Effects Of Media Violence Research Articles

The Immediate Effects of Media Violence on Behavior

The purpose of this paper is to review experimental studies and short‐term field investigations designed to assess the influence of violence in the mass media on aggressive behavior in the viewer. Major trends in research over three decades are examined. We also indicate some possible connections between research on effects of media violence and recent developments in cognitive psychology. The evidence is organized according to the theoretical explanations that have been given for the commonly reported correlation between the viewing of violence and aggressive behavior. The paper is divided into three major sections. In the first, evidence from laboratory experiments is reviewed. In the second section, studies carried out in field‐experimental settings are summarized. In the third section, evidence from nonexperimental studies is analyzed. In our conclusion we suggest that the cognitive‐neoassociationist hypothesis serves as the best explanatory model for the overall findings and we indicate how some of the other hypotheses concerning media effects on aggression may be subsumed by it.

Naturalistic Studies of the Long‐Term Effects of Television Violence

Carefully controlled experimental laboratory research has provided valuable evidence about the effects of media violence on the aggressive behavior of viewers in laboratory settings. However, social stimuli such as media violence can produce different effects in laboratory as compared to nonlaboratory settings. Thus, research in laboratory settings may not provide accurate information about the effects of media violence in nonlaboratory settings. Some types of quasi‐experimental field research can provide evidence about the long‐term socialization and developmental effects of media violence on viewer's behavior in naturalistic settings. The present article summarizes the findings from a number of quasi‐experimental studies about the effects of naturally occurring media violence. We conclude that the balance of the findings are consistent with the hypothesis that television produces a long‐term increase in the aggressive behavior of boys but not of girls.

The Effects of Media Violence on Attitudes, Emotions, and Cognitions

This review specifies mediating factors between the viewing of violent media and aggressive behavior. The role of cognitive and emotional arousal processes are discussed, as is the interplay among these factors and attitudes toward aggression. Included in these discussions are attention and comprehension of the viewed media, as well as attributions and moral evaluations of the perpetrators of violence. Finally, the effects of media exposure on arousal, emotional desensitization, and the excitement of the observer's aggressive inclinations are presented.

Television violence and aggression: The debate continues.

VioLit summary: OBJECTIVE: The purpose of this study by Friedrich-Cofer and Huston was to provide a literature review and critique of the debate over the relationship between viewing violence on television and actual aggression and violence. This study supported the majority of the literature which claims that a relationship does exist and discussed the theoretical and social policy implications of this relationship. METHODOLOGY: This literature review was divided into two main areas of research: experimental studies (both laboratory and field studies) and correlational studies (primarily longitudinal studies). Within each of these areas, methodological issues were addressed, as well as results, trends and potential sources of bias. FINDINGS/DISCUSSION: Laboratory studies have had high internal validity because of their random sampling techniques and their ability to manipulate the independent variables. However, three major questions were raised about their external validity: the degree to which the stimuli used reflected actual television programming; the degree to which their artificial settings lead to experimenters' influencing the subjects' responses; and the degree to which the measures of aggression in the laboratory actually mirror real aggression. This study refuted these questions by explaining that a content analysis of laboratory studies showed that most use actual television programming; that aggression seemed, in most studies, to increase when experimenters' left the room not when they remained; and that a variety of measures were used in studies for aggression, many of which reflected real-life situations. Field experiments were considered a good methodological choice because they utilize the random assignment and experimental manipulation of variables in a natural setting, increasing both internal and external validity. However, questions again were raised in the same three areas of stimuli, setting and measures of aggression: there has been the problem of finding control (nonviolent) stimuli which the subjects would voluntarily watch on a regular basis; most of the settings of field research have been within residential institutions for adolescents where often their television viewing is monitored or controlled, not voluntary; when measuring aggression, problems arose in distinguishing between serious and playful aggression, as well as distinguishing between aggression related to group dynamics (within a group home setting) and individual aggression. Correlational studies have been primarily very large and involving a diverse range of subjects in terms of race, ethnicity and socioeconomic status. They have had high external validity but low internal validity. Longitudinal studies have allowed for analysis across time and have seemed to support a small but consistent positive relationship between violent behavior and violent television viewing. Problems with these studies, however, were pointed out such as the possibility of selection bias; bias caused by other variables not being statistically controlled; or self-reporting biases. This study pointed out, however, that most correlational studies have rigorously controlled for as many other variables as they could and have attempted to find as many alternative measures of aggression to self-reporting as possible. This study claimed that the literature overall supported a bi-directional model of relations between viewing television violence and aggressive behavior. In other words, there appeared to be both a tendency for violent television viewing to positively influence aggressive behavior and a tendency for violent behavior to positively influence the desire to watch violent television shows. AUTHORS' RECOMMENDATIONS: The authors suggested that social scientists must acknowledge that the dialogue about violence on television must broaden to include a discussion on societal values. They also stressed the urgent need for connecting policy issues with theoretical and research issues. (CSPV Abstract - Copyright © 1992-2007 by the Center for the Study and Prevention of Violence, Institute of Behavioral Science, Regents of the University of Colorado) KW - Television Viewing KW - Television Violence KW - Literature Review KW - Media Violence Effects KW - Aggression Causes KW - Violence Causes KW - Child Aggression KW - Child Offender KW - Child Violence KW - Juvenile Aggression KW - Juvenile Offender KW - Juvenile Violence KW - Exposure to Violence Language: en

