Show Summary Details

Page of

 PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, CRIMINOLOGY AND CRIMINAL JUSTICE ( (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 22 October 2017

Surveys and the Study of Crime, Mass Media, and Popular Culture

Summary and Keywords

The survey, a research methodology in which variables are measured through the answers to questions on a data collection tool called a “survey questionnaire,” has been used to investigate several potential relationships between mass media and crime. In almost all studies, the relationship hypothesized is media influence, in which the media images change the consumer’s attitudes, beliefs, and behavior. Most commonly the hypothesized change is behavioral: individuals who consume violent, sexual, or otherwise suspect media images purportedly have a heightened risk of engaging in criminal acts. Other hypothesized effects of media consumption that are studied frequently include racist or otherwise biased attitudes, which could impact decision making during the criminal justice process; inaccurate beliefs about crime, with an impact on fear of crime and resultant changes in public policy; and the CSI Effect, an exaggerated belief in the efficiency of forensic science in identifying criminal offenders, which could sway jury decisions.

Survey-based research studies have rarely found that mass media consumption has a significant impact on criminal behavior or biased attitude, and its impact on beliefs about crime are contingent on many more significant factors. While these poor results may suggest that there is indeed minimal impact, the problem may lie within the methodology itself, in the ability of survey-based questions and answers to adequately measure the independent variable (media consumption) and the various dependent variables. Additionally, the problem may lie within the essential premise of media influence, with its passive, naive viewer and a radical disjunction between media and real life.

Keywords: crime, CSI Effect, juvenile delinquency, mass media, media influence, moral panic

Surveys in the Study of Mass Media and Crime

The survey is research methodology in which variables are measured through the answers to questions solicited from a large group of respondents through a data collection tool called a survey questionnaire. Questions are typically closed-ended, requiring respondents to choose an answer from a list of two or more. The responses are then quantified and subjected to statistical analysis. For instance, instead of asking “Describe your experience in playing violent video games,” as in an ethnographic interview, a survey questionnaire may ask high school students to determine if they play violent video games frequently, infrequently, or never. The responses, quantified as frequently = 2, infrequently = 1, and never = 0, can then be used as an independent variable in a correlational analysis with a dependent variable such as “engages in aggressive acts (yes = 1, no = 0).” If 78% of respondents who frequently play violent video games engage in aggressive acts, but only 28% of respondents who never play do so, then the researcher can conclude that playing violent video games is a predictor of aggressive acts. It may even be a cause, barring sampling error and antecedent variables that cause both, such as a subculture of violence or lack of parental supervision. Sensitive tests such as the chi-square can measure the statistical significance of a causal link to within a hundredth of a percentage point (Walker, 1999, p. 144; Sirkin, 2006, p. 398; Babbie, 2016, p. 465).

The survey is very adaptable. It can consist of 20 or 200 questions. It can draw respondents from a single classroom, a single school, or an entire nation. It can be administered in several stages over time, and in person, over the telephone, or via email. It can be combined with any other quantitative or qualitative research methodology. Ethnographic interviews are sometimes conducted with a subset of survey respondents to clarify responses and add depth and color to the dry statistical analysis, but probably the most common mixed methodology is the survey experiment, in which questions measuring the independent variable differ in two or more versions of the survey questionnaire, allowing respondents to be divided into control and experimental groups after the fact.

Although there are precursors of the survey in the census, used to determine characteristics of a national population since ancient times, and in the “man in the street” public opinion polls popular in journalism since the 1700s, survey methodology itself dates only from the late 19th century. It was rarely used until the 1930s, when, energized by new developments in statistical analysis, or perhaps hoping to acquire the prestige that physics, chemistry, and biology enjoyed, scholars like Talcott Parsons in sociology, Rensis Likert (who invented the five-point Likert Scale) in psychology, and Harold F. Gosnell in political science petitioned for a quantitative turn in their disciplines. They argued that the quantitative data derived from surveys would be better, more detailed, and more empirically valid than data derived from the established methodologies of observation, interviews, and experiments (Groves, 2011). Paul Lazarsfeld (1901–1976), who founded the Bureau of Applied Social Research at Columbia University, is often revered as the father of survey methodology, as he adapted the survey to research topics as varied as unemployment, presidential politics, and the impact of radio commercials on listeners.

The development of two major statistical computer programs, SPSS in 1968 and SAS in 1973, permitted researchers easy access to sophisticated quantitative analysis methods, such as linear and logistic regressions and probit and tobit models. At the same time, the introduction of national and international surveys, such as the General Social Survey (U.S., every two years since 1972), the National Crime Victimization Survey (U.S., annually since 1973), and the British Social Attitudes Survey (annually since 1981), made thousands of questions and answers available for secondary data analysis. The survey soon became the preferred method in nearly all of the social sciences except psychology, where experiments took precedence. Indeed, some of the most prestigious journals in criminology and sociology accept nothing else, while other journals require that authors explicitly justify their use of non-survey methodologies (King, 1991).

