the wind out of him put him out of practice for the day put him out of practice for a week wrack him up for a good long time put him in the hospital for a long time 1 Your friend John is in a bind and needs some help to move some things this Saturday. You would like to help him but you can't spare much time because you have to help your dad fix the garage roof. How much time would you help John? I would help him: about one hour 2—3 hours 4—5 hours 6-7 hours more than 7 hours 120 Your class has gone to a larger classroom to watch a movie. In the desk drawer where you're sitting there are rubberbands, paper, and a box of straightpins. Although the room is dark, you notice that the guy who tripped you in the hall yesterday is sitting a couple of seats ahead of you. You really don't like him very much and would like to get even with him for tripping you. What would you do? I honestly would: do nothing throw paper balls to bother him shoot him with some rubber bands shoot him with some spit wads bend a straightpin and shoot him with it. The guys are playing a game of baseball over the noon hour. Imagine that Bill, your friend, is pitch— ing and a guy he doesn't like comes up to bat. This guy has been giving Bill's girlfriend a bad time. Yesterday he pulled her books out of her hands which almost knocked her down the stairs. He is a real creep. How should Bill pitch to him? Bill should: pitch as usual try very hard to strike him out throw the ball fast and close to scare him try to hit him with the ball throw the ball fast and hard to hit him in the head 121 While dressing in the locker room, the local joker snaps you hard with a towel. You really don't like him and none of the other guys do either. When they see him do it, they grab him and hold him so that you can snap him back. It really hurt when he snapped you. What would you do? I honestly would: tell them to let him go snap him once snap him two or three times snap him four or five times snap him as hard as I could until he cried A student in your class has become very ill and needs special treatment. His family doesn't have much money so your class voted to give them some of the money that was earned in the car wash last Saturday. Of the $50 that was earned, how much should the class give? The class should give: less than $10 122 7. While playing volleyball, your friend Jim was lined up near the net waiting for the game to get started. Before the game starts and without good reason, the guy across from him spikes the ball hard and hits Jim in the face. Jim dislikes the guy anyway; what should he do? Jim honestly should: do nothing and ignore it tell him to watch what he is doing grab the ball and throw it back at him charge under the net and slug him one charge under the net and beat him up wait to catch him alone after school and really beat him up good 8. You have just bought a new transistor radio. On your way home a guy you don't like asks to look at it. He takes off running and will not give the radio back to you. What would you try to do? I would really try to: ignore him, hoping he would give it back later keep asking him for it until he gave it to me catch him and take it back even if I had to shove him around catch him, take the radio and slug him a couple of times catch him, take the radio and punch him five or six times catch him, take the radio and beat his head in good 123 FOR THE PART or THE TV snow you WATCHED, PLEASE THINK ABOUT WHAT YOU SAW. DO YOU THINK: What you saw was. . . . . A VERY GOOD THING TO WATCH A PRETTY GOOD THING A PRETTY BAD THING A VERY BAD THING TO WATCH .IS it . . . . . . . . . . VERY RIGHT FOR PEOPLE TO BE THIS WAY A LITTLE RIGHT NOT VERY RIGHT NOT RIGHT AT ALL FOR PEOPLE TO BE THIS WAY Vkis it. . . . . . . . . . A WONDERFUL SHOW A PRETTY GOOD SHOW A PRETTY BAD SHOW A TERRIBLE SHOW Werwe the people . . . . . NOT VERY ANGRY A LITTLE ANGRY VERY ANGRY EXTREMELY ANGRY Was that you saw a show like . . . . . . . . . . YOU REALLY LIKE TO SEE YOU SOMETIMES LIKE TO SEE YOU DON'T LIKE TO SEE VERY MUCH YOU DON'T LIKE TO SEE AT ALL Was what you saw. . . . . A VERY GOOD THING TO DO A PRETTY GOOD THING A PRETTY BAD THING A VERY BAD THING TO DO 24 H Was what you saw . . . . . NOT VERY VIOLENT PRETTY VIOLENT VERY VIOLENT EXTREMELY VIOLENT Was what you saw . . . . . NOT VERY SERIOUS A LITTLE SERIOUS PRETTY SERIOUS VERY SERIOUS IS it. . . . . . . . . . . VERY NICE FOR PEOPLE TO ACT LIKE THIS PRETTY NICE NOT VERY NICE NOT NICE AT ALL FOR PEOPLE TO ACT LIKE THIS Mkis what you saw . . . . . NOT VERY CRUEL A LITTLE CRUEL PRETTY CRUEL VERY CRUEL l 125 WE WOULD NOW LIKE TO KNOW SOME THINGS ABOUT YOU. BELOW ARE SOME THINGS THAT OTHER KIDS YOUR AGE HAVE SAID ABOUT THEMSELVES. PUT AN 'X' ON THE LINE THAT BEST DESCRIBES YOU. 10. 11. 12. Often I have trouble getting my breath. Usually, I don't get very mad at people. I like games where you can get rough and nobody cares. I have trouble swallowing. I think a fight is a good way to settle an argument. When I get angry. it usually leaves me shaking inside. I get tired very easily Sometimes I get the urge to pound someone just for fun. It doesn't bother me to watch if someone has been badly hurt. I get headaches. I can't think of a good reason for ever hitting anyone. It makes me nervous to see two of my friends fighting. A LOT A LITTLE NOT LIKE LIKE ME LIKE ME ME AT ALL l3. r14. 15. 16. 17. 18. 119. 2C). 21. .,22. 23. 126 A LOT A LITTLE LIKE ME LIKE ME I worry quite often. Anyone who says bad things about me is asking for a punch in the nose Arguing nearly always leads to trouble. I usually don't hit back, even if someone hits me first. Sometimes I don't sleep well. Sometimes you have to get tough with other kids to get your way. I don't mind a bit if my remarks hurt some- one's feelings. There's too much blood and gore in the news these days. Fighting is not a good way to get even with someone you hate. I'd rather not have my own way if I have to punch a guy to get it. I get hungry very often. NOT LIKE ME AT ALL 127 BELOW WE WOULD LIKE TO KNOW SOME GENERAL THINGS ABOUT YOU. 1. What is the name of your favorite TV show? 2. About how many hours did you watch TV yesterday? hours 3. About how many minutes did you spend reading a newspaper yesterday? minutes 4. About how many radio news programs do you listen to each day? news programs 5. About how many TV news programs do you watch each day? news programs 63. About how often would you say you read TIME or NEWSWEEK magazines or US NEWS AND WORLD REPORTS? never once in a while about every week —— —.—-—-————- 128 PUT AN 'X' NEXT TO THE SHOWS YOU WATCH EVERY WEEK OR ALMOST EVERY WEEK. My Three Sons _____Men at Law Mission: Impossible _____Perry Mason ___It Takes a Thief ebien From Shiloh Mannix _____Medical Center i h Hogan's Heroes _____Hawaii Five-O E _____The FBI _____Lancer fl Bonanza _____Ironside _____The Bold Ones _____Adam-l2 Gunsmoke ____iDan August Doris Day _____The Interns All in the Family _____Name of the Game _____Mod Squad _____Strange Report Here are some final questions about your family. What kinds of jobs do your parents have? What sort of work do they do? (For example: "Sales clerk," "Runs a gas station,” ”Works in a car factory," or "Works on a farm.") Mother-— Father-— THANK YOU VERY MUCH! "I11111111111111.1111?!“