Television violence and aggression: A rejoinder.

Television violence and aggression: A rejoinder.

Television, fantasy and vicarious catharsis

Reviewers and critics of research on the effects of media violence have rejected the catharsis hypothesis for a lack of empirical evidence. This essay suggests that catharsis research has been based on an unsound formulation of how catharsis functions, ignoring the degree to which viewers are able to fantasize. That ability is posited as being crucial to experiencing vicarious catharsis. A discussion of the construct of television viewing suggests that fantasy is not in the media content but in the person. Assumptions in previous research that (1) catharsis should take place equally for all, and (2) cathartic fantasy will be violent in nature are challenged.

Same Time, Next Year: Aggregate Analyses of the Mass Media and Violent Behavior

VioLit summary: OBJECTIVE: The purpose of this article by Baron and Reiss was to argue that imitation effects attributed to mass media events were statistical artifacts of the mortality data, the timing of media events, and the methods employed in past research. METHODOLOGY: This study was itself a quasi-experimental analysis of aggregate data from other quasi-experimental investigations purporting to show that mass media coverage of violent events caused imitative responses among the public. The authors utilized Phillips's (1983) investigation linking boxing to U.S. suicides, and Bollen and Phillips's (1982) work linking publicized news stories about prominent suicides to total U.S. suicides. Those earlier studies relied almost exclusively on time-series regression methods to explore linkage between media coverage of violent events and real-world respondent behavior. The authors noted that the researchers lacked a theory specifying how many days would transpire prior to media depicted violence inducing real-world violence, and this was accordingly prone to induce Type I errors (falsely rejecting the null hypothesis.) Thus, the authors have deduced that the use of more cautious and appropriate testing procedures would have yielded significantly less definitive conclusions about the effects of prize fights on homicides and would almost certainly have led to a rejection of the claim that television suicide stories cause suicides. To test their conclusions, the authors replicated the previous studies' data by enlarging and extending the set of eliciting stories examined in order to reassess the earlier findings. The authors applied stringent statistical manipulation of Phillips's data on relationships to prize fights and suicides, and concluded that the lacking overall replicability and generality criterion indicated a misspecification argument showing the earlier study was flawed due to a mistaken cause-effect relationship specification theory. The authors also noted that other researchers had extended Bollen and Phillips's analyses of imitative suicide by enlarging or extending the set of eliciting stories examined. However, those studies could not replicate the imitative effects. The authors themselves undertook new analyses aimed at generalizing the imitative hypothesis to a larger set of media events and they too conformed to the authors misspecification hypothesis. FINDINGS/DISCUSSION: The authors found statistically significant evidence indicating that the imitative effects purported by Phillips lacked replicability and generality. Adopting the same specification as Phillips, none of the lag coefficients denoting imitative responses to prize fights were statistically significant. The authors concluded that the previous research examined issues that primarily reflected theoretical underdevelopment and was lacking in a comprehensive or credible account of how, where, when, and why mass media violence affects real-world imitation activity. The authors concluded that current formulations emphasizing imitation or suggestion provided little basis for identifying the aspects of media stimuli that affected behavior, the specific forms of behavior affected, or the social and psychological mechanisms by which imitation occurred. AUTHORS' RECOMMENDATIONS: The authors reiterated that no credible theory of media-induced imitative violence exists. Thus, quasi-experimental studies of aggregate data invite ex-post reinterpretations of what such a theory might really expound. The authors noted their single intent of trying to show that researchers may have prematurely rejected the null hypothesis of no media effects. The authors concluded with a wholehearted concurrence with Bollen and Phillips that future research might focus on micro studies rather than macro analyses. (CSPV Abstract - Copyright © 1992-2007 by the Center for the Study and Prevention of Violence, Institute of Behavioral Science, Regents of the University of Colorado) KW - Violence Causes KW - Media Violence Effects KW - Media Coverage KW - Media Portrayal KW - Sports Viewing-Related Violence KW - Television News KW - Television Violence KW - Television Viewing KW - Televised Sports KW - Adult Violence