For all of its scientific panache, the survey is not without its methodological problems. In order for the results to be generalizable, applicable to everyone in the population (all children, all citizens of Denmark, all police offices, and so on), a random sampling strategy must ensure that every member of the population has an equal and likely chance of being selected as a respondent. For many populations, such as junior high school students, criminal justice professionals, and fans of violent video games, randomization is unfeasible or impossible (Babbie, 2016, p. 245). Even if a random sample is drawn, professional ethics requires that participation be voluntary, and the percentage of selected respondents who agree to participate varies from 50–60% in telephone surveys to less than 1% in email surveys. Those who agree often have strong opinions on the issue, while those with more moderate opinions opt out.

An additional problem arises in the operationalization, how the variables in the study are measured. What constitutes “frequent violent video game playing,” one hour per week or ten hours per week? One researcher may operationalize crime as “one or more convictions for a serious violent or property crime,” and another, more broadly, as “any violent, property, public order, or drug offense, whether or not it resulted in criminal charges.” Thus, comparing the results of different studies becomes difficult, if not impossible.

The fallibility of the respondents produces another set of problems. Respondents who don’t remember or don’t know an answer will guess anyway. They interpret vague terms like “often” and “very often” in ways far different from what the researcher intends. They mark whimsical responses. They deliberately mark answers that present themselves in the best possible light, a phenomenon known as “desirability bias” (Krumpal, 2013). For instance, on Self-Reports of Offending, juvenile boys often admit to assaults and burglaries that they haven’t actually committed, in an attempt to present themselves as “tough” and “bad” (Simon, 1995; Kim, Fendrick, & Wislar, 2000).

The most serious problem in the survey methodology is the opportunity for researcher bias. The researcher writes the questions and answers based on assumptions about causal links that may or may not be justified. Variable categories can be omitted or emphasized. The wording of questions can push respondents toward specific answers. In a famous example, the General Social Survey regularly asks respondents about a member of an excoriated group, such as a white supremacist, who wants to make a speech in their town. Asked if he “should be permitted,” most respondents say “no.” Asked if he “should be forbidden” from making such a speech, most respondents also say “no.” Even placement in the survey questionnaire can alter responses (McFarland, 1981). Same-sex marriage receives more support when it is placed within a group of questions on civil rights than when it is placed within a group of questions on social problems.

However, these methodological problems are no more severe than those faced by researchers who use experiments or ethnography, and the quantification of the results, with its mathematical rigor and practical generalizability, more than outweighs the drawbacks. Therefore, surveys have been used extensively to investigate the relationship between mass media and crime. Generally, surveys are used to measure media influence, a theoretical model in which violent, sexual, or otherwise problematic media images have a direct, usually negative impact on viewers’ attitudes, beliefs, or behavior (Jewkes, 2004, p. 12; Carrabine, 2008, p. 29).

Risk of Criminal Behavior

By far the most commonly investigated hypothesized media influence is on viewer behavior, especially the risk of physical violence. The hypothesis states that individuals, usually juveniles, who read about physical violence in novels or comic books, watch depictions of it in movies or on television, or pretend to engage in it while playing video games, will experience a change in perception or aggressiveness sufficient to increase their risk of engaging in real-life physical violence. They are not expected to engage in precisely the same gleeful serial killing or grisly acts of dismemberment and cannibalism they see in the media; the real-life violence is typically operationalized as getting into fights or “trying to hurt someone” (see the variable list in Kilpatrick & Saunders, 2009).

Although sometimes a direct imitation of the media image is theorized, as in the case of so-called copycat crimes (Surette, 2002), the main theoretical argument in linking media images with violent behavior derives from social learning theory, first developed by Edwin Sutherland in the 1930s. It argues that individuals, particularly at a young age, treat the fictional characters of mass media products as if they were real-life friends and loved ones. When they read about or see their fictional loved ones engaging in violent behavior, the viewers learn a violent “definition of the situation,” and carry it with them as a potential response to real-life disagreements and slights (Sutherland, 1939, p. 77).

Worries over the impact of mass media images on juvenile physical violence have a long history. In 1865, early sociologist Harriet Martineau toured a reform school for juvenile delinquents and reported with shock that “every boy had a penny dreadful in his back pocket” (Martineau, 1865). Her implication was that reading penny dreadfuls, the cheap story magazines sold for a penny, caused the delinquency. In 2007, a Washington Post poll asked respondents why there is so much gun violence in America. The most common response was “the influence of popular culture, such as movies, television, and the internet,” chosen by 43% of respondents, followed by “the way parents raise their children,” chosen by only 31% (ABC News & Washington Post, 2009).

