—s~nd 7... 3):... . ‘H. ‘ .. :ng ..~...,.', 4.: 4‘: {I‘r‘r‘u 13 . 4-. r‘“~‘- s. .. “4,. , .... "4.3% 7.1; $344... .4_ :4 V“, : 1,‘ -¢;m"4u’."u-. .i-x, “(A 1. - ’3'...” 1311:"? 4" . . '3' . ,... .4 ‘r‘ ‘K , ~ . '31}: ”4:? W, ‘ ’V- :.'P~’L4" qr . 'v ’ _ IV . .L '4. f t».{4'4'.';'4' '41.. ,4 . ,I‘. : 1 “1,24; 1244.“; “~73: . ’; VII" ’I -‘~ ‘4’ ‘- ‘12-: "/“l'N/hr .-4' '7” " «w 44"» '4"). .w 3,21; v'- 43;, ':-’4!.—.,..4 .- ~.4, .4- . .4 ,4 “In 4. 1;” 4*!“ A“: .4 a ”'~»'l.u rr 3.}! ”9 - Vr(‘: . H 'j”, 4 .44 fin, I “44),“ ; '..‘ .'..'..;4..r “ WNW”, W '. .! .. aw . .- i' : y . ; fi 4.”; 4 A .1.» .4..:, r ."'J. :g- 45435.2; : ”l4: ,4 1:1! r 2"":- 1“ (:- 'qu Dr try-«tug,» rm .4314’14.:. . ,‘flrm '/"”":": r 4-; ~ 4 N $444.”. 2.4::- ’l . .(4-4-p..,.... (" 'K 1‘ II " ' (‘5 f 4 n) I .- -,4n-v»4uu4‘~.44..}, 4 r’ "_ “J, 4.4 ”r a VJ, (I. 4 4-... -o:~rv4:44r4, ) - , ’4‘! "f4 29"} ‘ chm "ll’l44110”}., ., 9'. :- rv‘yyfrnn’am fly)... .; {4'14” _. , r 1 ,r 4- Air,” .4...” 4:4, 4: 4 4. "0,7,1. ’9‘)" “4» -~ ‘44 4,. .-4~.4-.4L 54" “w- 4 hub}; and. " ' 4 ”an «no . 4r mo ”'H -r - Jan" 3 44,143 4 4-. , . .14. '4' ya; 2355'}: , 1:1: cup-n;- ' w z 4% 332% -= fig J‘mv' M?"€W; ’5' an '72:: 34.44:,” mm» 4. _ WY- 4‘-’ "27 .-. 'w-I: J‘s??? “3m: ' 56,5"3;zzrwf ”gm W1: ' .nlm T.’ "1.433;": fwyg. .4, “my“. ”W." t I-"vu."'n4’4""‘ .U,‘ ,.4 4 in. é’dd‘rmléw in“ 72' yu'wrrr IW “4:4: "’4' na- {fa-rm Own-r: fix... an}. ' I. . ..,'.., 4 ...-.‘”v.‘”.’ rm “N’ffi‘lzi 1:; 4442.. ';IID H. a... ,fi: ,. 44.: w»? 1:»va mu- *- y W V —- Wat: Mun-MM m4 fads...“ .4 - 4...“... W’s-"-w 5....-. , ‘27“ :224- Vvv 3&3 _£m vaM-— 4 aw... .4 M 7w.“ 24......5 .. I r. ’ 4?“ J... . ' 1&4..«.;3‘3""9::::Wg§‘ 1' + A , V. .1“ 'r 9.11,? {4492433435, &!P ’MJ ff “2“. fidm:;wa&a ’M-~,f‘m ' ,4 4... N v film ”1 10?. :4 . , an em 4." .5422: ‘3”:3'4 ""1”... WW - , 4—..4414 ”2-4;: 1'0"“: gm 1.. ’ v! 4W4.» '37??qu - 4w 4... - .. .4; 4 r5. 4 ‘ 'y‘.:;‘fi*' 4, 44.:w’,’;v- “2mg?“ 4 . 3:14;; ‘ ””‘v~.~«~:trum.m&zm ‘ .. 4 :24?” mw?w “4......” .. ~44 4.. w. .. . g: ' f‘ 1;. .' I1???) Ti n . -. c . H ‘ \. 34.5.! L u 'I*W‘EM'WEEWE‘ELW“ ‘ 9 3 2 6 MACE 2 ”m " {339 g 999 W 3.?56657 1 2- 5 211““ MA; , Q o& 0 0 ‘2 I )7 0H >3, OASP m: g) a m:’ R 18% 22002 ABSTRACT THE INFLUENCE OF SOCIAL CLASS, THE FAMILY, AND EXPOSURE TO TELEVISION VIOLENCE ON THE SOCIALIZATION OF AGGRESSION By Joseph R. Dominick This study examined the influences of three ante— cedent variables--exposure to TV violence, perceived fam- ily attitudes toward violence, and socio-economic status-- on the socialization of attitudes toward aggression. The respondents were 434 boys from ages 9-11. Support was found for the following hypotheses: (1) Boys from families where attitudes toward aggres- sion were left undefined were: (a) more approving of violence; (b) more willing to use violence; (c) more likely to suggest violence as a response to a conflict situation; and, (d) more apt to believe that violence was an effec- tive way to achieve some goal than were boys who reported their families to be against violence. Joseph R. Dominick (2) Children who were frequently exposed to TV violence differed from boys not so regularly exposed in the following manner. They were: (a) more willing to use violence; and, (b) more likely to perceive violence as effective. (3) The interaction of high exposure to TV violence and little exposure to counter-information about violence from the family led to: (a) more approval of violence; and, (b) more willingness to use violence. A hypothesized secondeorder interaction among the three antecedent variables was significant for two of the dependent measures—-approval of aggression and suggesting violence in response to conflict situations. For a third variable--willingness to use violence-~this interaction showed a strong similar tendency (p < .10). In each case, the obtained pattern of means indicated that exposure to TV Violence and low exposure to counter-information pro- duced the greatest degree of acceptance of violence among middle class children. These two variables did not inter- act significantly among the lower class group. We had ex- pected that both social classes should demonstrate an Joseph R. Dominick interaction between.these two variables with the interac- tion being more pronounced among lower-class boys. The following hypotheses were not supported: (1) Children with a high degree of exposure to TV violence should: (2) (3) (a) (b) The and show more approval of violence; and, suggest violence more often as a response to conflict situations. interaction of high exposure to TV violence little exposure to counter~information about violence from the family should lead to: (a) (b) The low (a) (b) (C) (d) more suggestions of Violence in conflict situations; and, more tendency to perceive violence as effective. interaction of high exposure to TV violence and socio-economic status should lead to: more approval of violence; more willingness to use violence; more suggestions.of violence in conflict situa- tions; and, more tendency to perceive violence as effective. Possible reasons for the incomplete confirmation of the hy- potheses are suggested. Research extensions of the present Joseph R. Dominick study and implications.for future research concerning the media and socialization are discussed. THE INFLUENCE OF SOCIAL CLASS, THE FAMILY, AND EXPOSURE TO TELEVISION VIOLENCE ON THE SOCIALIZATION OF AGGRESSION BY Joseph R. Dominick ii A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Communication 1970 «u :3“; (r The research upon which this dissertation is based was performed pursuant to Contract No. HSM 42—70-32 with the National Institute of Mental Health, Health Services and Mental Health Administration, Department of Health, Education, and Welfare. ACKNOWLEDGMENTS The-author would like to express his thanks and appreciation to his committee chairman, Dr. Bradley S. Greenberg, for his valuable assistance and guidance throughout the planning and writing of this thesis. The author would also like to extend his gratitude to other members of his committee, Dr. Thomas Baldwin, Dr. R. Vincent Farace, Dr. Randall Harrison, and Dr. Frank Pinner. Helpful criticism and suggestions were also made by several members of a seminar group concerned with vio- lence and the media; namely, Dr. Natan Katzman, Thomas Gordon, Rober Haney, Stuart Surlin, and Mantha Vlahos. Mrs. Joyce Haney was instrumental in arranging the pretest. For their assistance in data collection, the author would like to thank Gary Bogart, Cynthia Guy, and Judy Robinson. Michael Fisher assisted both in the pre- test and in the final collection and analysis of the data. The author would also like to thank Mr. Roger Fee— man, Mr. John Peruzzi, Mr. Douglas Thomasma, Mrs. Beatrice Forward, Mr. Lewis Bradway, Dr. Norman Taylor, Dr. Lawrence iii Reed, Mr. Leon Roberson and numerous teachers in the Battle Creek and Jackson area for their co-operation in collecting the data. iv TABLE OF CONTENTS Page LIST OF TABLES o o o o o o o a o o o o o o o o o o 0 vii LIST OF FIGURES. o o o o o o o o o o o o O o o o o o Viii Chapter 1. INTRODUCTION. . . . . . . . . . . . . . . . . 1 Statement of the Problem. . . . . . . . . . 1 Conceptual Model. . . . . . . . . . . . . . 3 Review of Literature. . . . . . . . . . . . 5 Characteristics of Television and the Child [g Antecedent Variables. . . . . . . . . . . . l7 Dependent Variables . . . . . . . . . . . . l8 Hypotheses. . . . . . . . . . . . . . . . . l9 2. MTHODS O O C O O O O O O O O O O O O O O O O 27 sample. 0 0 O O O O O O C O O O O O O C O O 27 Variables and Operationalizations . . . . . 28 Analytic Procedures . . . . . . . . . . . . 34 3. RESULTS 0 O O O O O O O O I O O O O O O O O O 36 Approval of Aggression. . . . . . . . . . . 37 Willingness to Use Violence . . . . . . . . 42 Use of Violence in Conflict Situations. . . 48 Perceived Effectiveness of Violence . . . . 54 4. DISCUSSION. 0 O O O O O O O O O O O O O O O C 59 Summary . . . . . . . . . . . . . . . . . . 59 Discussion. . . . . . . . . . . . . . . . . 63 TABLE OF CONTENTS (cont.) Chapter Page Implications. O O O O O O O O O 0 O C O C 0 68 Research Extensions . . . . . . . . . . . . 72 BIBLIOGRAPHY O O C O O O O C C C O C O O O O O O O O 75 APPENDIX 0 O C O O 9 O O O C- C O O O I O O O O O O O 79 vi Table 10. LIST OF TABLES Intercorrelations among the four dependent variables. 0 O O O O O 0 O O O O O O ,. O O Three-way analysis of variance for approval of aggression scale. . . . . . . . . . . . Two-way analysis of variance by social class for approval of aggression scale . . . . . Three—way analysis of variance for willing- ness to use violence scale . . . . . . . . Two-way analysis of variance by social class for willingness to use violence scale. . . Three-way analysis of variance for use of violence in conflict situations scale. . . Two-way analysis of variance by social class for use of violence in conflict situations scale. 0 O O O I C O C C C O O O C C O O O Three-way analysis of variance for perceived effectiveness of violence scale. . . . . . Two-way analysis of variance by social class for perceived effectiveness of violence- scale. 0 O O O O O O O O C O O O O O O O 0 Summary of probabilities values of F for each dependent variable. . . . . . . . . . vii Page 39 43 46 47 50 53 55 57 58 62 LIST OF FIGURES Figure Page 1. Predicted pattern of means compromising second-order interaction . . . .v. . . . . 24 2. Comparison of predicted and obtained cell means for the approval of aggression scale 40 3. Comparison of predicted and obtained cell means for use of violence in conflict situations scale . . . . . . . . . . . . . 51 viii CHAPTER I INTRODUCTION Statement of the Problem The process whereby an individual acquires the cul— ture of his group and internalizes its social norms can be thought of as socialization. Socialization is a continu— ous process-~extending from childhood through old age. Some norms, such as basic rules about food and eating, are transmitted to the individual as a child. Other norms, such as rules about dating behavicmy are postponed until later. Still others may involve continuous instruction throughout life. Responsibility for socialization is found in speci— fic people or institutions-~called agencies—~depending on the normative area involved. The mother, for example, usu- ally directs toilet training while other members of the profession supervise occupational training. fSocialization is often deliberate (Elkin, 1960), but occasionally it occurs inadvertently when an individual picks up cues about 1 social norms from his environment without receiving spe- cific instructions about them. ‘Kx Certain data indicate that the mass media may be playing an increasingly important role in the socializa- \ tion process. The North American child, for example, spends about one-sixth of his waking hoursiwatahingmtele- vi 8 i on . , Thi s i s a lmoggflas.mughmtime..as...he.,..sp.endsminlsghonl . and more time than he devotes to any gther actiyigngxgep; sleep and play (Schramm, Lyle, and Parker, 1961). Data on the exposure patterns of children, while helpful, cannot alone establish the nature of the media's socializing influences. In other words, there is no neces- sary correlation between the amount of time a child spends with a.mass medium and the impact this exposure may have on him. To understand the role of the media as socializa- tion agencies, we need to determine (Wright, 1959): (l) the relative influence of the media compared to other agencies of socialization, e.g., the family, the school, with regard to some normative area; and, (2) what differences might result if the major part of socialization is carried out by the impersonal mass media rather than by more primary agents. These questions are complex. Developing into an adult is probably too complicated a task to make it likely that any one influence (such as the school, the family, or the media) will predictably produce a specific effect. In- stead, we must examine the meshing of influences from var- ious agencies and certain background characteristics of the child before we can reliably evaluate what effects the media may have had (Schramm, g£_al., 1961). This research is a first step in suggesting answers to such questions. The effects of one mass medium-~televi- sion--are assessed in relation to one other agency--the family-~concerning one specific aspect of socialization-- the attitudes of a child toward the use of physical force or violence. The Conceptual Model As a child grows up, he is forced to deal with the task of processing a massive amount of information about his environment. Much of this information is gathered /” through direct contact with the outside world. (Other information, dealing primarily with parts of the environ- ment that the child cannot experience first hand, is passed A x (12- 51», a. .. ‘ on to him through other persons or institutions. This available information is used by the child to accumulate knowledge, to form values, attitudes, and opinions, and perhaps to guide some portion of his behavior. At a gen- eral level, this process can be represented schematically 4" n ,' l. I;- '. . i as the following: Family vanowledge School\\\\\\‘ i::::;Attitudes \\\\\? -¥> Friends////7 \\\\g£§rceptions .. Behaviors //”\\ Mass Media I 1‘5" ‘PI'H. The arrows in the above diagram represent channels of com- munication and information processing available to the child. For example, as a child is socialized into American society, he is expected to become competent as a citizen in his relation to the government. The information he needs to do this, however, is often difficult for him to obtain directly. As a result, he may learn in school how a bill becomes a law. His family may teach him about voting. Television might show him political conventions and the mechanics of electing a President. From all these sources, the child learns about the political system. The same sources might also influence the attitudes a child develops about politics. His family might indoctrinate him about the strengths of a certain political party. His friends might suggest that all politicians are crooked. All these sources serve to shape the individual's political attitudes. Of more relevance to the present paper is the situa- tion where the linkages between agency and child are weak or even non—existent. In this instance, the most influen- tial socialization agency might be the mass media. Review of Literature Several studies have investigated this situation. Siegel (1958) using a sample of children from a semi-rural area who had little previous contact with taxi drivers, ex- posed second graders to a series of radio programs about taxi operators. Half the children heard stories in which the drivers responded aggressively in a given situation while others heard a version in which the driver responded /" non-aggressively. The group hearing the aggressive version expected taxi drivers in real-life situations to behave in the same manner as the radio taxi drivers. In the absence of competing information, the programs had shaped the chil— dren's expectations about taxi drivers. More recently, Gerson (1963) investigated the ef- fect of the media as socialization agents on the dating be- havior of adolescents. Under.several control conditions used in the analysis, black teen-agers were more likely to use the media more for socialization than were whites. This supported Gerson's reasoning that blacks were unlikely to receive this kind of information from other sources and as a result would rely more on the media. These findings sug- gest the media may be effective when there is a discontin- uity in the socializing functions of traditional institu- tions. Finally, additional evidence of the media's effec- tiveness in forming beliefs in the absence of competing in- formation is found in a study by Greenberg and Dominick (1969). Three groups of teen-agers, low-income blacks from a ghetto area, low-income whites, and middle-income whites from a suburban area, were asked a series of questions de- signed to measure how much they believed that life as shown on television was similar to the way things really are. The- group of teen-agers with the least contact with middle-class society (the world portrayed primarily by TV) should show the most faith in its reality while the group having the most experience should show the greatest disbelief. The results supported this expectation with the greatest degree of belief being shown by the low-income blacks, the next highest by the low-income whites, and the lowest by the middle-income whites. The pattern of effects that television may have on a child depends on the interaction of two factors: what television presents to the child and what the child brings to television. Several researchers (Schramm, gt_al., 1961; Himmelweit, gt_al., 1958; Maccoby, 1964) have specified when-the media, particularly television, are likely to have maximal influence. Their generalizations reflect the inter- play of these two factors. The media ought to be most in- fluential when the child: (1) is exposed to a set of ideas or behaviors which recur from program to program; (2) is a heavy user of the medium; and, (3) is likely to have limited contact with other socialization agencies and con- sequently is not supplied with a set of values against which to assess the views of the media. To apply this reasoning to the present problem two steps are necessary. First, we must examine what televi- sion presents to the child concerning force or violence. Secondly, we should examine what other agencies are likely to be important in shaping the ideas the child brings with him to television. Then, we may present hypotheses which deal with probable effects of the interaction of these two dimensions. Characteristiqs of Television ‘ and the Child Violence as Presented on Television Several content analyses indicate that the TV world is a violent one. Most of the studies discussed below have used different definitions of violent content. Despite these differences, the studies are informative as to tele- vision program content. {9%79 A 1955.5tudy (U.S. Senate) found that almost one- fourth of the programming in the nation's largest cities featured violent content. Later, another Senate committee (U.S. Senate, 1968) found that almost one—half of the shows in the 4-10 p.m. period (peak viewing hours for children) were action—adventure shows—-the program-type most likely to portray violence. A further analysis of the shows in this category disclosed that acts of violence outnumbered acts of protective behavior by four to one. More recently, two studies indicate that the level of violence on television is still high, at least through the 1968-69 season. In a survey conducted by the Christian Science Monitor six weeks after Robert Kennedy's assassina- tion, researchers found that in 85—1/2 hours of prime time and Saturday programming, 84 killings took place. This survey found that the most violent evening hours were 7:30 to 9 p.m. when approximately 27 million children from 2-17 were watching. During these hours, violent incidents occurred once every 16 minutes. There was a murder or killing every 35 minutes. 10 An extensive study by Gerbner (1968) further sub- stantiates these findings. Acts of violence occurred in eight of ten shows. Dramatic shows had seven violent epi- sodes while cartoon shows had three times that number. For an entire week, the study found that 400 people were killed. More germaine to the present analysis are studies which examine violence as it specifically relates to problem-solving. Stampel (1969) analyzed one week of net— work programs examining the means used to solve problems. For the entire week, 202 problems were counted. Of these, almost 60% were solved by violent tactics; about one-third were solved non-violently while the remainder went unsolved. Larsen, Gray, and Fortis (1968) used a broader con— text in analyzing TV violence. They identified "program goals" and the means by which these goals were achieved. Violent means were the most prevalent. The authors also found that childrens' shows were even more likely than adult shows to use violence to achieve goals. Gerbner (1969) analyzed the personalities of vio- lent characters on the basis of coders' judgments. He re- ported that violent performers were judged to be more ll logical and efficient than non-violent characters. This reiterates the findings that violent action is a widely- used and efficient method of problem—solving. These studies lead to the following generalizations about the television world: (1) A chi1d“whgiwarchesmaniayerageiamonnt of TV is likfiixisgiseeraisubstantialiEEQHBE.ofwviolent contgnp. (2) Typigsllyi violence is Exe§ented-as a highly §RQ§§$S£U1 means of 99%;:sghievement. (3) As of the 1968-69 season, violence ismphe pre— dominant means of conflictfresolutionfoundin EYMQFBWSfi This, then, is what television is presenting to the child. We now examine another component of the problem. Shaping the Child's Attitudes Toward Violence What the child brings to television will be the result of what he has learned from other socialization agencies. Research indicates that theifamilymismtherkey source_foria childis attitudes toward violence. To under- stand the family's role in this process, it is necessary to review some research findings from the field of child development. 12 Most likely it will be a family member that becomes the target for the child's first attempts at violence (Sears, Maccoby, and Levin, 1957). As a child grows older, most con- flicts occur with siblings inside his home.- Again, it will be the parent who will have to either reward or punish these aggressive behaviors. (Sears, et_alf, 1957) Not only does the family administer positive or negative reinforcement for aggression, it also serves as a source from which the child can learn models of problem- solving that are essentially non—violent. The family, for example, could provide the child with examples of democratic decision-making, arbitration, or compromises--all of which expose the child to some alternative to violence for problem—solving. There are variations among families in the degree to which they use these methods. An early study (Sewell, Mussen, and Harris, 1955) extracted a pattern analysis from the Fels Parent Behavior Rating of 125 families. One major pattern which emerged was a measure of democracy vs. auto- cracy. Frequent parent-child conversations and formalized techniques for solving family conflicts were found in l3 democratic households. The authoritarian households did not use these techniques for problem—solving and parent— child interactions were less frequent. A recent series of studies (McLeod §E_al., 1966; 1967) has suggested that communication patterns within fam- ilies can affect the socialization of the child. These re- searchers constructed four family types. Most relevant to this research is a family type labelled "pluralistic." In this family, a child is exposed more often to both sides of an issue. Discussion of controversial matters is encour- aged. A child may be more frequently exposed to alterna— tives to violence in this family type. These homes are perhaps more likely to foster techniques that do not empha- size aggression to solve problems. Parents may also influence the way a child thinks about violence in a way more directly connected with tele- vision. AguitmcommentsuonTV content... cassettesas import- wgpgsssafisasnildren (HickS: 1968)~ If a Parent comments on the inappropriateness of violence while watch- ing an aggressive scene with his children, the child may develop more negative attitudes toward violence. ”'54) Conversely, a child who observes his parents watching vio- lence while calmly eating dinner or performing some other routine activity might come to accept violence as normal. In other words, parents have the opportunity to either counteract or legitimatize aggression while viewing TV violence With their children (Sakuma, 1968). To summarize and to draw conclusions from the above research: (1) The family is the first agency which deals with a child's aggressive behavior. (2) Families can influence the way a child thinks about violence by (a) providing positive or negative feed— back when the child himself is aggressive; (b) by using specific methods of problem-solving within the family; and, (c) by giving feedback to the child about scenes of violence shown on TV. (3) Families vary in the extent to which they use the three techniques mentioned above. The Influence of Social Class For several reasons, socio-economic background may also influence the pattern of effects that stem from expo- sure to TV violence. A child from amlowaES family is more “may up“-m h‘u - 1335.212 thinlamidsle classishlludflto b? a heavy Viewer of.._'I.-3Y. s5) (Schramm, et a1., 1961; Greenberg and Dominick, 1969; 1970). Consequently, a lower—class child is probably exposed to a larger number of violent episodes than his middle-class counterpart. The conceptual framework outlined above sug- gests that the media have more of a potential for influenc— ing children who are heavy users (Maccoby, 1964). Social status may also affect what the child brings With him to teleViSion. Allinsmith ( found that w— aishilirsiisre mprs likelyrthan theifmidleIClBSR Peers to ressoiridiizs.pgtentislly“frustrating ééféustions with the most direct forms of aggression.“ Moreover, these lower- income youngsters habitually expressed more aggressive be— havior than their middle-class peers. Further: MEET??? 9’? Fhe P9??.-f=hi;1¢ outside his home is more likely to contain frequent acts of physi— cal yiolenge (U.S. Government, 1968). Fighting with peers, incidents involving violence among neighbors, and disputes with police are events that characterize the environment of a low-income youngster. This greater exposure and fa— miliarity With real—life aggression among low-SES children ! might make them more tolerant of violence and more willing to use it to solve problems. 1‘ \ / ( . 16;; k/‘ The family may be less of an influential socializ— ing agent among lower-class children. Among low—income families, parent-child interactions are erratic and incon- sistent. Parents and children see each other on a non- systematic, disorganized basis (Minuchin et a1., 1967). Many lower-class husbands also leave the family unit for various reasons. As a result, mothers are forced to work, further emphasizing the fragmented interaction between parent and child. This kind of environment seems less likely than a middle-class home to foster the awareness and development of those techniques which could serve as alternatives to violent behavior. Consequently, the ef- fects of TV violence would seem to have the greatest po- tential among low-SES children. Again, to summarize: Social class should influence the effects of TV violence on the child's attitudes toward aggression inasmuch as-- (l) Low-income youngsters are more apt to see more vio- lent TV content. (2) Low-income youngsters are more likely to be exposed to real-life aggression. (3) Low-income families are less likely to be as effec- tive as middle-income families in providing alterna- tives to violent behavior. 17 Antecedent Variables The conceptual framework for this research suggests that the effects of television violence will be at its maxi- mum when a lower—income child is highly exposed to TV ag— gression and is not provided with a set of values or com— peting information that will offset the media emphasis. Conversely, a middle—class child, with low exposure, and a strong set of alternatives provided by other sources should show the least effect. iables are: (1) the d99r€9WPQNWhich the child is eXPOSed to Viqignt COBFPP’C; ' ~ . (2) the degree to which a child is providedwwi h §,$et of alternative views about violence by other agen- cies-—in this instance, the family; and, (3) the child's sociojeconomic status. Consider the following typological scheme: Low Income Middle Income Family attitude Family attitude toward violence toward violence Negative Undefined Negative Undefined Exposure Low A B Low A' B' to TV violence High C D High C' D' 18 Those children in cells D and D'-—high exposure to TV vio— lence and little exposure to offsetting information——ought to be the most influenced in their respective social classes by TV violence. These children might be identified as "tele- vision socializees" who have come to accept the norms of the television world. Dependent Variables From prior content analyses of television, we can derive what appear to be the norms of TV violence: (1) Violence is an approved method of dealing with problems as evidenced by its frequency. (2) Violence is effective. (3) Violence is the most frequently used means of problem-solving. \x These norms suggest four behaviors which sould be dependent on exposure to TV violence: (1) Approval of violence--a measure of what the child thinks about certain forms of violence that are generally considered unacceptable in our society. (2) Willingness to use violence--a measure of the will- ingness of the child to report that he would use violence to deal with his own problems in real life. l9 (3) Perceived effectiveness of violence--an index of how effective the child perceives violence to be as a means to solve problems. (4) Ability to suggest solutions to conflict situations-— an attempt to gauge the number of solutions a child would suggest when presented with a potentially frustrating situation. Hypotheses Prior research (Sears e£_al., 1957; Sears, 1961) indicates that sex differences are present in the atti— tudes a child has toward violent behavior. Accordingly, data will be presented for males only. The hypotheses de— pict differences among boys. Approval of Violence, Willingness to use Violence and Perceived Effectiveness The three main antecedent variables which should influence the child's attitudes toward violence are expo- sure to television violence, exposure to countering infor- mation about violence, and socio—economic status.- Each of these should exert a separate effect. 20 The conceptual framework summarized by Maccoby (1964) suggests that the more a child watches television the more we will accept the point of view expressed by TV content. Applying this to violent content, we hypothesize: H1: Youngsters with a high degree of exposure to TV violence will: (la) indicate higher approval of violent behavior; (1b) be more willing to use violence; and, (1c) perceive violence to be a more effective way of solving problems than will youngsters with a low degree of exposure. The child—rearing research cited above suggests that the degree to which the child's family provides him with information about alternatives to violent behavior should also influence the way he looks at violence. Con— sequently, H2: Youngsters with a high degree of exposure to neg- ative information about violence will: (2a) indicate less approval of violence; (2b) be less willing to use violence; and, (2c) perceive violence to be a less effective way of solving problems than will youngsters with a low degree of exposure. 21 Further, the effects of exposure to media violence should be affected by the interaction of both a child's social class and by the influences of interpersonal agents. As emphasized by Schramm g£_al., (1961), the potential ef- fects of television ought to be the reciprocal of the in- fluence of more personal sources. Within the present con— text, we should expect those children from both social classes who are frequently exposed to violent content and have little contact with countering information about vio- lence to be most influenced. We hypothesize: H3: The interaction of more exposure to television violence with low exposure to counter—information leads to more: (3a) approval of violence; (3b) willingness to use violence; and, (3c) perceived effectiveness of violence. The rationale also indicates that the effects of exposure to TV violence will not be independent of social class. Low—income children are more exposed to television and are more likely to possess favorable attitudes about violence than are middle—class children. As a result, 22 H4: The interaction of more exposure to television violence with low socio-economic status leads to more: (4a) approval of violence; (4b) willingness to use violence; and, (4c) perceived effectiveness of violence. The joint interaction of the first two variables discussed above--exposure to TV violence and exposure to counter—information about violence-~should also be influ— enced by soc1al class. In contrast to a middle—income youngster, a low-income child is surrounded by an environ— ment characterized by violent acts. The low-income child is more likely to belong to a peer group which emphasizes physical acts or toughness (U.S. Government, 1968). The violence a low—income child sees on television probably reinforces the prevailing attitudes of his peer group. Consequently, the lower-class youngster who is a heavy viewer of violence and receives little countering information from his family should accept the norms of the world of TV violence more than should his middle-class peer. Television violence should have more of an effect on this individual since it portrays values more congruent 23 with the life-style of a low-income youngster. In other words, the pattern of interaction between exposure to vio- lence and exposure to counter-information should vary from class to class. It is hypothesized, ' H5: The interaction of more exposure to TV vio- lence with low exposure to counter—informa— tion and low socio-economic status leads to more: (5a) approval of violence; (5b) willingness to use violence; and, (5c) perceived effectiveness of violence. Figure 1 contains a diagram of the probable pattern of means likely to result from these predictions.* Ability to Suggest Solutions to Conflict Situations Social class and the degree to which the family provides various examples of problem—solving are variables *The research design also allows us to test two additional hypotheses of secondary interest to the study. They are: (l) Low-income youngsters will indicate more approval of violence, be more willing to use violence, and perceive violence as more effective than middle—income youngsters. (2) The interaction of a low degree of exposure to counter- information about violence with low socio—economic status 24 High Low competing Dependent information variable LowernC1ass High competing information Low High Exposure to TV violence High Dependent . l . variab e Middle Class Lowacompeting information / ”,,,1,#r*””’””#;i;h competing information Low High Exposure to TV violence Figure 1 Graph of predicted pattern of means comprising second—order interaction. 25 more relevant to this issue than TV. Therefore, the hy- potheses in this section are of secondary interest. Children from families which use arbitration, com- promise, or mediation, should suggest more alternatives to problem-solving than children not so exposed. Youngsters exposed to these techniques have a greater likelihood of using them in conflict situations since they have had more opportunity to see them. They also have been more exposed to adult models for examples-~a condition which should re- sult in more imitation. The key variable seems to be how familiar the child is with different options, a condition more likely affected by his family. A parent can intervene directly in disagreements among playmates or siblings and can pro- vide immediate advice and alternatives that are suitable for the situation. The impersonal nature of TV limits its effectiveness and its flexibility. We hypothesize: H6: Children who have a high degree of exposure to alternatives of problem solving will mention more solutions than children with a low degree of exposure. leads to more approval of violence, more willingness to use violence, and more perceived effectiveness of violence. 26 Social class should be influential in this area also because of the more limited parent—child interaction in lower—class homes, we hypothesize: H7: Middle—income children will offer more different solutions to problem situations than low-income children. CHAPTER II METHODS The total sample was made up of 434 boys. Each respondent was then assigned to one of eight sub-groups according to the amount of TV violence he viewed regu— larly, his family's attitudes toward violence, and his social class. Sample Questionnaires were given to fourth, fifth, and sixth graders in six schools located in and around Battle Creek and Jackson, Michigan. The questionnaire was read aloud to the children during class sessions on May 1 and May 8, 1970. The schools in the Battle Creek area were in a district with families representing a wide social class range. One Jackson school was in the inner city while a second was in a middle-class residential district. 27 28 Completed questionnaires were obtained from 434 boys. About nine percent of the sample was black. Variables and Operationalizations Antecedent Variables Three antecedent variables were examined——the child's exposure to television violence, the child's perceptions of his family's attitudes toward violence, and the family's socio—economic status. Exposure to TV violence. Each child was given a list of 28 TV programs avail— able on his local television stations. Twenty of these shows had been judged by a sample of newspaper and magazine critics to contain at least some violent activity (Greenberg and Gordon, 1970). Coders recorded the number of these 20 shows that each child reported watching regularly. Obtained scores on this variable ranged from 0 to 20 and were norm- ally distributed with a standard deviation of 3.7. 29 Family attitudes toward violence. The children were asked a series of seven questions about how they thought their parents felt about various forms of violence, e.g., Suppose you got into a fight with one of your friends. How—do you think your parents would feel about it? Suppose your teacher told your parents that you had been fighting in school. What do you think your parents would do? Other items asked about whether or not the child had ever been punished for fighting, whether the parents usually intervened to help the child solve problems with his friends and whether or not the parents ever commented on television violence. All seven items in this section were found to inter- correlate significantly with one another, with correlations ranging from .38 to .70. The obtained scores were summed per respondent to form an index ranging from 7 (low approval of violence) to 17 (high approval). Social class. Each respondent was asked to write down what kinds of jobs his parents had. The reported occupation was then 3O assigned a position on a 13 point scale of occupational prestige (Troldahl, 1967). Approximately 15% of the youngsters gave either no response or a response which was illegible. These were assigned the modal occupational prestige rating of all children in the particular school attended. Dependent Variables Four sets of dependent variables were used. Approval of violence. This measure was derived from Sears' aggression scales. Specifically, eight modified items from the scale labelled "antisocial aggression" were given to the children. These items were declarative sentences referring to aggres— sions that are generally socially disapproved in this cul- ture (e.g., "I see nothing wrong in a fight between two teen-age boys." "It's all right if a man slaps his wife."). Response categories consisted of three alternatives, "I agree," "I'm not sure,‘ and "I disagree." A score of three was assigned to the alternative indicating maximum approval 31 of violence; a score of one was assigned to the opposite alternative. Scores were summed across the eight items yielding an index which ranged from 8 (low approval) through 24 (high approval). Willingness to use violence. This index measured the child's willingness to use violence in real life. Scales were adapted from the Buss- Durkee Hostility Inventory (Buss, 1957). The format of these items is similar to that of the Sears' agQPéésion” _scales. IFive declarative sentences had response scales that permitted the child to indicate his agreement or dis- agreement. In these items, however, the sentences dealt with whether or not the individual would use some sort of physical violence in certain situations, (e.g., "Anybody who says bad things about me is looking for a fight." "People who keep on bothering me are looking for a punch in the nose.") Again, a single index per subject was formed by summing across the items. A score of 5 indi- cates low willingness to resort to violence while a score of 10 denotes high willingness. 32 Perceived effectiveness of violence. Five items were constructed to measure how effec- tive the child perceived violence to be as a means to solve problems. These items described violence in terms of its efficacy to gain desired objectives. For example, "Some- times a fight is the easiest way to get what you want." "A fight is the best way to settle an argument once and for all." The response categories were "I agree,” "I'm not sure," and "I disagree," and were coded in the same manner as the Sears' scale. Scores were summed across items with five representing low perceived effectiveness and 15 high perceived effectiveness. Ability to suggest solutions to conflict situations. The hypotheses for this variable were originally conceived in terms of the number of solutions a child would offer to solve a conflict. Pretesting indicated that only a few children were able to come up with more than one problem situation. Consequently, an adequate measure was not achieved. Hypotheses #6 and #7, then, were not tested. 33 A revised format, however, was included in the final questionnaire. In four open-ended questions, a po- tentially frustrating situation was described to the child. He then wrote down the one thing he would most likely do in that situation. For example, "Pretend somebody you know takes something from you and breaks it on purpose. What would you do? or "Pretend somebody you know tells lies about you. What would you do?" I A score of "l" was assigned if the child's response was judged to be non-violent; a score of "2" was assigned if the response was judged violent. (For these questions, violence was defined as behavior which would produce phys- ical pain in another person.) An index was created for each child with a score of four indicating all noneviolent responses and a score of eight representing all violent re- sponses. In its new form, the measure resembled the "willing- ness to use violence" scale and corresponding predictions were made. 34 Analytic Procedures The sample was divided into eight sub-groups. The first divisionesa median split--was made according to the occupationaliprestige of the child's family. All children in the three lowest categories of the l3-step prestige scale were classified in the low income category; children with a rating of four and above were placed in the middle- income group. The mean occupational prestige rating for the lowfincome group (n=218) was 2.80 while the mean rating for the middle-income group (n=216) was 5.72. Each sub-group was then divided according to the number of violent shows each child watched each week. The distribution on this variable ranged from 0 to 20 and was divided at the median. Children who watched eight or fewer violent shows per week were placed in the low-exposure group while the remaining children were classified as high expo- .'1 1‘ sure. (‘ Finally, each sub—group was divided according to the child's perceptions of his family's attitudes toward violence. This variable had a range from 7 (low approval) to 17 (high approval). The distribution on this index was 35 skewed toward the low approval end of the scale. Children with scores of 7-10 were put in the low-approval group. _More than ninety percent of the remaining children indicated that they were unsure or didn't know how their parents felt about violence. Less than ten percent reported that their families gave high approval to violence.< All children scor- ing 10 or above on this scale were labelled as "undefined." In other words, the two groups of children did not repre- sent the poles on this variable. The "low approval" group can be thought of as children who definitely perceive their families to be anti-violence. The "undefined" group con- sists of young people whose parents had not demonstrated disapproval of violence to their children. It The three divisions resulted in eight groups of the following sizes: Middle Class Lower Class Family attitude Family attitude toward violence toward violence Low approval Undefined Low approval ( Undefined Exposure to TV Low: n = 47 n = 60 n = 62 n = 40 violence High: n = 57 n = 52 n = 50 n = 66 CHAPTER III RESULTS The results are presented in terms of the four main dependent behaviors: l) the respondents' approval of aggres- sion; 2) their willingness to use Violence; 3) use of vio- 1ence in conflict situations; and, 4) perceived effective- ness of violence. Exposure to violence correlated .10 with family attitudes toward aggression and r.09 with social class; family attitudes correlated —.10 with social class. Table 1 contains the intercorrelations of the four dependent variables. For each variable the hypotheses followed a stepwise progression from predictions about main effects, through first-level interactions, to the second-order interaction. Specific predictions were made about the pattern of means that would constitute the higher-level interactions. For consistency, the results will be presented in the same for- mat, although it is recognized that a statistical interac- tion significantly qualifies statements about main effects. 36 37" Approval of Aggression It was hypothesized that more approval of various forms of aggression ought to come from children who: (Hla) are frequently exposed to TV violence, and, :pJHZa) are not provided by their families with counter- ‘ information about violence. Moreover, more approval of aggression should stem from: (H3a) the interaction of high exposure to TV violence and low exposure to counter—information; and J$\(H4a) the interaction of high exposure to TV violence and low social class. Finally, more approval of aggression should result from: H§f(H5a) the joint interaction of high exposure to TV violence, low exposure to counter-information, and low social class. A three-way analysis of variance with unequal cell sizes (Snedecor, 1956) was performed on the approval of aggression scale. The results of this analysis are in Table l. The first hypothesis predicting higher approval of aggression among the group with high exposure to TV 38 violence was not supported. There was no main effect for the TV violence variable. Support was found for the second hypothesis posit- ing higher approval scores from children whose parents do not provide counter-information about violence. The F ratio for main effects for the family attitudes variable ”stun..- ‘n.._..-~-~" , . q "" ‘r‘fiw‘l‘. —‘- ismsignificant beyond the .0005 level. An examination of the means in Table 1 shows that in all cases children from families who were classified as "undefined" scored from approximately one to two scale units higher than children from families who are anti-violence. A significant main effect was also found for the social class variable with lower—class children generally scoring higher on approval of aggression. This finding supports those of Allinsmith (1960) mentioned earlier. The predicted interaction between exposure to vio— lence and social class (Hypothesis 4a) did not quite reach significance (p < .10). The interaction between exposure to violence and family attitude stated in Hypothesis 3a was significant. The second order interaction (Hypothesis 5a) was also significant as predicted. An examination of Figure 2, 39 TABLE 1 Intercorrelations among the Four Dependent Variables Approval of Willingness Perceived aggression to use Vio— effectiveness scale lence scale of violence scale Approval of aggression scale - Willingness to use vio- lence scale .43 - Use of vio— lence in conflict situations scale .22 .31 Perceived effective- ness of violence scale .33 .43 .36 4O High High Approval Approval of Low competing of Aggression information Aggression H' 1 i . .lgl compet ng Low competing information _ . information High competing W Low High Low High Exposure to TV violence Exposure to TV violence Lower Class Middle Class . . . . . . . . . . . Predicted Pattern of Means High High Approval Approval of of Aggression Aggression Low competing information Low competing information High competing information . . High competing information Low High Low High Exposure to TV violence Exposure to TV violence Lower Class Middle Class . . . . . . Obtained Pattern of Means Figure 2 Comparison of predicted and obtained cell means for the approval of aggression scale. 41 however, reveals that the pattern of obtained means did not follow the predicted pattern. In order to help interpret these results, a 2X2 analysis of variance was performed on both the lower and middle class groups. The results of this analysis are in Table 2. For the middle class chil- dren, a significant main effect was found for the family attitude variable. The TV violence variable was signifi- cant at the .10 level. The interaction of these two vari— ables was significant with those children who were highly exposed to TV violence and whose families provided little counter-information to violence showing the greatest appro- val of aggression. Among the lower class children, however, only the family attitude variable yielded significant differences. There was no significant main effect due to exposure to TV violence nor was the interaction significant. To summarize, in neither social class was there a significant relationship between exposure to TV violence and approval of aggression. Both lower andfmidd E class QEQQRS “_. "‘u~..——..-u--_" a“... digwshgwwamsignificant relationship between family attitudes toward aggression and approval. Further, social class .— ““""“"‘-v-.._..~.- - -. 42 affected the waymtheweimflgiyariahlefiiinteracted. Among middle class children, those highly exposed to TV violence from families where anti-violence-was not stressed showed the most acceptance of aggression. This pattern of inter- action was not present among the lower class children. Willingness to Use Violence The hypotheses for this variable followed the same pattern as those in the previous section, More willingness to use violence ought to be evidenced by children who: ti “‘THZb) are (Hlb) are frequently exposed to TV violence; and not provided by their families with counter- information to violence. Further, more 2:5(H3b) the ” and ‘\ "t (H4b) the and willingness to use violence should stem from: interaction of high exposure to TV violence low social class; and, interaction of high exposure to TV violence low counter information to violence, 43 TABLE 2 Three-way Analysis of Variance for Approval of Aggression Scale Cell Means (The higher the score, the more approval of Aggression Scale) Exposure Middle Class to TV Family attitudes violence toward aggression Low approval Undefined Low 14.13 15 03 (n=47) (n=60) High 14.14 16.52 (n=57) (n=52) ANALYSIS OF VARIANCE TABLE Source of variation MS _93_ Exposure to TV violence 4.0 1 Family attitudes 223.0 1 Social class 54.0 1 TV violence X Family attitudes 28.0 1 TV violence X Social class 24.0 1 Social class X Family attitudes 1.0 1 Violence X Family X Class 30.0 1 Error 6.74 426 433 *Omega represents the proportion of the variance explained by each possible source of variation in Lower Class Family attitudes toward aggression Low approval Undefined 15.29 (n=62) 14.68 (N=50) F p. 0.59 n.s. 33.14 .0005 7.86 .025 4.15 .05 3.56 .10 0.01 n.s. 4.46 .05 16.65 (n=40) 16.17 (n=66) Omega* .07 .02 .01 .01 .01 the dependent variable 44 Finally, maximum willingness to use violence should result from: ./ I \\\KH5b) the interaction of high exposure to TV violence, low counter— information about violence, and low social class. The results of the three-way analysis of variance are in Table 3. The first hypothesis predicting that chil- dren who are frequent watchers of TV violence ought show more of a willingness to use violence was supported. The main effect for the exposure variable was significant at the .05 level. The high TV violence group scored higher overall than did the low Violence group. Family attitudes again showed a strong relationship with willingness to use violence in the direction predicted by Hypothesis 2b. Children from families who are explicitly against violence score lower than do children from families whose attitudes toward aggression are undefined. The main effect for social class was also signifi- cant with children from lower class homes tending, on the whole, to be more willing to use violence. 45 The predicted interaction between exposure to TV violence and family attitudes was also significant. For both social classes, children who were frequent viewers of TV violence and whose families were not anti-violence scored the highest on the willingness to use violence in- dex. The interaction between violence and social class, however, was not significant. The predicted third order interaction was significant at the .10 level. The two-way analysis for each social class is help- ful in interpreting the above findings (Table 4). Among middle class children, the main effects for viewing violence and for family attitudes as well as the interaction between the two are all significant. Among lower class children, however, only the main effect for the family attitude var- iable reaches significance. It would seem that the signif- icant differences in the three-way analysis of variance for the interaction between TV violence and family attitudes are due primarily to differences that exist among middle class children. 46 TABLE 3 Two-way Analysis of Variance by Social Class for Approval of Aggression SCale Middle.C1ass ANALYSIS OF VARIANCE TABLE Source of variation MS _§£ F _p Exposure to TV violence 21.75 1 3.27 .10 Family attitudes 146.00 1 21.95 .0005 TV violence X Family attitudes 29.25 1 4.39 .05 Error 6.65 212 Lower Class ANALYSIS OF VARIANCE TABLE Source of variation MS df F pA Exposure to TV violence 5.00 l 0.73 n.s. Family attitudes 107.82 1 15.82 .005 TV violence X Family attitudes 0.20 1 0.00 n.s. Error 6.83 214 47 TABLE 4 Three-way Analysis of Variance for Willingness to Use Violence Scale Cell Means (The higher the score, the more willingness to use violence) Exposure Middle Class Lower Class to TV Family attitudes Family attitudes violence: toward aggression toward aggression Low approval Undefined Low approval Undefined Low 7.27 7.70 7.77 8.42 (n=47) (n=60) (n=62) (n=40) High 7.28 8.60 7.64 8.53 (n=57) (n=52) (n=50) (n=66) ANALYSIS OF VARIANCE TABLE Source of variation MS _§f- F __p___ nggg_ Exposure to TV violence 6.9 1 4.06 .05 .01 Family attitudes 68.5 1 40 34 .0005 .08 Social class 16.0 1 9.41 .005 .02 TV violence X Family attitudes 12.2 1 7.18 .025 .01 TV violence X Social class 1.2 l 0.70 n.s. - Social class X Family attitudes 0.1 1 0.10 n.s. - Violence X Family X Class 5.2 1 3.06 .10 .01 Error 1-71 426 Total 433 48 In s um, boothsocialiclasses-showedisignifi cant-wre- liatiqnsh198~~6e tween. familyai ti.tu.daS--towar.~d-.aggression... and willingnesswtowusewaggression. Among midle class children, those who were more frequent viewers of TV violence re- ported more willingness to resort to violence. Further, the interaction of high exposure and little offsetting in- formation from the family led to the highest level of will- ingness. These latter findings did not appear among lower class children. Use of Violence in Conflict Situations The hypotheses for this variable paralleled those of the preceding section. Violent solutions ought to be suggested more often by children who: (ch) are frequent viewers of TV violence; and, (H20) are not provided by their families with counter— information about violence. Further, this tendency to suggest Violence as a solution to a problem ought to be enhanced by: 49 (H3c) the interaction of high exposure to TV violence and low social class; and, (Hue) the interaction of high exposure to TV violence and low counter—information about violence. Finally, more violent solutions ought to be prompted by: (HSC) the interaction of high exposure to TV violence, low counter—information about violence, and low social class. Table 5 indicates that the predicted relationship between exposure to TV violence and the tendency to suggest violence as a means to solve problems was not present. EXposure to little counter—information about violence, however, was significantly related to using violence more often to solve problems as predicted by Hypothesis 2c. Again, the effect of social class was evident with lower class children suggesting violent means of problem solution more often than middle class children. There was no significant interaction between TV violence and family attitudes nor between TV violence and social class. The second order interaction was significant and predicted by Hypothesis 5c. An examination of the means, however, reveals that the cell means di( 50 TABLE 5 Two-Way Analysis of Variance by Social Class for Willingness to Use Violence Scale Middle Class ANALYSIS OF VARIANCE TABLE Source of variation MS df F p Exposure to TV violence 6.80 l 4.00 .05 Family attitudes 41.60 1 24.54 .0005 TV violence X Family attitudes 10.60 1 6.24 .025 Error 1.70 212 Lower-Class ANALYSIS OF VARIANCE TABLE Source of variation MS _d£___;§ ._JE__ Exposure to TV violence 0.40 1 0.23 n.s. Family attitudes 31.89 1 18.71 .0005 TV violence X Family attitudes 0.70 l 0.41 n.s. Error 1.71 214 .511 High "19” Use of . violence in; C°mpfatm9 in conflict n ormation situations , ‘ Low competing High competing informatio information High competing information Low High Low High Exposure to TV violence Exposure to TV violence Lower Class Middle Class . . . . . . . . Predicted Pattern of Means . . . . . . . . . . High 4 High Use of violence in conflict - . situations To" competing information High competing Low competing information information High competing information Low High Low High Exposure to TV violence Exposure to TV violence Lover Class Middle Class ... Obtained Pattern of Means . . Figure 3 Comparison of predicted and obtained cell means for the use of violence in conflict situations scale. 52 not fall into the expected pattern. In other words, expo- sure to violence, the family‘s attitudes toward violence, and social class all interact together to influence a child's tendency to suggest violence as a means of conflict resolution. The pattern of interaction among these three variables, however, did not achieve the expected configur— ation. The two-way analysis of variance in Table 6 for each social class shows that the results correspond to that found for the approval of aggression scale. Neither social class showed a significant main effect for the TV violence variable. The family attitude variable is a strong predictor in both groups of children. Further, among middle class youngsters, the joint operation of the family attitude variable with exposure to TV violence re- sulted in the expected pattern of means. Middle class children who watch a great deal of violence and who come from families where their parents have not communicated their feelings about aggression are significantly more likely to offer violence as a means to solve problems. This was not so among lower class children where family attitudes was the only consistent predictor. 53 TABLE 6 Three-way AnalYSlS of Variance for Use of Violence in Conflict Situations Scale Cell Means (The higher the score, the more often the child uses violence to solve conflicts) Exposure Middle Class Lower Class to TV Family attitudes Family attitudes violence: toward aggression toward aggression Low approval Undefined Low approval Undefined Low 4.57 4.65 4.85 5.25 (n=47) (n=60) (na62) (n=40) High 4.49 5.26 4.78 5.10 (ns57) (n=51) (n=50) (n=66) ANALYSIS OF VARIANCE TABLE Source of variation MS df F g) QEEEE Exposure to TV violence 2.0 l 2.08 n.s. - Family attitudes 14.0 1 14.58 .005 .03 Social class 7.0 l 7.29 .025 .02 TV violence X Family attitudes 2.8 l 2.96 n.s. - TV violence X Social class 2:2 1 2.29 n.s. - Social Class X Family attitudes 0.5 l 0.19 n.s. - Violence X Family X Class 4.5 l 4.68 .05 .01 Error 0.96 425 Total 432 54 Perceived Effectiveness of Violence It was hypothesized that violence ought to be per- ceived as being more effective by children who: (Hld) are frequent viewers of TV violence; and, (H2d) are not provided by their families with counter- information about violence. Further, this tendency to see violence as effective ought to be increased by: (H3d) the interaction of high TV exposure and low social class; and, (H4d) the interaction of high exposure to Violence and low counter-information about violence. Finally, seeing violence as effective ought to be further enhanced by: (H5d) the interaction of high exposure to TV violence, low counter-information about Violence, and low social class. Table 7 contains the three-way analysis of variance for this index. The results support Hypothesis ld. Chil- dren who are frequent viewers of violence are significantly 55 TABLE 7 Two-Way Analysis of Variance by Social Class for Use of Violence in Conflict Situations Scale MiddleVClass ANALYSIS OF VARIANCE TABLE Source of variation MS df F _p Exposure to TV violence 2.15 l 2.77 n.s. Family attitudes 9.65 l 12.39 .005 TV violence X Family attitudes 7.20 l 9.26 .005 Error 0.77 211 Lower Class ANALYSIS OF VARIANCE TABLE Source of variation MS df F _;2_ Exposure to TV violence 0.05 l 0.04 n.s. Family attitudes 6.73 1 5.90 .025 TV violence X Family attitudes 0.72 l 0.63 n.s. Error 1.14 214 56 more likely to believe that violence is effective as a prob- lem solver than are children who View TV violence infre- quently. Hypothesis 2d is also supported. Children who come from families that are anti-violence see aggression as being less effective than children from families where the prevailing attitude toward violence is less clear. The effect of social class is again apparent as lower-class children are more likely to see violence as effective. None of the predicted first order interactions nor the second order interaction were significant. The two-way analysis of variance for each social class parallels the above findings. In each economic level, both violence and family attitudes show a significant effect while the inter— action is non-significant (Table 8). 57 TABLE 8 Three—way Analysis of Variance for Perceived Effectiveness of Violence Scale Cell Means (The higher the score, the more violence is seen as being effective) Exposure Middle Class Lower Class to TV Family attitudes Family attitudes violence: toward aggression toward aggression Low approval Undefined Low approval Undefined Low 7.83 9.22 8.68 10.90 (n=47) (n:60) (n=62) (n=40) High 8.67 11.08 9.54 11.50 gnz57) (n=52) (n=50) (n=66) ANALYSIS OF VARIANCE TABLE Source of variation MS _d_f F _p__ Q_Ir£g_a_ Exposure to TV Violence 135.0 1 15.79 .005 .03 Family attitudes 416.0 1 49.81 .0005 .10 Social class 93.0 1 10 93 .005 .02 TV violence X Family attitudes 12.0 1 1.44 n.s. - TV violence X SOCial class 0.5 l 0.01 n.s. - SOClal class X Family attitudes 5.5 l 0.64 n.s. - Violence X Family X Class 12.0 1 1.44 n.s. - Error 8.55 426 Total 433 58 TABLE 9 Two-way Analysis of Variance by Social Class for Perceived Effectiveness of Violence Scale Middle Class ANALYSIS OF VARIANCE TABLE Source of variation MS _df F p Exposure to TV violence 78.21 1 9.28 .005 Family attitudes 194.50 1 23.07 .0005 TV violence X Family attitudes 14.29 1 1.69 n.s. Error 8.43 212 Lower Class ANALYSIS OF VARIANCE TABLE Exposure to TV violence 67.14 1 7.74 .025 Family attitudes 227.90 1 26.33 .0005 TV violence X Family attitudes 0.96 1 0.11 n.s. Error 8.67 214 CHAPTER IV DISCUSSION Summary This study examined the influences of exposure to TV violence, perceived family attitudes toward violence, and socio-economic status on the socialization of aggres- sion attitudes. 9.11. The respondents were 434 boys from ages Support was found for the following hypotheses: (1) Young boys from families where attitudes toward aggression were undefined were: (a) more (b) more (c) more to a (d) more tive approving of violence; willing to use Violence; likely to suggest violence as a response conflict situation; and, apt to believe that violence was an effec- way to achieve some goal than boys who re- ported their families to be against violence. (2) Children who were frequently exposed to TV violence differed from boys not so regularly exposed in the following manner. They were: 59 60 (a) more willing to use violence; and, (b) more likely to perceive violence as effective. (3) The interaction of high exposure to TV violence and little exposure to counter-information about violence from the family led to: (a) more approval of violence; and, (b) more willingness to use violence. The hypothesized second order interaction among the three antecedent variables was significant for two of the dependent measures--approval of aggression and suggesting violence in response to conflict situations. For a third variable--willingness to use violence—-this interaction showed a strong similar tendency (p < .10). In each case, the obtained pattern of means indicated that exposure to TV violence and low exposure to counter-information produced the greatest degree of acceptance of violence among middle class children. These two variables did not interact sig- nificantly among the lower class group. We had expected that both social classes should demonstrate an interaction between these two variables with the interaction being more pronounced among lower—class boys. (l) (2) (3) 61 The following hypotheses were not supported. Children with a high degree of exposure to TV vio- lence should: (a) (b) The and show more approval of violence; and, suggest violence more often as a response to conflict situations. interaction of high exposure to TV violence little exposure to counter-information about violence from the family should lead to: (a) (b) The low (a) (b) (C) (d) more suggestions of violence in conflict situa- tions; and, more tendency to perceive violence as effective. interaction of high exposure to TV violence and socio—economic status should lead to: more approval of Violence; more willingness to use violence; more suggestions of violence in conflict situa- tions; and, more tendency to perceive violence as effective. 62 TABLE 10 Summary of the Probability Values of F for Each Dependent Variable Approval Willing- Use of Perceived Antecedent of Vio— ness to violence. _ effective- . lence use Vio~ in conflict ness of Variables . . . lence Situations Violence ------------- Dependent variable---—--—-----—- Exposure to TV violence n.s. .05 n.s. .005 Family attitudes .0005 .0005 .005 .0005 Social class .025 .005 .025 .005 TV violence X Family .05 .025 n.s. n.s. TV violence X Class .10 n.s. n.s. n.s. Family X Class n.s. n.s. n.s. n.s. Violence X Family X Class .05 .10 .05 n.s. 63 Discussion A discussion of these results first should pOint out some of the limits of this research. Replication with chil- dren from more varied backgrounds is necessary before claims of wide generalizability could be made. Secondly, these data do not demonstrate a pattern of causation. Research dealing with the "effects" of tele- vision implies some causal relationship. Moreover, the an- alysis of variance design used is commonly associated with ' experimental settings where statements about "cause" and "effect" can be made with more confidence. This research does not permit us to make such inferences. Measurement problems might also be involved with the exposure to TV violence variable. Tabulating the number of shows which others have judged to be violent obscures other factors which might be important in understanding the ef— fects of exposure to violence. For example, it is possible that a child might be receiving an anti-violence message from what were labeled violent shows. Moreover, using an entire program as the unit of analysis does not allow us to specify what it is about the message that is having an 64 effect. Further, a satiation effect might make the impact of TV violence decline with age. By the time a child is ten years old he may have already seen so much TV violence that its effects may-become less evident as he gets older. In other words, there may be a critical period during which media violence is most influential. At best, the measure of TV violence is a rough approximation. The nature of the dependent measures also require some reservations. Measures of aggression in general, and self- reported measures in particular, are still.in a rough stage of development (McLeod, 1970). Considerably more needs to be done in the refining of the reliability and validity of such scales. Moreover, although the F values associated with the ante— cedent variables indicate that these results are probably not due to chance, an examination of the omega coefficients shows that the amount of variance accounted for by these variables is rather small. At best, the three variables explain about 15 per cent of the variance of the dependent variable labelled perceived effectiveness of violence. In other words, we may conclude that while an association does “as exist between the antecedent and dependent variables, the magnitude of this association is not sizable. The bulk of the variance remains unexplained. Finally, this research was concerned with the atti— tudes children reported toward violence. No data were gathered about the actual behaviors of the boys in the sample. The results indicate that, given certain condi- tions, exposure to TV violence is associated with more favorable attitudes toward aggression. Whether favorable attitudes toward aggression will prompt a child to engage in more aggressive behavior requires additional research. No support was found for one of the major hypothe— ses of this study—-that the interaction of high exposure to TV violence and low exposure to negative information about violence ought to produce the most positive attitudes about violence among less advantaged children. This lack of support may be due to both operational and theoretical considerations. 66 On-the operational level, the dependent variable scales may not have been sensitive enough to detect any further increment in attitudes toward Violence among the low-income group. The lower class boys scored signifi- cantly higher on all four dependent measures than did the middle class group, thus exhibiting a possible "ceiling effect." Given that low-income youngsters were already at a rather high point on each measure, their potential for an increase as measured by the items was limited. If the measures were less accurate at higher levels of approved aggression, further approval by the low-income sample could have gone undetected. At the theoretical level, there are two major rea- sons why the predicted results may not have been obtained. The first relates to a direct vs. indirect socialization effect. Compared to middle-class children, low-income youngsters are more likely to have had more personal exper- iences with violence. Perhaps the lower class child is more influenced in his attitudes toward violence by his direct experiences with it. In contrast, a middle-class youngster, having relatively less direct experience with 67 violence, may be more prone to the influence of his family or television, which pass this information on to him second— hand. If the attitudes of a lower class child have been more fixed by personal experience, TV violence may have little additional effect on him. In other words, if the child already brings with him a set of well-defined atti- tudes about violence to television, TV's effects may be minimal. Second, in both social class levels, perceived family attitudes were strongly related to the child's atti- tudes. The conceptual framework which led to the hypothe— ses suggested that the family should not be as effective a socializing agent among low—income children since family interactions with children are generally less frequent, haphazard, and inconsistent. Where aggression is concerned, however, these generalizations may not hold. Findings from child psychology regarding social class differences in the way parents socialize aggression are inconsistent (Maccoby gt_§l,, 1954; Sears et a1., 1957). The present data sug— gest that the family is at least as effective in lower-class families as in middle class homes. If true, then the 68 situation among lower class children which led to these particular hypotheses may not have existed. Implications Having explored the principal weaknesses of the study, let us turn to its strengths. The following discus- sion is based on two key assumptions. It assumes that the child perceives TV violence in much the same way as the re- searchers who derived the propositions about the world of television violence from content analyses. Secondly, it assumes a pattern of causation which suggests that televi— sion is shaping what the child thinks about violence; it is influential in transmitting attitudes to the child and as such is producing an effect. In this sense, the dis- cussion goes beyond the actual data contained in this re- port. Let us begin at a theoretical level to suggest ten- tative answers to Wright's questions which prompted this study. A. What is the relative strength of television com- pared to other agencies of socialization? 69 In terms of the socialization of aggression atti- tudes, it would appear that among the three antecedent variables the family plays the strongest role in influenc- ing how a child thinks about violence. Only for the per- ceived effectiveness of violence scale did both family attitudes and exposures to TV violence exert separate effects. In general, however, it seems safe to say that the family supercedes television in the socialization of. aggression, but it would be unwise to conclude from this that TV is therefore unimportant. B. What differences might result if the major part of the socialization process is carried out by television instead of by more primary agents? Again, with regard to the socialization of aggres- sion, when family attitudes about aggression are unknwon by the child, young boys who are highly-exposed to TV vio- lence tend to endorse the norms of televised violence. In short, certain children display particular patterns of attitudes which suggest that they may have been socialized by what they've seen on TV. These children's attitudes indicate that they accepted the prevailing attitudes of 70 the socializing agent; in this case, they seem to endorse the lessons contained implicitly in televised violence. They thought it was more effective, more approved, and were more likely to use violence than were other children. A further word should be said about the perceived effectiveness of Violence scale--the only variable to show a significant effect due to exposure to TV Violence when family attitudes toward violence were negative. It may be that the family has not given the child information which ran counter to what he saw on TV concerning this aspect of violence. If such information were made expli— cit, perhaps this effect might not be found. In any case, this finding again suggests that television may be an im— portant source from which the child learns about various aspects of violence. It might also be noted that if TV can teach a child that violence is effective, there is no reason why it cannot also teach him that other non- violent forms of problem-solving are equally as effective. On a more practical level, this research has im— plications for both parents and TV programmers. One main finding illustrated the strong influence of the family on 71 the child's attitudes toward aggression. Parents who make explicit to their children their negative feelings toward violence will probably find that their children also hold Violence in low esteem. Families who are concerned with the possible effects of TV violence on their children can find some assurance in the results of this study. It appears that watching TV violence will probably have no great effect on a child who has already acquired a set of values which are antagonistic with what he sees on TV. It is possible for the TV industry to look upon these findings as both good and bad news. In one sense they indicate that any influence that TV violence may have on children is probably mitigated by the teaching of more primary agents. The family can probably offset any lessons that the child picks up about violence from television. 0n the other hand, this study also suggests that there are certain types of children who possess the potential of be- ing more affected by TV violence. If this is the case, then perhaps some alterations in the treatments of Violence on television may be in order. 72 Besearch Extensions All survey research examining socialization will have to deal with a common problem. The nature of this research indicates that some multi-variate analysis scheme is needed. An investigator would primarily be interested in knowing about the relative influence of the media in relation to the family, the peer group, the school, and perhaps other agencies. And, as the present study shows, there are probably several interactions taking place. Multiple regression is inappropriate for this situation. Another technique, used here, involves setting up analy— ses of variance designs that look directly for interac- tions. This method, however, is difficult to analyze and interpret with a large number of antecedent variables. Further, dichotomizing or trichotomizing respondents into large groups, a common practice with this type of design, also represents a loss in precision. Other techniques to handle a large number of antecedent variables (Morgan and Sunquist, 1963; Lewis, 1962) may show promise. At any rate, subsequent research dealing with the interactions 73 of several socialization.agencies and their joint effects on socialization will have to come to grips with this problem. One direction that future studies might take would be to assess TV's role in the socialization of aggression against different sets of socializing agents. How does television operate in conjunction with peers or with the school, for example. TV's influence in relation to other agencies may fluctuate throughout the child's life. Per- haps there is an optimum age at which TV is most influen- tial. The current study examined the nature in which television may influence various aspects of a child's attitudes. Subsequent studies might define those situa- tions in which socialization by television affects the childfls behavior. What conditions, for example, will make it likely that the child will imitate behavior learned from television? Finally, another type of research suggested by this study would examine the influence of different as- pects of TV content on other areas of socialization. For 74 example, Catton (1970) suggests that perhaps the televi- sion commercial is socializing children into accepting a norm of immediate gratification of desires and is discour- aging the norm of delayed gratification. Hollander (1970) presents data that suggest that television may be the most. influential source in shaping young people's attitudes about war. The role that the various "family shows" on TV are playing in instructing children about the "proper" behavior of family members might be another area for in— vestigation. BIBLIOGRAPHY Allinsmith, B., cited in D. Miller and G. Swanson, Inner Conflict and Defense, New York: Holt, Rinehart, and Winston (1960), p. 315. Buss, A. B., and Ann Durkee, "Assessing Different Kinds of Hostility," Journal of Consulting Psychology, Vol. 21 (1957), PP. 343-348. Catton, William, "Value Modification and the Mass Media," ‘ 'Iflxin Robert Baker and Sandra Ball, Mass Media and Violence, U. 8. Government (1970). ”BlkiEL_Frederick, The Child and Society, New York: Random ’ House (1960). Gerbner, George, "Dimensions of Violence in Television g/waw- Drama," paper prepared for the National Commission on the Causes and Prevention of Violence-(1968). "The Case for Cultural Indicators," paper delivered to American Political Science Association, September (1969). Gerson, Walter, "Mass Media Socialization Behavior: Negro- White Differences," Social Forces, Vol. 42 (1966), pp. 40-50. Greenberg, Bradley S., and Joseph Dominick, "Race and So- cial Class Differences in Teen-Agers Use of Tele- vision," Journal of Broadcasting, Vol. 13 (Fall, 1969), pp. 331-44. and , "Television Behavior among Disad- vantaged Children," paper presented to the Inter- national Communication Association (1970). 75 76 and Thomas Gordon, "Critics' and Public Percep- tions of Violence in TV Programs," Michigan State University (June, 1970). (Mimeographed.) Hicks, D., "Imitation and Retention of Film-Mediated Peer -~m»mwand Adult Models," Journalof Personality.and-So- cial Psycholggy, Vol. 2 (1965), pp. 97-100. Himmelweit, H., A. N.—Oppenheim, and P. Vince, Television ”Wand the Child, New York: Oxford University Press (1958). Hollander, Neil, "Adolescents and War: The Sources of Socialization," paper presented to the Associa— tion for Education in Journalism (1970). Larsen, 0. N., L. Gray, and J. Fortis, "Goals and Goal Achievement Methods in Television Content: Models for Anomie?" Social Inquiry, Vol. 33 (1963), pp. 180-96. Maccoby, B., "Effects of the Mass Media," in M. Hoffman and L. Hoffman, Review of Child Development Re— search, New York: Sage Foundation (1964). and P. Gibbs, "Methods of Child Rearing in Two Social Classes," in W. Martin and C. Stendler, Readings in Child Development, New York: Har- court, Brace and World (1954). McLeod, J., Steven Chaffee, and H. S. Eswara, "Family Communication Patterns and Communication Research," paper presented to the Association for Education in Journalism (1966). , __ , and Daniel Wackman, "Family Communi- cation: An Updated Report," paper presented to the Association for Education in Journalism (1967). McLeod, J., "Children and Television Violence,‘ unpublished 73paper (1970. 77 Minuchin, S., B. Braulio, J. Gurney, B. Risman, and F. Schumer, Families of the Slums, New York: Basic Books (1967). Morgan, J., and Sunquist, J., "Problems in the Analysis of Survey Data," Journal of the American Statisti- cal Association, Vol. 48 (June, 1963), pp. 415—34. Sakuma, Arline, "Values and the Mass Media," unpublished paper, Syracuse University (1968). Schramm, W., Lyle, J., and Parker, E., Television in the Lives of Our Children, Stanford University Press (1961). Sears, R. R., E. Maccoby, and H. Levin, Patterns of Child ,,,»« Rearing, New York: Row, Peterson (1957). Sears, R. R., "Relation of Early Socialization Experiences ‘r”' to Aggression in Middle Childhood," Journal of Abnormal and Social Psychology, Vol. 63 (1961), pp. 466-92. Sewell, W. H., P. H. Mussen, and C. W. Harris, "Relation- ships among Child Training Practices," American Sociological Review, Vol. 20 (1955). pp: 136-48. Siegal, A., "The Influence of Violence in the Mass Media upon Children's Role Expectations," Child Develop- ment, Vol. 29 (1958), pp. 35-56. Stempel, G., unpublished study cited in B. Greenberg, "The "7 Content and Context of Violence in the Mass Media," paper prepared for the Commission on the Causes and Prevention of Violence_(l968). Troldahl, V., "Occupational Prestige Scale," Department of Communication, Michigan State University (1967). (Mimeographed.) 78 U. S. National Commission on Civil Disorders, Report of the National Commission on_Civi1 Disorders, New York: Bantam (1968). U. S. Senate, Television and Juvenile Delinquency, U. S. Government Printing Office (1955). U. S. Senate, Effects on Young People of,Violence and(Crime as Portrayed on Television, U. S. Government Print- ing Office.Xl963). Wright, Charles, Mass Communication: A Sociological Approach, New York: Random House (1959). APPENDIX QUESTIONNAIRE WE'RE TRYING TO FIND OUT WHAT YOUNG PEOPLE THINK ABOUT A LOT OF DIFFERENT THINGS ..... Today we'd like to ask you some questions about what you watch on television, what you do with your friends, and how you feel about some other things. Parents are often asked for their opinions and we think it is about time somebody asked young neonle what they think. This is not a test...so there are no right or wrong answers. Your answers « will not be shown to your principal or your teachers. What we want is your honest opinion on each question. So that your answers are kept private, please do not even put your name on this booklet. Please follow along with us as we read each question out loud. If you have a question or don't understand something, please raise your hand. Don't go on to the next question until everyone has finished. ' We will ask you to write down a few things. Don't worry about perfect spelling...but do try to write or print so that we can read your answer. Remember, you don't need to tell us who you are. Thank you for helpinv us. We appreciate it very much. -2- WE WILL READ OUT LOUD AND YOU CAN FOLLOW ALONG WITH US. AFTER WE HAVE READ A QUESTION YOU WILL HAVE TIME TO THINK ABOUT YOUR ANSWER AND THEN TO MARK IT DOWN. IF YOU HAVE ANY QUESTIONS, PLEASE RAISE YOUR HAND AND ASK US. 1. What is the name of the TV show you like the best? Yesterday, how1mvf'hours did you watch shows on television. I watched hours and minutes. I‘ On Saturday morning, about how many hours do you usually spend watching TV before noon? I usually watch hours and minutes. Put an "x" next to each show you watch every week or almost every week: ___Land of the Giants ___Virginian ___fBI ___Daniel Boone ___Nannix ___Laugh In ___Pamily Affair ___Uod Squad ___Mission Impossible ___Then Came Bronson _~_Get Smart ___Beverly Hillbillies .__Bonanza ____Adam - 12 ___It Takes A Thief Name of the Game ___Tom Jones ~__Jim Nabors ___Hawaii Five-O __*Dragnet __*_Lancer ___Ju1ia ___Doris Day ___Bill Cosby Show Gunsmoke Ironsides Dark Shadows Perry Mason -3- NEXT, HERE ARE SOME QUESTIONS ABOUT YOU AND YOUR FRIENDS: 1. Suppose you got into a fight with someone. How do you think your friends would feel about it? (CHECK ONE ONLY) They would think it was bad. They wouldn't say anything about it. They would think it's OK. I don't know what they would think. 2. Suppose somebody slapped you. What do you think your friends would want you to do? (CHECK ONE ONLY) They would want me to walk away. I'm not sure what they would want me to do. They would want me to slap him back. 3. Suppose that you had been fighting in school. What do you think your friends would say? (CHECK ONE ONLY) They would tell me it was OK. They would tell me it was a bad thing to do. They wouldn't say anything. I'm not sure what they would say. 4. Suppose you saw some guys fighting each other after school one day. What would you and your friends do? (CHECK ONE ONLY) We would watch the fight. he would break it up. We would cheer on the fighters. We would ignore it. HERE ARE SOME THINGS THAT HAPPENED TO OTHER YOUNG PEOPLE. WE WOULD LIKE TO KNOW WHAT YOU WOULD DO IF THESE THINGS HAPPENED TO YOU. PLEASE TELL US WHAT YOU WOULD HONESTLY DO: 1. Pretend somebody you know takes something that belongs to you and won't give it back. What would you do? (WRITE YOUR ANSWER BELOW) 2. Pretend somebody you know tells lies about you. What would you do? (WRITE YOUR ANSWER BELOW) 3. Pretend that Somebody you know-takes something from you and breaks it on purpose. What would you do? ' (WRITE YOUR ANSWER BELOW) 4. Pretend somebody you know just got a new bicycle and lets everybody in the neighborhood take a ride on it but you. what would you do? (WRITE YOUR ANSWER BELOW) HERE ARE SOME STATEMENTS MADE BY YOUNG PEOPLE. -5- THEN? HERE IS THE FIRST: 1. WHAT DO YOU TPINK ABOUT "It is perfectly natural for boys to want to fight sometimes." What do you think? I agree I'm not sure I don't agree "It's OK for a teacher to hit one of his students." What do you think? I don't agree ,. I'm not sure I agree "It's all right if a man slaps his wife.” What do you think? I agree I'm not sure I don't agree "I see nothing wrong in a fight between two teen-age boys." What do you think? I don't agree I'm not sure I agree "I think it's wrong for a policeman to shoot someone who has What do you think? I don't agree I'm not sure I agree "It's OK with me if two of my friends get into a fight." What do you think? I agree I'm not sure I don't agree "Fighting is one thing I never approve of." What do,you think? I don't agree I'm not sure I agree "I don't think it's right for grown-ups to hit one another." What do you think? I agree I'm not sure I don't agree escaped." ~6- BELOW ARE A LIST OF STATEMENTS WADE BY OTHER YOUNG PEOPLE. WE WANT TO KNOW IF YOU AGREE OR DISAGREE WITH EACH OF THE STATEMENTS. HERE IS THE FIRST ONE: 1. "There is no good reason for ever hitting anyone." agree Do you: or disagree "People who keep on bothering me are asking for a punch in the nose." agree Do you: or disagree "Anybody who says bad things about me is looking for a fight." agree Do you: or __~_disagree "I get into fights about as often as the next person." agree Do you: or disagree "I think fighting is a waste of time.” agree Do you: or disagree HERE ARE SOME THINGS OTHER YOUNGSTERS HAVE SAID. THINK ABOUT WHETHER YOU AGREE OR DON"T AGREE WITH THESE THINGS. THEN PUT AN "X" ON THE LINE BESIDE WHAT YOU THINK. HERE IS THE FIRST ONE: 1. "Sometimes a fight is a good way to settle an argument." What do you think? I agree I'm not sure I don't agree 2. ”The best way to deal/with someone who keeps bothering you is to rough.him up a little." What do you think? I don't agree I'm not sure I agree 3. "Sometimes a fight is the easiest way to get what you want." What do you think? I agree I'm not sure I don't agree 4. "A fight is the best way to settle an argument once and for all." What do you think? I don't agree I'm not sure __:::I agree 5. "Fighting is a good way to get even with somebody you don't like." What do you think? I agree I'm not sure I don't agree u. I \ I NEXT, HERE ARE SOME QUESTIONS ABOUT YOU AND YOUR FAMILY: 1. Suppose you got into a fight with one of your friends. How do you think your parents would feel about it? (CHECK ONE ONLY) They would think it's OK They wouldn't say anything about it They would think it's bad 'm not sure what they would do I l 2. Suppose your teacher told your parents that you had been fighting in school. What do you think-your parents would do? (CHECK ONE ONLY) They would punish me for it They wouldn't do anything They would say it was a good thing to do I don't know what they would do I l 3. Suppose you and your parents were watching a TV show together and one of the people on TV shot another person. What do you think your parents would say? (CHECK ONE ONLY) They would say shooting someone is a wrong thing to do They wouldn't say anything I'm not sure what they would say I 4. Suppose one of your friends hit you. What do you think your parents would want you to do? (CHECK ONE ONLY) They would want me to hit my friend back. They WOuid ......t m to walk away Al'm not sure what they would want me to do 5. Have your parents ever told you that fighting with someone was a bad thing to do? yes no 6. Have your parents ever pun ihed you because you were fighting with someone? yes no 7. Have your parents ever helped you settle a problem you were having with another person? yes no FINALLY, HERE ARE some QUESTIONS ABOUT you.... 1. Are you a boy or a girl? boy girl 2. Do you have any brothers or sisters? I have brothers and sisters. 3. What kind of jobs do your ;&qx:13 have? What sort of work do they do? (For example: "Sales clerk,” ”Runs a gas station," ”Works on a farm," or "Waits on people in a clothing store.") Mother-- Father-- THANK YOU VEPY MUCH... HICHIGQN STQT | \ riwfl’llh 3129300649 LIBRARIES IWIWIWHI . 8475