Videogames, Television Violence, and Aggression in Teenagers

VioLit summary: OBJECTIVE: The aim of this paper by Dominick was to examine the effects of videogames and television violence upon levels of self-esteem and aggression in adolescents. The study examined both the stimulation theory of viewing violence, which suggests that viewing such scenes could increase the probability of commission of actual violence, and catharsis theory, suggesting that viewing violence reduces levels of aggression by purging the viewer of aggressive feelings. METHODOLOGY: The author employed a quasi-experimental cross-sectional design with a non-probability sample of tenth and eleventh grade adolescents at three high schools in northeast Georgia. Of the 292 children who agreed to participate, 250 survey questionnaires that could be used in the study were returned. Measures included total amount of time spent per week playing videogames at arcades, and the total amount of money spent per week in these activities. Viewing of television violence was measured by how often each subject watched each of 24 programs, 16 of which were rated as above average in violence by the National Coalition on Television Violence. General television viewing was measured by number of hours spent watching television on an average day, as well as the amount of time spent watching on the day before the survey. School performance was measured with average grades received by the subject, and a measure of socio-economic status (SES) was obtained via use of the Occupational Prestige Scale. Aggression was measured with three types of question. The first involved hypothetical responses to various hypothetical situations, ranging from backing out of the situation to fighting. The second consisted of six items that dealt with interpersonal relations with others and manifest physical aggression, and attitudes to fighting. The third measure of aggression was an index of aggressive delinquency, according to the frequency with which the subject engaged in three violent acts - fighting with several people, hurting someone for revenge and fighting with another student. Self-esteem was measured with four items based upon a scale developed by Rosenberg, with subjects agreeing or disagreeing with statements about self-worth. Analysis was conducted both for the overall sample and separately for boys and for girls, and included correlations, examination of frequencies, ANOVA and Chi-Square. FINDINGS/DISCUSSION: The author found that 24% of the girls and 18% of the boys did not play videogames, with average time playing being less than an hour for girls and an hour and a half for boys. Girls spent an average of $1.50 each week on videogames, with only 6% playing alone, whilst boys spent just over $2.00, with 21% attending the arcade by themselves. 36% of the girls' families owned home systems, with 55% of the boys' families having videogames at home - a variable that was negatively related to school performance for boys. Boys viewed more television violence than did girls, although this finding was only weakly significant (p AUTHOR'S RECOMMENDATIONS: The author recommended future research be conducted using subjects from various age groups in different settings with more accurate and thorough measures of aggression. EVALUATION: This study represents an interesting examination of the relationships among videogame playing, television viewing and aggression. However, the use of a sample from one location with subjects of the same age group might have resulted in either an overestimation or an underestimation of the true effects of violent media upon aggressive behavior. The use of self-report data might have led to inaccurate findings, and the measures of aggression might not have covered all types of aggressive tendencies. Internal validity, therefore, might not be of the highest order. However, the study provides a good basis for further research. (CSPV Abstract - Copyright © 1992-2007 by the Center for the Study and Prevention of Violence, Institute of Behavioral Science, Regents of the University of Colorado) KW - Georgia KW - Grade 10 KW - Grade 11 KW - Senior High School Student KW - Late Adolescence KW - Early Adolescence KW - Exposure to Violence KW - Video Game Violence KW - Aggression Causes KW - Juvenile Aggression KW - Media Violence Effects KW - Television Viewing KW - Television Violence KW - Juvenile Self-Esteem