Hundreds of surveys conducted over the past 60 years have investigated the hypothesized link between depictions of physical violence in mass media and real-life physical violence, with disappointing results. Statistical analysis generally reveals no significant impact, particularly after eliminating the influence of antecedent variables (e.g., low parental supervision) (Walker & Morley, 1991) or intervening variables (e.g., exposure to real-life violence). Some surveys have even found that watching television (Barnes, Hoffman, Welte, Farrell, & Dintcheff, 2006) or playing violent video games (Ferguson, 2011) actually deter juveniles from delinquency. When a substantive media-violence link is found, it usually fades away over time in longitudinal studies (Wiegman, Kuttschreuter, & Baarda, 1992), or design flaws in the survey methodology make the results unreliable (Surette, 2006, p. 80). Operationalizations are over-extended, such as defining ordinary arguments as “violence.” Variables are measured with questions that are nearly impossible to answer accurately, such as “How many violent acts have you witnessed on television during the last year?” Sometimes a tiny percent difference, such as 38% of delinquents and 35% of non-delinquents, is promoted as indicative of a causal link (Coyne, 2007).

Research on media influence on the physical violence of adults is minimal. This seems strange, because adults commit far more violent acts than children, so one would expect more concern over the possible impact of media images on their behavior. However, perceptions of children as uniquely vulnerable and suggestible may increase the concern, and thus the number of the research studies (Jenks, 1996, p. 72). Instead, most surveys concerning adults investigate the hypothetical impact of viewing depictions of sexual behavior, usually the explicit sexual behavior in pornographic films, on the risk of engaging in real-life rape and sexual assault.

Again, fears of a possible media–sexual assault link extend far back in history. The rationale behind the first criminalization of obscene books was that “weak-minded” readers would thereby have their “prurient interests” aroused and be unable to control their sexual passions (Heins, 2007, p. 28). Survey investigations began during the 1970s and have continued unabated. Sometimes the dependent variable is operationalized with a question about whether the respondent has engaged in rape or a sexual assault, but because it is difficult to get approval from Institutional Review Boards to ask about respondents’ criminal activities, most surveys substitute the answer to a hypothetical question (“If I was certain that I could get away with it, I would have sex with someone against their will”). Others measure the respondents’ belief in “rape myths,” such as “women who say ‘no’ actually mean ‘yes’” (the woman’s initial refusal is commonplace in heterosexual pornography). Presumably a male respondent who has acquired such beliefs from media depictions is likely to ignore a rejection of his sexual advances.

The results of these surveys have been as disappointing as the results in the media–physical violence surveys. One study even found heavy consumers of pornography to be more punitive than light consumers in their attitude toward marital rape: they were significantly more likely to suggest imprisonment for the offender rather than counseling or “no penalty” (Davies, 1997). As with the media–physical violence link, when a significant causal impact is announced, flaws in the study’s methodology usually negate the validity of the results (Ferguson & Hartley, 2009).

Risk of Nonviolent Delinquent Behavior

Since the 1920s, scholars, teachers, and parents have been worrying that juveniles who are exposed to the non-explicit sexual situations in fictional media are at heightened risk of engaging in real-life sexual behavior, a status offense for juveniles under the age of consent and a significant predictor of nonviolent delinquency. In Middletown, their 1929 sociological study of small-town Muncie, Indiana, Robert and Helen Lynd were shocked to find teenage boys and girls “taking liberties” with each other. They concluded that this unprecedented and unnatural erotic interest during adolescence was caused by listening to jazz music, reading trashy magazines, and “the constant witnessing of sex films,” that is, the silent movies of the pre–Hays Code era, which are rather sedate by contemporary standards (Lynd & Lynd, 1956, p. 138).

The methodological hurdle of using adolescent respondents, plus, no doubt, researcher squeamishness over thinking about juveniles as sexually active, has limited the number of surveys measuring the hypothetical link between the “witnessing of sex films” and the decision to engage in underage sexual activity. The dependent variable is usually operationalized through answers to questions about the juveniles’ age at their first sexual act, or if they have not engaged in sexual acts, their interest in doing so. The results of the existing surveys seem promising: juveniles with an early age of “sexual onset” or an expressed interest in engaging in sexual acts are more likely than their peers to have consumed media with non-explicit sexual content. However, the methodological flaws seen elsewhere are present here as well, especially faulty correlation: often adolescents planning to engage in sexual activity seek out media depictions for tips on the proper procedure, thus reversing the hypothesized causal link (Pardun, L’Engle, & Brown, 2005). In other studies, a significant causal value is present only when the sexual media is consumed with friends, not when it is consumed alone, thus making the social context the operant independent variable: peers are the actual agents of socialization, the “intimate associates” of social learning theory, and media images mere tools for conveying group expectations (L’Engle, Brown, & Kenneavy, 2006).