Family Patterns and Television Viewing as Predictors of Children's Beliefs and Aggression

VioLit summary: OBJECTIVE: The purpose of this study by Singer et al. was to examine the ways in which children's family life and TV viewing combine to predict aspects of their conscious experience as well as their social interaction patterns and behavior. METHODOLOGY: A quasi-experimental design was employed using 63 children whose average age was four when data were first obtained in 1977. When final data were obtained in 1982, the average age was close to nine. Initially, the children had been observed during play and rated on their aggressiveness. Parents had maintained detailed TV logs for at least three two-week periods during the previous year. The average weekly TV viewing was 24 hours. In 1980, mothers filled out the Hoffman Child rearing questionnaire which included a series of items about discipline and control, power assertion, physical punishment, and forceful control of the child. Mothers described in detail the family's daily routines, and activities. The measure of parental values was derived from a multitrait-multimethod analysis of self descriptions. The variables hours of sleep and emphasis on outdoor activities were also included. Parents were also asked about the family's use of TV and any restrictions they put on their children. The children were interviewed and tested for cognitive skills, IQ, and knowledge of TV characters. In 1982, the children came to the Yale TV Center for further interviews and to complete a Scary World questionnaire. Dependent variables were beliefs about the world (high scores indicate scary world), physical aggressiveness, behavior in school (high score indicates the child has been disruptive), and motor activity (high score indicates self restraint and lack of restlessness). Multiple regression was used to analyze the data. FINDINGS/DISCUSSION: Multiple regression analysis was used to estimate how much combined family and TV variables could predict the scores on the dependent variables. Belief system showed significant positive correlations with TV environment (p AUTHORS' RECOMMENDATIONS: The authors suggested that parents, educators and the media industry take a serious look at the possible consequences for the early school-age child of unrestrained viewing of violence on TV. EVALUATION: This study has a strength in being a longer term study than most experiments. A sample of 63 is a large enough one to draw correlations from, though the usual weaknesses of experiments do apply here--artificiality, possibility of a narrow population for generalization, and other impact of the experiment itself. The measures hinged chiefly on parental reports--which the authors argue would underestimate occurrence. However, the possibility of sensitization of the parents to the condition of being part of an experiment cannot be overlooked, for whatever effect it may have. The authors argue for a restraint of the child's exposure which, in some respects may work, but still does not get at the heart of the process through which violence on television affects youth. (CSPV Abstract - Copyright © 1992-2007 by the Center for the Study and Prevention of Violence, Institute of Behavioral Science, Regents of the University of Colorado) KW - 1970s KW - 1980s KW - Child Aggression KW - Child Development KW - Television Viewing KW - Television Violence KW - Media Violence Effects KW - Exposure to Violence KW - Family Relations KW - Aggression Causes KW - Child Perceptions

Effects of Realistic TV Violence vs. Fictional Violence on Aggression

Effects of Realistic TV Violence vs. Fictional Violence on Aggression

Film violence and verbal aggression: a naturalistic study.