A final area of hypothesized media influence on behavior is the “glorification” of alcohol and tobacco, and more infrequently illicit drugs, encouraging juvenile decisions to begin using them. Social learning theory suggests that we imitate the acts engaged in by people we admire, such as the glamorous, exciting stars of movies and television programs, in an attempt to gain their favor and share in their prestige. However, surveys have found that the use of alcohol and tobacco by characters in fictional media, as well as advertisements for alcohol and tobacco, have a negligible impact on juvenile decisions to drink or smoke. Peer and parental influences are far stronger. An exception may be the advertising juveniles view at sporting events or in other social situations, where, as with sexual behavior, peers are the actual agents of socialization, using the advertisements to promote their expectation that the juvenile should begin using alcohol or tobacco (Ellickson, Collins, Hambarsoomians, & McCaffrey, 2005).

Fear of Crime and Crime Misinformation

Many surveys from the 1920s to the present have revealed that the general public has serious misconceptions about the nature and extent of crime. They are very likely to believe that the crime rate is going up, when actually it is going down. They estimate that their risk of crime victimization is staggeringly high. Their information about specific crimes is faulty at best. They believe that most murders are perpetrated by serial killers, that the number of school shooting incidents has increased dramatically, that rapists are strangers to their victims, that thousands of children are abducted and murdered by pedophiles every year. Media producers enhance these misconceptions by creating a “discourse of fear,” driving up television ratings and newspaper readership through the canny use of fear-inducing vocabulary, such as referring to attempts to control crime as a militarized “war on crime” (Altheide, 2003, p. 31). These misconceptions can have a deleterious impact on both individual behavior and crime policies, as elected officials are pressured to ignore more serious problems and spend time and money on unnecessary interventions, such as installing metal detectors at the entrance to schools rather than addressing the endemic physical and sexual assaults that occur on school property (Dowler, 2003).

Moral panics, sudden media-driven surges in perceptions of the seriousness of a public threat that is actually of minimal importance or even non-existent, can result in social, political, and economic catastrophe. For instance, the crack panic of the 1980s was a media-driven perception that crack cocaine is far more dangerous than powder cocaine, instantly addictive, and causes instant psychosis in both the users and their unborn children, who will grow up to be super predators. In fact, the two forms of the drugs have practically identical properties: neither causes instant addiction, psychosis, or super predators. However, the purported danger of crack cocaine overwhelmed news programs and fictional media, and was to a great extent responsible for the War on Drugs that has given the United States the highest incarceration rate in the world (Reinarman & Levine, 1997).

We know that most people with no academic background in criminology cannot access official crime reports or criminological research studies, or understand how to interpret them, so they must rely on mass media news sources, which increasingly use “infotainment,” emphasizing the most sensational stories (Anderson, 2004). Content analysis, a research methodology involving analysis of media texts, has demonstrated that television news programs regularly exaggerate the incidence and severity of crime, for instance, by saying the number of heroin users in the city has “skyrocketed,” when it has, in fact, increased from 200 to 220, or that two drive-by shooting incidents in six months constitutes an “epidemic of violence.” In addition, in the absence of disconfirming information, media consumers may also accept fictional crime movies and television programs as accurate representations of real-life dangers. It makes intuitive sense, therefore, that heavy consumption of mass media results in a skewed belief that the world is more violent, chaotic, crime-ridden, and dangerous than more empirical data warrants. The main theoretical arguments derive from the heuristic-systematic model of social psychology, in which media images provide us with mental “shortcuts” with which to make sense of ambiguous or confusing situations (“strangers are usually evil”), and from media studies themselves, with the cultivation theory developed by George Gerbner during the 1970s.

Surveys conducted during the 1970s and 1980s often found that heavy media consumption was correlated with heightened fear of crime victimization, but more recent surveys have mixed results, varying depending on the genre, the type of crime, whether the crime occurs locally or in another area, and the age, race, and gender of the consumer: among heavy consumers of television crime news, elderly, white, and female respondents are significantly more likely than young, nonwhite, and male respondents to perceive a heightened risk of crime victimization. As with juvenile sexual behavior and alcohol and tobacco use, the social setting in which the media is consumed may be the causal variable, not the media image itself. Even the most uninformed respondent has a variety of sources to draw information on crime from, including observation, logical deduction, folk wisdom, and word of mouth, so the impact of media images is more complex than merely believing what one sees on TV (Chiricos, Eschholz, & Gertz, 1997).

Biased Attitudes and the Criminal Justice Process

The criminals depicted in both crime news and fictional crime dramas are disproportionately black and Hispanic. Fictional serial killers are usually depicted as stereotypic gay men. Racial and religious minorities, lower- and working-class persons, and members of other disenfranchised groups are regularly excoriated in mass media portrayals as sneaky, duplicitous, deviant, potentially violent, and generally suspect. As with fear of crime, the heuristic-systematic model of social psychology and cultivation theory of media studies hypothesize that individuals who have little exposure to members of these groups in everyday life use media images to fill in the blanks in their knowledge, thus acquiring biased attitudes, believing negative stereotypes to be accurate. Then, when they see members of the group on the street, they are likely to perceive their behavior as suspect and call the police. The police are more likely to stop and search, or issue an arrest. Judges and juries are more likely to convict.

Anti-defamation groups have been condemning examples of media stereotyping since at least 1913, when the Anti-Defamation League of the B’nai B’rith protested the “objectionable and vulgar” Jewish stereotypes in the New York Times. However, surveys specifically attempting to measure the impact of biased media images on individual biased attitudes date only from the 1970s, and only a few have specified criminal stereotypes, the belief that members of the target group are more violent or otherwise criminal. The results have been poor, primarily due to faulty correlation: respondents who are already biased seek out biased media, rather than learning new biases. They may also ignore disconfirming data.

Some surveys have skipped over the creation of biased attitudes and moved directly to behavior, measuring the impact of biased mass media images on the process of making criminal justice decisions, with more promising but still limited results. For instance, one survey experiment had respondents watch fictional TV programs about a white criminal in the control group and a black criminal in the experimental group, and then read about a purportedly “real” criminal case with a black defendant. Then respondents were asked to issue a guilty or not guilty verdict, and if guilty, determine the length of the prison sentence. There was no significant difference in the guilty verdicts, or in the prison sentences suggested by female respondents, but male respondents were more likely to find the black defendant guilty, and proposed a significantly longer prison sentence. Other studies have found that biased decision-making is so heavily dependent on demographics, cultural context, and other antecedent variables that the impact of media bias is negligible, especially when compared to the undeniably strong impact of parental and peer socialization, subculture, and the intersection of race, class, and gender privileges (Oliver, 2003).

The CSI Effect

In fictional crime investigation television programs, cases—usually involving homicide—are always solved within an hour, through a trial that clears the innocent defendant and draws a confession from the real killer by means of extensive use of forensic evidence. Every crime scene is “cooperative,” with DNA, fingerprints, hair, and other fibers readily available, easily detected, and conclusive. Fingerprints can be drawn from snowbanks, or from knives pulled out of the ocean. Special detectors can identify the offender by analyzing hair, pillow fabric, knife grooves, and bits of moss found in cracks in the sidewalk. The CSI Effect, named after the popular television program CSI (Crime Scene Investigation, 2000–2015), which has spawned several other similar shows, is the worry that jurors who are heavy viewers of crime investigation television programs will expect real-life forensic investigation to be equally spectacular, extensive, and conclusive, and therefore be more likely acquit the defendant because no DNA is available, or because no fingerprints were lifted from a snowbank.

Judges and attorneys increasingly believe that the CSI Effect is responsible for many guilty defendants receiving not-guilty verdicts from gullible juries. They interrogate prospective jurors about their television viewing habits during voir dire. They even ask pointedly, “Do you think CSI is real?”

Concern over the possible effects of CSI and other forensic investigation programs is so new that there hasn’t been time to conduct many research studies. The few surveys that have been conducted generally measure the dependent variable through the acquittal rate in a hypothetical court case, but at least one has used the results of real jury decisions. The results are considerably different from expectations: heavy forensic investigation television viewers are actually less likely to acquit, and more likely to be satisfied with the forensic evidence presented in the real-life courtroom, than light viewers (Podlas, 2005). They hardly believe that “CSI is real.” Indeed, they often have characteristics desirable in a juror: a keen interest in the criminal justice system and a good knowledge of real-life forensic investigation and its limitations. We can see a parallel in the fans of science fiction television programs like Star Trek and Babylon 5, who are often interested in developments in real-life astronomy and physics, perhaps more so than non-viewers. They are able to enjoy stories involving warp drive and transporter beams without expecting them to be real.

Media as Practice

Sixty years of survey-based research studies have been unsuccessful in finding a significant causal impact of media consumption on violent and sexual criminal behavior, and only marginally successful in finding a causal impact on biased attitudes, fear of crime, belief in crime myths, and unrealistic forensic expectations. This may suggest that media consumption does not readily change one’s attitudes, beliefs, and behavior, or it may suggest a problem with the survey methodology itself.

The independent variable, media consumption, is typically measured through questions about the frequency of consumption (for instance, “How often do you watch crime TV?” or “How many violent movies have you seen?”), an operationalization that ignores the many ways in which we watch TV or movies: alone or with friends, paying close attention or not paying attention at all, as an avid fan or a casual viewer, dispassionately, ironically, critically. The dependent variable, a change in attitude, beliefs, or behavior, is typically measured through responses to hypothetical situations, an operationalization that may not adequately address how people would respond to a similar situation in real life, or self-reports of minor offenses, which may not adequately address the etiology of serious offending. The correlational nature of the research, in which media have a direct causal link with beliefs or behavior, is also problematic, because media consumption and real-life behavior are both ongoing phenomena.