VioLit summary: OBJECTIVE: The purpose of this study by Sebastian et al. was to examine: (1) the extent of verbal aggression expressed after having watched violent movies; (2) whether or not the results of laboratory experiments on the relationship between viewing film violence and subsequent observer aggression can be generalized to real-life situations; (3) the duration of aggression-facilitating effects of media violence, and (4) the effect of multiple exposure to media violence. METHODOLOGY: This was an experimental static group comparison of 74 males, ages 12-19 years, who were randomly assigned to four cottages at a minimum security penal institution for juveniles. An additional 15 participants lived in the cottages, but were excluded from the study (2 refused, 5 did not cooperate, and the data was lost on the last 8). To ensure comparability of the boys, analysis of their characteristics was carried out. The boys in each of the four cottages were shown commercially available movies: one group saw 5 violent movies, one each night of the week, another saw 5 non-violent movies, one each night of the week, another saw 1 violent movie, and the last group saw 1 non-violent movie. After viewing each movie, the boys rated its content. The next day, after all boys had seen the movies, a trained confederate peer criticized each boy as they put together a puzzle while a trained judge observed the participant's response. The confederate peer either mildly or severely harassed and criticized the boy, depending on the random assignment of the boy to the exposure. Each boy experienced each distraction condition. The level of verbal aggression expressed after watching the movie(s), with and without reinforcing verbal aggression towards the subjects, was measured. The boys perceived the aggressive movies as significantly more violent than the non-aggressive movies (all t-tests were two-tailed and significant at least at the .02 level). The objective analysis of the aggressive content of the movies was also statistically significant for the 5 movie case (t8=4. 59, p FINDINGS/DISCUSSION: (1) Boys who saw the aggressive movie(s) expressed significantly more intense verbal aggression than those who viewed non-aggressive movie(s) (F1,66=3. 72, p=0. 058). (2) Boys who were severely harassed and criticized were more verbally aggressive than those who were mildly distracted (F1,66=34. 95, p AUTHORS' RECOMMENDATIONS: Future research on the effects of multiple exposures would have to control for the viewer's already substantial media exposure and pay attention to the similarity of stimuli and time between exposures. (CSPV Abstract - Copyright © 1992-2007 by the Center for the Study and Prevention of Violence, Institute of Behavioral Science, Regents of the University of Colorado) KW - Media Violence Effects KW - Film Violence KW - Exposure to Violence KW - Juvenile Male KW - Juvenile Inmate KW - Juvenile Aggression KW - Early Adolescence KW - Late Adolescence KW - Inmate Aggression KW - Inmate Studies KW - Incarcerated KW - Juvenile Correctional Institution KW - Aggression Causes KW - Verbal Aggression KW - Male Aggression KW - Male Inmate Language: en

Does television violence enhance program popularity?

Does television violence enhance program popularity?

Facilitative effects of media violence on helping.

The present experiment was undertaken to test the generality of Tannenbaum and Zillmann's arousal-excitation transfer model. 83 female subjects were insulted, praised, or treated in a neutral manner by the experimenter and then watched a neutral, arousing, or aggressive film clip. After viewing the film the subjects were given the opportunity both to aggress against and later to help the experimenter. Insulted subjects were more aggressive but no less helpful. Subjects helped significantly more following the aggressive film than either the arousing or neutral film.

Television Violence and Violent Behavior

Television Violence and Violent Behavior

Children's Viewing of Crime Shows and Attitudes on Law Enforcement

VioLit summary: OBJECTIVE: The purpose of this study by Dominick was to examine relationships between media exposure to and law enforcement and attitudes about police and among elementary school children. It also attempted to bridge the gap between research that has focused solely on mass media socialization and research that has focused only on young people's attitudes about police and crime. METHODOLOGY: During May of 1972, questionnaires were completed by 371 fifth graders in New York City schools. Two schools were in blue collar communities and two were in white collar communities. A content analysis was done to pick crime aired in the 1970-1971 television season. Nothing was mentioned about how the schools, children, or TV sample episodes were chosen. Crime shows were defined as shows that had at least two crimes in the sample episode. Respondents were asked about how often they watched these shows. Scaling went from 1 (never) to 4 (almost always watch). The Crime Show Viewing Index was used as an independent variable, and is calculated by simply summing these weights for the 21 shows. The TV Importance Index was also used as an independent variable. In a pretest, respondents were asked about their most important source of information about police, crime,and law enforcement. Four sources of information (TV, parents, friends, and personal contact with police) were used in a pair comparison technique adapted from White. The number of times TV was chosen as the most important source was then tabulated, with scores ranging from 3 (high importance) to 0 (low importance). Perceived attitudes of friends toward police was also used as an independent variable. This was indexed by three inter-correlated items. Weights were assigned to each option and summed. Parents' attitudes towards police was also used as an independent variable. This measure used two items with revised wording from the Friend's Index. Personal contact with police was used as an independent variable and an index was created from three yes-no questions concerning the respondent's experience with police. The Occupational Prestige scale was also used. This is a 13 step index described by Troldahl. Eight questions were then asked designed to measure attitudes and knowledge about police. These were the dependent variables: identification with TV characters, occupational prestige ranking, evaluation of police, efficacy of police, reality of TV police, perceptions of and police violence, reporting of witnessed crimes, and knowledge of the legal system. For criterion variables measured by nominal methods, independent variables were dichotomized by means of a median split and chi square tests were performed. For ordinal techniques, variables were transformed into ordinal scales and Spearman rhos were calculated. Product moment correlations were used where interval scales were the criterion variables. FINDINGS/DISCUSSION: The inter-correlation among the six dependent variables was generally low. Contact with police was significantly correlated with show viewing (CSV) (p AUTHOR'S RECOMMENDATIONS: The author implied that more research is needed in TV as an information source and other non-TV influences on childrens' attitudes toward law enforcement. EVALUATION: This study begins an interesting exploratory look at the relationship between children's viewing of shows and their knowledge/beliefs about law enforcement. The sample size, 371, provides a large number of respondents, but there was no mention of how the sample was selected. The use of questionnaires by fifth grade school children raises questions of how these were administered. This is not given in the article. Other definitions and selection procedures such as crime and white/blue collar communities raise more methodological questions. The analysis, relying totally on correlation matrices, is associative only and cannot help to establish causal links or sort out inter-item influences. This study was published in the spring of 1974 which makes the findings quite dated. The issues raised are interesting ones which need more sophisticated analysis, a broader respondent base, and more recent replication. (CSPV Abstract - Copyright © 1992-2007 by the Center for the Study and Prevention of Violence, Institute of Behavioral Science, Regents of the University of Colorado) KW - 1970s KW - New York KW - Grade 5 KW - Elementary School Student KW - Late Childhood KW - Exposure to Violence KW - Media Violence Effects KW - Television Viewing KW - Television Violence KW - Child Attitudes KW - Child Perceptions KW - Child Ideology KW - Attitudes Toward Police KW - Law Enforcement

Media violence research: Interpreting the findings

The pages which follow are an attempt to stand back and assess what we know about the effects of media violence in an effort to sort into a usable pattern the results of hundreds of reported studies. Timothy P. Meyer is assistant professor of speech in Mass Communications at the University of Massachusetts, Amherst; James A. Anderson is associate professor of radio‐television and director of the Broadcast Research Center at Ohio University, Athens.

Desensitization of children to television violence.

Desensitization of children to television violence.

Effect of media violence “justification” on aggression

Effect of media violence “justification” on aggression

Effect of exposure to symbolic aggression on aggressive behavior.

VioLit summary: OBJECTIVE: The purpose of this study by Lˆvaas was to examine the effect of exposure to symbolic aggression on the behavior of children. The author hypothesized that exposure to aggression would produce aggressive behavior. METHODOLOGY: Experimental design was used for this study. Three experiments were done. The first experiment was conducted in the Gatzert Institute of Child Development at the University of Washington. Twelve children from the Institute's nursery school were used as subjects. The subjects were above average in intelligence and came from families with fathers in occupations above the means of the general population. They were randomly selected and matched by gender into two groups of six and were observed for three trials spaced about four days apart. Each subject was accompanied to a playroom by an assistant who invited the subject to play some games and look at a movie. The assistant then showed the subject how to operate the dependent measure, a set of dolls that hit each other when a bar was depressed. Then the assistant showed the subject how to operate the movie (the movies were set up to stop at ten second intervals, and only continued when the subject pressed a bar - this insured continued interest in the movie). Subjects were shown one of two films on a 9 by 12 inch screen. The aggressive film was cut to give an almost continual display of aggression inflicted by one figure upon another. The nonaggressive film depicted three bear cubs and a mommy bear engaging in pleasant humanlike play. Running time was five minutes for both films. The subjects were then told to some more with the dolls (dependent measure), and the number of times the bar was depressed was counted via an electronic counter. This was a two minute period of continuous reinforcement where a doll hit the other one every time the bar was pressed. The second and third trials were identical to the first with two exceptions. First, the operant period was eliminated. Second, doll was immediately followed by a one-minute extinction period. One subject from each group refused to return for the third trial. The second experiment used 20 children from 4-6 years of age. These children were chosen from low income families with working mothers. Subjects were randomly matched by age and sex into two groups of ten. The experiment was conducted in a trailer. The same independent stimuli and dependent measures were employed, and the procedure was exactly the same as the first experiment, except that a 4 minute extinction period followed the 2 minute continual reinforcement period. The third experiment used the same subjects from experiment 2 with conditions reversed (the group who had seen the aggressive film now saw the nonaggressive film). The experimental model was modified to minimize the interfering effect of following instructions. During the dependent measure, another toy was placed next to the dolls. This toy was activated the same way that the dolls were (by pressing a lever), only instead of producing an aggressive effect, it simply sent a ball to the top of a cage. The boxes were placed side by side so that subjects could use with both toys simultaneously if desired. The procedure was identical except that subjects were told they could with either toy after viewing the film. ANOVA was used to analyze the data. FINDINGS/DISCUSSION: In the first experiment there were no significant differences between the mean number of responses on the dolls after seeing the films. Despite the increase in sample size in experiment number two, none of the differences on response between the groups were significant. There was a significant interaction between the kind of toy played with and the kind of film preceding the toys in experiment three. The children who watched the aggressive films played with the hitting dolls more than the nonaggressive group (p EVALUATION: This article presents more support for the media-violence argument and, with the focus being on very young children in this study, the very young ages at which children may be affected by violent behavior. With any experiment, artificiality, narrow pool for sample subjects, and relatively small numbers of subjects may cast some validity concerns on these findings. The researcher did try to cope with possible confounding effects of verbal instructions and, thus, tried to maintain methodological integrity. (CSPV Abstract - Copyright © 1992-2007 by the Center for the Study and Prevention of Violence, Institute of Behavioral Science, Regents of the University of Colorado) N1 - Call Number: F-364, AB-364 KW - Washington KW - Media Violence Effects KW - Television Viewing KW - Television Violence KW - Early Childhood KW - Child Aggression KW - Aggression Causes KW - Program-Film Content Language: en