There are also logistic problems, derived from the practical considerations of managing data collection from large groups. It is difficult to pursue longitudinal studies, measuring effects over time, due to participant and researcher drop-out. Adolescent respondents cannot participate without parental consent, and they are usually certain that their parents will have full access to their responses, leading to an overabundance of desirability bias. Finally, there may not be an adequate theoretical basis to create a coherent research hypothesis in the first place.

There is another possibility. Some scholars dispute the validity of the concept of media effects on real-life attitudes, beliefs, and behavior. (Bird, 2003, p. 2; Croteau & Hoynes, 2013, p. 260; Barak, 2012). It assumes a simple disjunction between real life and media images, the latter an imperfect simulacra that, at best, provides a momentary escape from the real, and more likely prohibits us from comprehending and acting upon the real. We play Grand Theft Auto, and go out to shoot our enemies as if we are still playing. We watch CSI, and when we are called to jury duty, judge the trial proceedings by its standards.

This simple, unidirectional cause and effect between two radically disjointed spheres of human activity has been subject to critique since the days of Adorno and Benjamin in the Frankfort School in the 1940s. The disciplines of cultural studies have moved on to reception theory, and mass communications, to the active audience theory, which argues that media is best conceptualized as something one practices rather than something one consumes (Couldry, 2004). We are active participants in television and movie watching and video game playing, bringing our own interests, beliefs, and social situations to the images, modifying them based on our understanding of the social world as often as we modify our understanding of the social world based on them. A more productive avenue for the use of survey research may be to measure not mere causality but the many ways in which media crime and real-life beliefs, attitudes, and behaviors interact with each other.

Literature Review and Primary Sources

T. M. Porter, The Rise of Statistical Thinking, 1820–1900 (Princeton University Press, 1986)D. Ross, The Origins of American Social Science (Cambridge University Press, 1992)S. Igo, “Subjects of Persuasion: Survey Research as a Solicitious Science,” in Social Knowledge in the Making (Camic & Gross, Eds., University of Chicago Press, 2011)

The history of survey methodology, in conjunction with the quantitative turn in the social sciences, is detailed in ; and , and in several general histories of the social sciences, notably .

The Practice of Social Research, by Earl S. Babbie (Wadsworth Publishing, 2016)Bradburn and Sudman, Polls and Surveys: Understanding What They Tell Us (Jossey-Bass, 1988)Kalton, Introduction to Survey Sampling (SAGE, 1983)Fowler, Improving Survey Questions: Design and Evaluation (SAGE, 1995)Teddlie and Taskakkori, Foundations of Mixed Methods Research: Integrating Quantitative and Qualitative Approaches in the Social and Behavioral Sciences (SAGE, 2008)

Of the many manuals on the writing and administering of surveys, social research methods textbooks often give the easiest, least obfuscated advice on such practical matters as increasing participation and reducing the desirability bias. , has been revised regularly since its first publication in 1975, and now includes information on the benefits and drawbacks of survey distribution on social media networks. For more specialized topics, the most essential texts are ; ; ; and .

Kappeler and Potter, The Mythology of Crime and Criminal Justice (Waveland, 2004)Kirsh, Children, Adolescents, and Media Violence: A Critical Look at the Research (SAGE, 2011)Flanagan and Longmire, Americans View Crime and Justice: A National Public Opinion Survey (SAGE, 1996)Goode and Ben-Yehuda, Moral Panics: The Social Construction of Deviance (Wiley-Blackwell, 2009)Covington, Crime and Racial Constructions: Cultural Misinformation about African Americans in Media and Academe (Lexington Books, 2011)Dennis, “The LGBT Offender,” in Handbook of LGBT Communities, Crime, and Justice (Peterson & Panfill, Eds., Springer, 2014)

There are few, if any, primary sources dedicated specifically to how survey methodology has measured the relationship between mass media and crime. The reader is directed to the texts on media influences on crime attitudes, beliefs, and behavior measured through all methodologies, in the Further Reading section, as well as the following. On media influence on criminal behavior, see ; and . On fear of crime and crime myths, see ; and . For the criminalization of racial and other minorities, see ; and .

Neil Postman’s AmusingOurselves to Death (Penguin, 1985)Media Unlimited: How the Torrent of Images and Sounds Overwhelms Our Lives (Picador, 2002)

was one of the first books to worry about the overload of media images in contemporary society, which, coupled with the merger of serious journalism with entertainment, threatened to destroy our critical reasoning abilities, rendering us passive, fearful, and infantilized. In , Todd Gitlin applies a sociological lens to the fear.

Debord, The Society of the Spectacle (Black & Red, 1967)Baudrillard, Simulacra and Simulation (University of Michigan Press, 1981)Cahoone, The Dilemma of Modernity (State University of New York Press, 1987)Calavita, Apprehending Politics: News Media and Individual Political Development (State University of New York Press, 2005)Staiger, Media Reception Studies (New York University Press, 2005)Kitzinger, A Sociology of Media Power: Key Issues in Audience Reception Research (Longman, 1999)Bird, The Audience in Everyday Life: Living in a Media World (Routledge, 2003)Croteau and Hoynes, Media/Society: Industries, Images, and Audiences (SAGE, 2013)Greg Barak’s article “Media and Crime” in the Routledge Handbook of Critical Criminology (DeKeseredy &Dragiewicz, Eds., Routledge, 2012)

Among the postmodern texts that problematize the disjunction of media and real life are ; and . See also . Some useful texts in contemporary reception theory are ; and . Active audience theory is described in ; ; and Chapter 8 of . summarizes some of the key critiques through the lens of criminology.

Further Reading


Trend, D. (2007). The myth of media violence: a critical introduction. Malden, MA: Blackwell.Find this resource:

Criminal and Delinquent Behavior

Hajdu, D. (2008). The ten cent plague: The great comic-book scare and how it changed America. New York: Picador.Find this resource:

Jones, G. (2003). Killing monsters: Why children need fantasy, super heroes, and make-believe violence. New York: Basic Books.Find this resource:

Kirsch, S. (2009). Media and youth. Malden, MA: Blackwell.Find this resource:

Kutner, L., & Olson, C. K. (2011). Grand theft childhood: The surprising truth about violent video games and what parents can do. New York: Simon & Schuster.Find this resource:

Lamish, D. (2007). Children and television: A global perspective. Malden, MA: Blackwell.Find this resource:

Sternheimer, K. (2003). It’s not the media: The truth about popular culture’s influence on children. Boulder, CO: Westview Press.Find this resource:

Fear of Crime and Crime Myths

Chadee, D. (Ed.). (2016). Psychology of fear, crime, and the media: International perspectives. New York: Routledge.Find this resource:

Glassner, B. (2000). The culture of fear. New York: Basic Books.Find this resource:

Leishman, F., & Mason, P. (2011). Policing and the media. Portland, OR: Willan.Find this resource:

Bias and the Criminal Justice Process

Entman, R. M., & Rojecki, A. (2001). The black image in the white mind: Media and race in America. Chicago: University of Chicago Press.Find this resource:

Holtzman, L., & Sharpe, L. (2014). Media messages: What film, television, and popular music teach us about race, class, gender, and sexual orientation. New York: Routledge.Find this resource:

Walker, S., Spohn, C., & DeLone, M. (2011). The color of justice: Race, ethnicity, and crime in America. Belmont, CA: Wadsworth Publishing.Find this resource:


ABC News & Washington Post. (2009, June 11). ABC News/Washington Post Poll, July 2007. ICPSR24589-v1. Ann Arbor, MI: Inter-University Consortium for Political and Social Research.Find this resource:

Altheide, D. L. (2003). Creating fear: News and the construction of crisis. Piscataway, NJ: Aldine Transaction.Find this resource:

Anderson, B. (2004). News flash: Journalism, infotainment and the bottom-line business of broadcast news. San Francisco: Jossey-Bass.Find this resource:

Babbie, E. S. (2016). The practice of social research (14th ed.). Boston: Cengage.Find this resource:

Barak, G. (2012). Media and crime. In W. S. DeKeseredy & M. Dragiewicz (Eds.), Routledge handbook of critical criminology (pp. 383–387). New York: Routledge.Find this resource:

Barnes, G. M., Hoffman, J. H., Welte, J. W., Farrell, M. P., & Dintcheff, B. A. (2006). Effects of parental monitoring and peer deviance on substance use and delinquency. Journal of Marriage and Family, 68(4), 1084–1104.Find this resource:

Bird, S. E. (2003). The audience in everyday life: Living in a media world. New York: Routledge.Find this resource:

Carrabine, E. (2008). Crime, culture, and the media. Malden, MA: Polity Press.Find this resource:

Chiricos, T., Eschholz, S., & M. Gertz. (1997). Crime, news and fear of crime: Toward an identification of audience effects. Social Problems, 44(3), 342–357.Find this resource:

Couldry, N. (2004). Theorising media as practice. Social Semiotics, 14(2), 115–132.Find this resource:

Coyne, S. M. (2007). Does media violence cause violent crime? European Journal on Criminal Policy and Research, 13(3–4), 205–211.Find this resource:

Croteau, D., & Hoynes, W. D. (2013). Media/Society: Industries, images, and audiences. Thousand Oaks, CA: SAGE.Find this resource:

Davies, K. A. (1997). Voluntary exposure to pornography and men’s attitudes toward feminism and rape. Journal of Sex Research, 34(2), 131–137.Find this resource:

Dowler, K. (2003). Media consumption and public attitudes toward crime and justice: The relationship between fear of crime, punitive attitudes, and perceived police effectiveness. Journal of Criminal Justice and Popular Culture, 10(2), 109–126.Find this resource:

Ellickson, P. L., Collins, R. I., Hambarsoomians, K., & McCaffrey, D. F. (2005). Does alcohol advertising promote adolescent drinking? Results from a longitudinal assessment. Addiction, 100(2), 235–246.Find this resource:

Ferguson, C. J. (2011). Video games and youth violence: A prospective analysis in adolescents. Journal of Youth and Adolescence, 40(4), 377–391.Find this resource:

Ferguson, C. J., & Hartley, R. D. (2009). “The pleasure is momentary … the expense damnable?” The influence of pornography on rape and sexual assault. Aggression and Violent Behavior, 14, 323–329.Find this resource:

Groves, R. M. (2011). Three eras of survey research. Public Opinion Quarterly, 75(4), 861–871.Find this resource:

Heins, M. (2007). Not in front of the children: Indecency, censorship, and the innocence of youth. New Brunswick, NJ: Rutgers University Press.Find this resource:

Jenks, C. (1996). Childhood. New York: Routledge.Find this resource:

Jewkes, Y. (2004). Media and crime. Thousand Oaks, CA: SAGE.Find this resource:

Kilpatrick, D. G., & Saunders, B. E. (2009). National survey of adolescents in the United States, 1995. ICPSR02833-v1. Ann Arbor, MI: Inter-university Consortium for Political and Social Research.Find this resource:

Kim, J. Y. S., Fendrick, M., & Wislar, J. (2000). The validity of juvenile arrestees’ drug use reporting. Journal of Research in Crime and Delinquency, 37, 419–432.Find this resource:

King, G. (1991). On political methodology. Political Analysis, 2, 1–30.Find this resource:

Krumpal, I. (2013). Determinants of social desirability bias in sensitive surveys: A literature review. Quality and Quantity, 47(4), 2025–2047.Find this resource:

L’Engle, K. L., Brown, J. D., & Kenneavy, K. (2006). The mass media are an important context for adolescents’ sexual behavior. Journal of Adolescent Health, 38(3), 186–192.Find this resource:

Lynd, R., & Lynd, H. (1956). Middletown: A study in modern American culture. New York: Harcourt Brace & World.Find this resource:

Martineau, H. (1865, October). Life in the criminal class. Edinburgh Review, 122, 337–371.Find this resource:

McFarland, S. G. (1981). Effects of question order on survey responses. Public Opinion Quarterly, 45(2), 208–215.Find this resource:

Oliver, M. B. (2003). African American men as “criminal and dangerous”: Implications of media portrayals of crime on the “criminalization” of African American men. Journal of African American Studies, 7(2), 3–18.Find this resource:

Pardun, C. J., L’Engle, K. L., & Brown, J. (2005). Linking exposure to outcomes: Early adolescents’ consumption of sexual content in six media. Mass Communication and Society, 8(2), 75–91.Find this resource:

Podlas, K. (2005). The CSI Effect: Exposing the media myths. Fordham Intellectual Property, Media, and Entertainment Law Journal, 16(2), 429–465.Find this resource:

Reinarman, C., & Levine, H. G. (1997). Crack in America: Demon drugs and social justice. Berkeley: University of California Press.Find this resource:

Simon, L. (1995, November). Validity and reliability of violent juveniles: A comparison of juvenile self-reports with adult self-reports. Paper presented at the annual meeting of the American Society of Criminology, Boston, MA.Find this resource:

Sirkin, M. (2006). Statistics for the social sciences (3d ed.). Thousand Oaks, CA: SAGE.Find this resource:

Surette, R. (2002). Self-reported copycat crime among a population of serious and violent juvenile offenders. Crime and Delinquency, 48(1), 46–69.Find this resource:

Surette, R. (2006). Media, crime, and criminal justice: Images, realities, and policies. Belmont, CA: Wadsworth.Find this resource:

Sutherland, E. (1939). Principles of criminology (3d ed.). Philadelphia: Lippincott.Find this resource:

Walker, J. T. (1999). Statistics in criminal justice. Gaithersburg, MN: Aspen Publications.Find this resource:

Walker, K. B., & Morley, D. D. (1991). Attitudes and parental factors as intervening variables in the television violence-aggression relation. Communication Research Reports, 8(1–2), 41–47.Find this resource:

Wiegman, O., Kuttschreuter, M., & Baarda, B. (1992). A longitudinal study of the effects of television viewing on aggressive and prosocial behaviours. British Journal of Social Psychology, 31(2), 147–164.Find this resource: