Skip to main content
SearchLoginLogin or Signup

Do Photos of Police-Civilian Interactions Influence Public Opinion about the Police? A Multi-Method Test of Media Effects

Purpose: To test whether exposure to news images depicting law enforcement affects public attitudes toward the police. Method: Participants drawn from a national online panel were randomly assigned to view one of three pictures that depicted a range of hostile to friendly ...

Published onJan 25, 2020
Do Photos of Police-Civilian Interactions Influence Public Opinion about the Police? A Multi-Method Test of Media Effects


Purpose: To test whether exposure to news images depicting law enforcement affects public attitudes toward the police. Method: Participants drawn from a national online panel were randomly assigned to view one of three pictures that depicted a range of hostile to friendly police-civilian interactions (compared to a control group who saw no pictures).  Dependent variables were perceptions of police officers’ effectiveness, misconduct, and bias.  Moderating variables were respondents’ experiential, ideological, or demographic characteristics. As a follow-up to the results of the experiment, regression analyses were employed to explore other factors that may influence perceptions of police or interact with the media effects. Results: Image exposure did not directly affect any dimension of attitudes toward the police, but there was one significant moderation effect.  Respondents who had been recently stopped by an officer and saw a picture of a friendly interaction between officers and a civilian perceived more frequent police misconduct than respondents in the same experimental condition who were not recently stopped.  Routine media consumption was significantly related to perceptions of police in the nonexperimental analysis. Conclusions: Findings indicate that brief exposure to static images of law enforcement disseminated by the media does not independently affect people’s opinions about the performance of police in society.  Rather, people’s global opinions about the police are shaped by their own beliefs, prior experiences with officers, and cumulative, self-selected media consumption. 

Keywords: Public opinion; Police; Media; Experiment; Protests; Police militarization


Scholars in the United States have long debated whether media coverage of law enforcement influences public opinion about the police.  Discussions of media effects frequently center on instances of police use-of-force or misconduct against civilians, such as the videotaped 1991 police beating of Rodney King that was broadcast nationwide by the media (Lasley, 1994).  More recently, confrontations between police officers and civilians during mass public protests in Ferguson, Missouri and Baltimore, Maryland received extensive media coverage (Gately & Stolberg, 2015; “Photo essay,” 2014).  The catalysts for these mass protests were the deaths of African American men at the hands of police officers (Michael Brown in Ferguson and Freddie Gray in Baltimore) that reignited long-simmering tensions between police and communities of color (Cobbina, 2019; Desmond et al., 2016; Holmes et al., 2019; Peck, 2015; Peffley & Hurwitz, 2010; Prowse et al., 2019; Rios, 2011; Tyler, 2006). 

Police tactics can be conceptualized as a spectrum defined by the police-civilian relationship (Weisburd & Eck, 2004).  On one side of the spectrum is community policing, which encourages officers to develop collegial relationships and maintain close communication with civilians in the communities they serve.  On the other side of the spectrum are more “aggressive” crime control tactics, like “stop and frisk” policing, that emphasize intense surveillance and frequent arrests for offenses.  The public protests in Ferguson and Baltimore became case studies in the kinds of hostile police-civilian interactions that stand at the furthest end of the “aggressive” side of the policing spectrum.  Evidence indicates that aggressive crime policing practices can damage citizen and community trust in the police, which in turn impedes citizens’ willingness to cooperate with police investigations (Braga et al., 2019).   Though conflicts between police and protesters like those that occurred in Ferguson and Baltimore are much rarer than interactions like stop-and-frisks, politicians and pundits worried that they might exert outside influence on the public consciousness.

Journalists and civilians (using social media) at the Ferguson and Baltimore protests provided visual documentation of police confronting protesters in full riot gear, frequently with guns drawn, sometimes from atop armored vehicles (Bosman & Apuzzo, 2014; Gately & Stolberg, 2015).  Reacting to the Ferguson protests, Walter Olson (2014) of the Cato Institute commented, “The dominant visual aspect of the story…has been the sight of overpowering police forces confronting unarmed protesters who are seen waving signs or just their hands.”  These events revitalized debates among politicians and media commentators about whether or not it was appropriate for domestic police forces to adopt the training, tactics, and weaponry of the military (Balko, 2013; Bowman, 2014; Kraska, 2007; Kraska & Kappeler, 1997).  Numerous pundits, policymakers, and criminal justice professionals debated whether media coverage of “police militarization” would erode the legitimacy of domestic law enforcement in the mass public consciousness (Institute for Intergovernmental Research, 2015).

While this hypothesis has commonsense appeal, empirical evidence of media effects on public opinion about the police is inconsistent, and scholars are paying greater attention to selection bias in media consumption and heterogeneous media effects across different facets of attitudes toward the police.  We contribute to this literature with an experimental study in which we exposed participants to images of police officers interacting with civilians paired with a nonexperimental regression analysis of common predictors of attitudes toward the police.  The experimental method alleviates concerns about media selection bias.  We find little evidence that brief media exposure affects public opinion.  Rather, attitudes toward the police are correlated with routine, self-selected media consumption and demographic, experiential, and ideological factors.


Most of the research in this area was based, implicitly or explicitly, on cultivation theory (Gauthier & Graziano, 2018).  Cultivation theory argues that the more media a person consumes, the more likely it is that he or she will endorse the perspectives that are presented in the dominant media narrative (Gerbner, 1970). Scholars criticized this theory over the decades since its inception, particularly arguing that media effects are likely to differ according to the type of programming one consumes rather than reflect a singular “media narrative” (Morgan & Shanahan, 2010). In terms of public views of crime, punishment, and related topics, the cultivation literature is mixed on the question of whether overall viewing habits or consumption of specific media genres (e.g., local news) impact one’s perceptions of crime (Romer et al., 2003; Weitzer & Kubrin, 2004).  Policing scholars frequently focused their studies on one type of media (or genre) at a time, which partially explains the heterogeneous findings of media effects (or lack, thereof).  Some studies focused on consumption of news coverage of police misconduct (Chermak et al., 2006; Jefferis et al., 1997; Kaminski & Jefferis, 1998; Lasley, 1994; Sigelman et al., 1997; Weitzer, 2002; Weitzer & Tuch, 2004a, 2004b), whereas others focused on consumption of fictional dramas or reality TV shows about police (Donovan & Klahm, 2015; Eschholz et al., 2002).1 Since the former type of media often has a critical tone against police while the latter has a positive valance in favor of police (see literature review in Intravia, Wolff, & Piquero, 2018), these studies often reveal opposite effects of media consumption on attitudes toward the police.  The fact that different types of media portray the police under different circumstances with different tones of coverage means that general measures of news or media consumption are theoretically-ambiguous in nature.  In other words, general measures of frequency of news consumption provide no indication of the specific nature of coverage about police a respondent was exposed to during her routine news viewing.  As such, it is unsurprising that tests of general media consumption present inconsistent or null evidence of media influence over public opinion about the police (Callanan & Rosenberger, 2011; Dowler, 2002; Intravia et al., 2018; Roche et al., 2016).

On top of the heterogeneity of media depictions of the police is the fact that people choose which types or sources of media to consume (e.g., Iyengar & Hahn, 2009).  This reality raises the ultimate challenge for causal inference: selection bias.  Do laudatory or critical media depictions of the police cause people to assess the police in a more positive or negative manner, or do people who hold more positive or negative attitudes toward the police selectively choose to consume media that is likely to reinforce their perceptions of law enforcement?  As such, scholars must demonstrate that media consumption or exposure significantly affects attitudes toward law enforcement above-and-beyond the influence of other common predictors of public opinion about the police, such as race, age, gender, education, and personal contact with the police (Brown & Benedict, 2002).  Correlational research designs that test the relationship between a measure(s) of media consumption and a measure(s) of attitudes toward the police while controlling for other typical predictors through a regression analysis can begin to address this causal inference problem, but even the most carefully specified cross-sectional nonexperimental designs remain vulnerable to omitted variable bias

Prior research also suggests that effects of media exposure may be moderated by people’s personal characteristics.  For example, Callanan & Rosenberger (2011) found that consumption of both news programs and crime-related reality programs increased confidence in the police, but only among white people (vs. Latinx and African Americans), and only among people who had not experienced a crime victimization or an arrest at the hands of the police.  Weitzer & Tuch (2004a) found that consumption of media coverage of police misconduct was positively related to African American respondents’ perceptions of the prevalence of police misconduct in their neighborhood and city.  This media exposure effect was significantly higher among black respondents who lived in high-crime neighborhoods, but media exposure did not affect the police misconduct perceptions of white or Latinx respondents.  Together, these studies suggest that personal or vicarious/communal experiences with crime and/or the police either supersede or condition the ability of the media to influence people’s perceptions of the police.  Such evidence led Gauthier & Graziano (2018) to argue that criminologists must develop a more sophisticated understanding of cultivation theory that analyzes not just media exposure but also audience characteristics.


Despite methodological challenges that impede scholars’ ability to generate estimates of the relationship between media consumption and attitudes toward the police with an unambiguous theoretical meaning, we cannot ignore the question of media effects because debates about the relationship between the police, the media, and the American public are only growing more salient.  With the advent of social media, smartphone cameras, and the increasing prevalence of officer body-worn cameras, the mass public is being exposed to images and videos of police officers interacting with civilians with greater and greater frequency.  In order to better understand how these audio and visual stimuli do (or do not) influence public attitudes toward the police, scholars are increasingly turning to experimental methods.  The body of experimental studies is still quite small and includes a variety of methodological designs.2 Several studies showed participants scripted videos of police officers interacting with civilians in which the behavior of the officer was manipulated to depict either standard/neutral behavior, behavior that implements the guidance of procedural justice theory, or hostile/confrontational behavior that violates procedural justice theory.  In general, these studies demonstrated that exposure to a video of an officer implementing procedural justice increased participants’ reported willingness to cooperate with the officer if they were in the situation depicted, feeling of obligation to obey the officer, and trust and confidence in the officer. In contrast, exposure to the video of hostile officer behavior decreased respondents’ trust, confidence, and willingness to cooperate with and obey the officer.  The treatment either had no effect on global attitudes of trust in the police generally, or effects that were weaker and more inconsistent than the effects on respondents’ attitudes toward the specific officer depicted in that specific situation (Johnson et al., 2017; Lowrey et al., 2016; Maguire et al., 2017).  Reisig, Mays, & Telep (2018) employed a similar design except rather than showing scripted videos of police-civilian interactions, they used written vignettes that described neutral, procedurally-just, or procedurally-unjust scenarios.  They also found that respondents who read the procedurally-just scenarios expressed more positive attitudes toward the officer described in the text; they did not measure more global attitudes toward the police.  Boivin, Gendron, Faubert, & Poulin (2017) found that respondents who saw scripted videos depicting police uses of force of varying severity reported heightened belief in the prevalence and brutality of police use of force, but video exposure did not affect more general assessments of the police.

Other studies did not manipulate the video content but rather manipulated the information that was given to respondents about the police-civilian interaction video they viewed.  Braga, Winship, Tyler, Fagan, & Meares (2014) and Culhane, Boman, & Schweitzer (2016) found that providing additional information about 1) the context of the police-civilian interaction, 2) the history of the depicted officer, and 3) the criminal history and/or mental health of the depicted civilian significantly affected respondents’ assessments of the officers’ behavior.  In a study of Chicagoan respondents, Graziano, Schuck, & Martin (2010) found that viewing a televised discussion about a recent racial profiling incident in the city that presented competing interpretations about the justifiability of the officer’s actions did not significantly affect respondents’ beliefs about racial profiling.  In a working paper, Testa and Dietrich  (2017) showed respondents a video of a real-life traffic stop between a white officer and a black civilian who claimed he was racially profiled, and they manipulated the information concurrently presented in order to state that the video either confirmed or refuted the civilian’s claim of profiling.  They found that respondents were more likely to conform their own assessment of the stop to align with the interpretation of the video, and this effect was heightened in a condition that further presented this information as if it were part of a newscast.  However, the authors found no effect of frame exposure on respondents’ more general attitudes toward the police.  Relatedly, Miethe, Venger, & Lieberman (2019) showed participants real-life videos of nonlethal police use of force, and they manipulated whether the video clip was attributed to network TV, a social media post, or source unidentified.  Respondents rated the videos attributed to national TV as more trustworthy than social media videos, but media source did not affect respondents’ opinions about the acts of force depicted in the videos.

Whereas a few of the previously-described studies depicted respectful, procedurally-just behavior by police officers, we are aware of only two studies that depicted officers going beyond respectful behavior.  One of the videos in  Lowrey, Maguire, and Bennett's (2016) study showed an officer engaging in “overaccommodating” behavior with the civilian he stopped; the officer used informal language and used a tone that was apologetic for impinging on the civilian’s time.  They found no significant differences in global or specific attitudes toward the police between respondents in the overaccommodation condition vs. control group respondents who viewed an officer with a professional but brusque manner.  Parry, Moule, & Dario (2017) showed participants videos of an interaction between a white officer and a black civilian, concurrently filmed by both men on their respective cell phones.  The officer stopped the civilian due to a call from a local resident.  The two men briefly spoke, the officer extended a “high five” to the civilian, and then the men parted ways.  Employing a pre-test, post-test design, the authors found that viewing the video improved respondents’ self-reported obligation to obey the law, willingness to cooperate with the police, and perceptions of the degree to which the police, in general, perform procedural justice and treat all citizens fairly.  This study is unique in that it presented a genuinely-friendly interaction, and it demonstrated a significant effect of media exposure on global attitudes toward the police, rather than respondents’ evaluation of solely the behavior of the officer depicted in the media stimulus. 

Finally, we are aware of only one experimental study that tested the effect of exposure to pictures of the police, rather than video.  In the Police Officer Perception Project, Simpson (2017, 2018, 2019) exposed participants to a range of pictures of real police officers that varied the officers’ dress (uniforms vs. civilian clothes), accoutrements (e.g., gloves, vests, batons), and style of patrol (foot, bicycle, vehicles either marked or unmarked).  He found significant differences across visual depictions on several dimensions of respondents’ assessment of the officers, such as their perceived friendliness and approachability.  He did not measure participants’ more global attitudes about the role of police in society.

In total, this small but growing body of experimental studies suggests exposure to images or video of police officers or police-civilian interactions can significantly influence public opinion about the police above-and-beyond people’s own ideology and media self-selection habits.  However, media exposure typically has a much stronger effect on people’s specific assessments of the officer(s) shown in the incident depicted in the media stimulus rather than their more global attitudes toward the police, in general; in fact, several studies find that media treatments have no significant effect on global attitudes (Johnson et al., 2017; Lowrey et al., 2016; Maguire et al., 2017; Testa & Dietrich, 2017).  Furthermore, some studies indicate that the manner in which videos of police-civilian interactions are framed (i.e., the information the viewer is given while viewing the video) also significantly shifts public opinion, which supports critics’ assertion that footage from police body-worn cameras (or civilian cell phones) cannot “speak for itself” (Phillips, 2018).

Much remains to be learned.  Very few of these studies employed true control groups in which respondents were not exposed to any media depiction of police; rather, most contrasted the opinions of respondents who saw videos that varied the behavior of the officers depicted.  Such studies cannot estimate “baseline” public opinion about the police.  Furthermore, many of these studies analyzed data from convenience samples of undergraduate college students, rendering their generalizability unclear  (Boivin et al., 2017; Johnson et al., 2017; Lowrey, Maguire, & Bennett, 2016; Maguire et al., 2017; Parry et al., 2017; Reisig et al., 2018; Simpson, 2017, 2018, 2019).  Scholars’ attention has predominantly shifted to examining the effects of videos even though static pictures continue to be embedded in news stories and can now be shared instantaneously by individuals via social media platforms; as such, we lack solid empirical evidence to scientifically judge whether media effects (if they exist) vary across media type. Generally, effects found for video tend to also be found for still pictures used in media stories (compared to text-only articles or statistical presentations) (Arguel & Jamet, 2009; Du et al., 2019; Sundar, 2000; Tran, 2015).  Depending on the outcome measure involved, video does not always outperform text in terms of impact on an audience (Adelaar et al. 2010). Finally, most studies of media effects on public opinion about the police focused on traffic stops or use-of-force against singular or small numbers of civilians.  We are aware of only two studies that tested how friendly police interactions influence public opinion, and we are not aware of any prior studies that tested how people react to images of “militarized” police confrontations with groups of protesters.  Our study contributes to this literature by 1) analyzing data from a national sample, 2) comparing the attitudes of respondents shown an image of police-civilian interactions to respondents who were not shown any image stimuli, and 3) testing the effect of exposure to images of militarized and friendly police-civilian interactions.



Echoing the concerns expressed by politicians and pundits during the Ferguson and Baltimore protests, we are most interested in testing whether seeing an image of confrontation between armed police officers and protesters affects public opinion.  However, conflicts between police and people participating in mass protests are relatively rare in the scope of the millions of interactions that occur between police officers and civilians each year (Davis et al., 2018).  As such, we employed an online survey experiment designed to expose participants to images that represent the full spectrum of police-civilian relations (Weisburd & Eck, 2004). The first picture was an image of two male police officers in riot gear atop an armored vehicle pointing a rifle at a group of protesters in the street with their hands raised; this was a picture taken during the 2014 protests in Ferguson, Missouri following the shooting of Michael Brown.  We chose this image to represent hostile, “militarized” conflict between police and civilians.  The second picture showed two smiling police officers (one male, one female), one of whom was giving a “high five” to a civilian seated on his porch.  We chose this image to represent positive, “community policing-style” interaction between police and civilians.3 Finally, the third picture showed two male police officers patting down two male civilians who had their hands pressed against a wall.  We chose this image to represent a confrontational, “stop-and-frisk-style” interaction between police and civilians.4

The experimental manipulations were embedded on the introductory page of the survey.  Above the survey’s title (“a study of public opinion about government and police”), treatment group participants saw a banner that contained three pictures.  Two of the pictures were constant across all three groups; these were a picture of Members of the U.S. House of Representatives delivering remarks at a podium emblazoned with the House seal and a picture of members of the U.S. National Guard in camouflage fatigues handing out care packages to a mother and child.  These constant images of government professionals were designed to slightly obscure the study’s central focus on images of the police in order to minimize the possibility that respondents would guess the intent of the pictures and reply as they thought the surveyors desired.  The third image, placed in the center of the banner, contained one of the three police-civilian interaction pictures.  Respondents were randomly assigned to receive one of the three treatment pictures or to receive a control condition that contained only the title text without any pictures.

Similar to other studies in this literature (Braga et al., 2014; Culhane et al., 2016; Graziano et al., 2010; Miethe et al., 2019; Parry et al., 2017; Tesla & Dietrich, 2017), we used a Google image search to choose actual pictures of police interacting with civilians taken by journalists.  This means that the treatment stimuli reflect the kinds of pictures that people routinely see online and on the news.  As such, the treatments possess a degree of external validity, but this comes at the cost of precision.  Since the pictures each contain numerous elements, we cannot specify precisely which facet of the images may generate a framing effect on people’s opinions about the police.  Due to the complicated racial dynamics at the center of the policing debate (Peck, 2015; Prowse et al., 2019), we ensured that the pictures represented racial diversity as much as possible.  No picture contained solely white people, and only the community policing treatment picture was racially-homogenous; both of the officers and the civilian in the picture were African American.

We verified treatment exposure in two ways.  First, the title screen was set on a five second delay before participants could proceed, so they could not immediately skip over the pictures embedded in the title screen banner.  Second, following the title screen, participants were reshown each of the three images in the title banner, one by one, and asked to briefly describe what they saw in the picture; participants in the control group proceeded directly to the survey questions.  Only about 1% of respondents in each experimental group provided no descriptions or wrote blatantly-inaccurate descriptions of the pictures they saw, indicating that respondents were successfully treated (Mutz, 2011).  Furthermore, there were no statistically significant differences across groups in regard to respondents’ political party affiliation (χ²[12, N = 1,068] = 19.25, p = 0.08), political ideology (χ²[12, N = 1,025] = 8.46, p = 0.75), self-reported degree of attention they paid to the news in the previous 7 days (χ²[6, N = 1,078] = 6.82, p = 0.34), race (χ²[12, N = 1,100] = 5.73, p = 0.93), or gender (χ²[3, N = 1,097] = 5.39, p = 0.15), which verifies successful random assignment.


We recruited respondents from Qualtrics’ national, online panel to participate in a survey to “study people’s opinions about how well the government and the police are addressing problems facing the nation.”  The survey contained a variety of original questions measuring respondents’ attitudes toward the police, their local government, and the federal government.  The survey was fielded in late April 2016, which means that data collection followed media coverage of the 2014 Ferguson protests and the 2015 Baltimore protests.  We procured responses from 1,100 participants.  Since this is a nonprobability sample, we do not claim that the point estimates of respondents’ attitudes toward the police are precisely generalizable to the national population (Baker et al., 2010, Thompson & Pickett, 2019).  Rather, we were interested in the causal relationship between image exposure and expressed opinion, and studies find that experiments administered to opt-in panel samples generate results that are similar to experiments administered to randomly-selected samples (Mullinix et al., 2015; Weinberg et al., 2014; Yeager & Krosnick, 2012).  We compare our sample to the 2016 U.S. Current Population Survey in the online supplementary materials.  Briefly, this sample over-represents people aged 20 to 34 and 55 to 64 as well as people with a college diploma or more education, but under-represents people aged 35 to 54 and 65 or older as well as people with a high school degree or less education.  The gender and household income distributions of the sample are comparable to the population.  We intentionally over-sampled African Americans and Latinxs in order to have sufficient statistical power for race moderation tests, so these groups are over-represented.


We present detailed information about the operational definitions and descriptive statistics of all our variables in Appendix A.  We employed factor analysis to combine participants’ responses to a series of questions into five dependent variables.  Public opinion about the police is multidimensional; people’s assessments of the performance of the police, in general, tend to be cognitively distinct from their assessments of their local police performing routine tasks (Brandl et al., 1994; Solomon Zhao & Ren, 2015).  We mirrored this distinction in our measures.  Two dependent variables, local and national police effectiveness, measured respondents’ opinions about the quality of the job that their local police or police generally across the nation are doing in routine responsibilities like preventing crime, maintaining order, and addressing citizens’ problems.  Higher values on these scale variables indicated a stronger perception that the police do a “good job” at their duties.  Two variables, local and national police misconduct measured respondents’ beliefs about the frequency with which local or national police engage in inappropriate behavior, such as stopping people without good reason, insulting civilians, or using excessive force.  Higher values on these scale variables indicated a belief that the police engage in improper behavior more frequently.  Finally, national police bias measured respondents’ beliefs about the degree to which police treat some groups of people in society better or worse than other groups.  Higher values on this scale variable indicated a belief that the police treat socially-advantaged groups in society better than socially-disadvantaged groups.

We began our analyses with a series of one-way ANOVA models to test whether or not exposure to one of the treatment images significantly affected participants’ responses to the dependent variable questions.  Following Gauthier & Graziano's (2018) argument that scholars need to pay attention to the ways in which different subgroup audiences in society may react differently to the same media stimuli, we then ran a series of two-way factorial ANOVA models to test whether or not the effect of image exposure significantly differed across respondents’ race, political partisanship, level of typical news consumption, belief about the extent to which police officers routinely face danger in their jobs, and experience being stopped by a police officer within the past 12 months (or lack, thereof).  Finally, both in light of null treatment effects and in order to better compare our findings with those of prior studies of media and public opinion about the police, we ran a series of ordinary least squares regression models that included a range of variables that are commonly correlated with public opinion about law enforcement (Brown & Benedict, 2002).


First, we find no evidence of a direct effect of image exposure.  The one-way, between-subjects ANOVA tests show no significant differences between groups in regard to local police effectiveness [F(3, 1044)=1.20, p=0.31], local police misconduct [F(3, 1025)=0.30, p=0.83], national police effectiveness [F(3, 1048)=0.08, p=0.97], national police misconduct, [F(3, 1041)=0.24, p=0.87], or national police bias [F(3, 1040)=0.14, p=0.94].  For the sake of parsimony, we will not report the full results of the various two-way factorial ANOVA models we ran to test for moderation effects.5 We found no evidence that the effect of image exposure was moderated by a respondent’s race/ethnicity, political party identification, typical volume of news consumption, or belief about the extent to which police officers routinely face danger in their jobs.  We found evidence of only one moderating effect.  The effect of image exposure on respondents’ perceptions of the frequency with which their local police or police across the nation engage in misconduct significantly differed between respondents who had been stopped by the police within the previous 12 months vs. those who had not been stopped [local misconduct: F(3, 1021)=3.16, p=0.02; national misconduct: F(3, 1037)=3.33, p=0.02].  We present this moderation effect graphically in Figure 1.

[Figure 1 about here]

The interaction effect is limited in scope but surprising in nature.  We expected that if image exposure were to particularly heighten perceptions of police misconduct among people who have recently experienced a police stop, it would have been exposure to the militarized police or stop and frisk images.  However, neither of these images triggered significantly different reactions between respondents who had vs. had not experienced a recent police stop.  Instead, only exposure to the image of an officer smiling and high-fiving a civilian triggered divergent reactions among participants, and the effect was counterintuitive.  Respondents who had been stopped by police within the past 12 months who saw the community policing image reported that they perceived significantly more frequent misconduct among local and national police than did respondents in the community policing condition who had not experienced a police stop.  We call this a backlash to the community policing picture.  This moderation effect appears to be robust. We reran the factorial ANOVA and added a Bonferroni correction for multiple comparisons, but the divergence across police stops remained statistically significant.  We also ran a three-way factorial ANOVA to test whether the interaction between image exposure and police stops also varied across the race of the respondents; we found no such three-way interaction.6,7 We ponder the implications of this finding in the discussion section.

Overall, our findings suggest that (brief) exposure to the kinds of pictures of police-civilian interactions that are frequently presented in the media has little effect on people’s opinions about the police.  Our treatment effects were limited to only two of the five dependent variables among only one subgroup of respondents.  So if media image exposure does not shape public opinion about the police, what does?  To answer this question, we ran a series of ordinary least squares regression analyses that included a range of demographic, experiential, and ideological factors that have been shown to be significantly related to public opinion about the police in other studies (Brown & Benedict, 2002).  We test whether these factors are significantly correlated with attitudes toward the police even controlling for experimental image exposure.

[Table 1 about here]

We present the results of the OLS regressions in Table 1.8 Consistent with prior studies, and mirroring the sole moderation effect we describe above, we found that people who were stopped by the police within the past 12 months perceived that the police engage in misconduct more frequently compared to people who did not experience a recent police stop.  We found a positive relationship between people’s perceived frequency of physical assaults against police officers and their assessment of the quality of the routine work done by police across the country, as well as a negative relationship with their perception of police bias against disadvantaged social groups.  However, we also found a positive relationship between perceived police risk and perceptions of local and national police misconduct.

In regard to political beliefs, we found that Democratic respondents perceived greater frequency of local and national police misconduct and slightly more police bias against disadvantaged groups than Republican respondents; we found no significant differences between Republicans and Independents.  We found more pervasive effects of political ideology than partisanship.  Compared to self-identified liberals, moderates expressed significantly more positive evaluations of local and national police and perceived less national police misconduct and bias against disadvantaged groups.  Compared to liberals, conservative respondents expressed significantly more positive evaluations of local and national police and significantly less police bias against disadvantaged groups.

In regard to demographics, non-Hispanic black respondents expressed significantly less positive evaluations of local and national police than whites, as well as greater perceptions of misconduct and bias.  Hispanic respondents only significantly differed from white respondents in their greater perceptions of police misconduct and bias, not their overall evaluations of the routine work done by police.  Asian and Pacific Islander respondents showed the fewest differences from white respondents, only expressing a significantly more negative evaluation of police across the nation.  Older respondents assessed local police more positively and perceived less misconduct than younger respondents, but surprisingly, older respondents perceived slightly more police bias.  Education was only related to evaluation of the local police; respondents with at least some college or more education expressed slightly more positive evaluations than respondents with a maximum of a high school degree.  Wealthier respondents expressed significantly more positive evaluation of police across the nation, and women perceived slightly more police bias than men.  Midwesterners and Westerners evaluated the local and national police more negatively than New Englanders, but there were no differences in perceptions of misconduct or bias.

Lastly, even controlling for experimental image exposure, general news consumption was significantly related to four of the dependent variables.  Respondents who consumed a greater volume of news from a greater variety of sources in a typical week both perceived more misconduct and evaluated police more positively at both the local and national levels compared to respondents who consumed less news.  We contemplate the contrast between our (lack of) media treatment effects and the significance of the general, self-selected measure of news consumption in the discussion section.


We reiterate that virtually all respondents passed our manipulation check by accurately describing the content of the image to which they were exposed, so it is very unlikely that our null experimental findings can be attributed to a failure to treat.  However, Osborne (2013) cautions that respondents who do not give a survey their full attention, either by speeding through the questions or becoming distracted and answering the questions in fits and starts, can degrade the validity and reliability of the data and dilute treatment effects.  To address this possibility, we identified respondents who fell within the fifth (about 6 minutes) and ninety-fifth percentiles (about 34 minutes) of the survey completion time distribution (the mean completion time was about 20 minutes, and the median time was about 12 minutes).  We assume that the most inattentive or distracted respondents were likely contained within these groups of the fastest and slowest respondents.  We dropped these 105 respondents and reran all analyses.  The substantive results were unchanged.


With the advent of social media and a renewed public debate about race and policing catalyzed by the Black Lives Matter movement (Lowery, 2016), it is more important than ever for scholars (and the police themselves) to properly understand the relationship between media consumption and public opinion about the police.  However, a significant challenge to this inquiry is the problem of selection bias: does media shape people’s perceptions of the police, or do they selectively choose to consume media that echoes and reinforces the perceptions of police that they already hold (Gauthier & Graziano, 2018)?  Experimental methods are the best tool in social scientists’ toolkit to overcome selection bias.  This study contributes to the small but growing body of experimental studies that test the effect of exposure to media stimuli on public opinion about law enforcement.

Our major finding is null results.  We found no statistically significant main effects of exposure to an image of police officers interacting with civilians on respondents’ perceptions of police misconduct, police bias, or evaluation of the manner in which police complete routine tasks and responsibilities.  Nor did we find that the effect of image exposure significantly varied across respondents’ race/ethnicity, political party identification, typical news consumption habits, or belief about the extent to which police officers routinely face danger in their jobs.  Rather, our non-experimental regression analyses suggest that public opinion about the police is primarily shaped by personal experiences with the police (through the experience of being stopped); personal demographic and ideological characteristics; and routine, self-selected news consumption habits.  In fact, our results suggest that group identities may be even more important than personal experiences with the police, as political partisanship, political ideology, race/ethnicity, and age were the most consistent predictors of variation on attitudes toward the police in our regressions.  Our findings mirror those of other scholars and suggest that Americans’ attitudes about the police are politicized and divided across racial groups and generations (Fine, Rowan, & Simmons, 2019; Gabbidon & Higgins, 2009; Graziano & Gauthier, 2019; Peck, 2015; Silver & Pickett, 2015).  This is consistent with the “neo-Durkheimian” perspective that many people primarily look to and evaluate the police as symbols of social order, so attitudes toward the police will likely mirror broader cleavages across groups in society at the nexus of beliefs about norms, values, and justice (Jackson & Bradford, 2009; Jackson & Sunshine, 2007).

It is important to delineate the scope of our findings, which are likely driven by a combination of the nature of our treatment and the nature of our dependent variables.  Other recent experimental studies have found that exposure to a media stimulus that depicts a police-civilian interaction does significantly affect respondents’ expressed opinions about the police.  It is important to note, though, that these prior studies employed video treatments (Boivin et al., 2017; Johnson et al., 2017; Lowrey et al., 2016; Maguire et al., 2017; Parry et al., 2017).  It is quite possible that exposure to a static picture of police-civilians interactions (such as we employed in this study) cannot make a sufficient impact to shift a person’s opinions about the quality of police tactics and behavior, even though photos are still routinely disseminated by both traditional and social media outlets (e.g., “Photo essay,” 2014).  Furthermore, it is important to note that these prior experiments that did demonstrate some kind of media stimulus exposure effect most commonly found that respondents’ opinions about the specific police-civilian interaction depicted in the video varied depending upon the nature of the content (i.e., officer and/or civilian behavior) shown.  In contrast, respondents’ more global assessments about the police, in general, were rarely affected by video exposure.  Knowing that global vs. specific attitudes toward the police are distinct (Brandl et al., 1994; Solomon Zhao & Ren, 2015), we did separately measure our respondents’ opinions about their local police forces vs. police across the nation, more generally.  We anticipated that image exposure might affect opinions about the police in general but not respondents’ local police.  However, we did not find this to be the case.  We also did not provide additional information about our treatment images, unlike some prior studies (Braga et al., 2014; Culhane et al., 2016; L. Graziano et al., 2010; Testa & Dietrich, 2017).  Finally, ours is the first study of which we are aware to test the effect of exposure to an image of police confronting protesters in a hostile manner, and only the third study to depict an explicitly friendly, rather than merely courteous, interaction between an officer and a civilian (Lowrey et al., 2016; Parry et al., 2017).  Though scholars and pundits expressed concern about the “militarized” response of the police forces to the protests in Ferguson and Baltimore, it is quite possible that most Americans perceived those confrontations as distinct from instances of officer misconduct.  In fact, many Americans may have seen nothing wrong with police forces’ tactics in these instances (e.g., Pew Research Center, 2014).  We did not probe our respondents’ interpretations or evaluations of the militarized police image, and so this remains an open, empirical question.

One alternative interpretation of our null findings is that a static picture is too weak of a treatment to alter public opinion.  We reject this interpretation for three reasons.  First, in a separate analysis of these data, we found that image exposure did significantly alter respondents’ [socio-political attitude and citation blinded for peer review].  Second, Simpson (2017, 2018, 2019) did find significant effects of exposure to static pictures of police officers, but he only assessed respondents’ ratings of the qualities and characteristics of the depicted officers.  Third, studies of media effects on other topics demonstrate that exposure to static pictures can generate similar effects as exposure to dynamic videos (Arguel & Jamet, 2009; Du et al., 2019; Sundar, 2000; Tran, 2015).  So it is not the case that exposure to a static image of police is completely incapable of affecting public opinion.  Rather, based upon the present findings and the findings of the other experiments we reviewed, it seems that neither image nor video exposure can alter global assessments of the police.

A second alternative interpretation is that respondents were already so “pretreated” by recent news coverage of police-civilian confrontations that our treatment could not possibly have an additional effect (Gaines et al., 2007).9 We cannot wholly rebut this critique since we have no empirical measure of the volume of respondents’ prior exposure to news stories of police-civilian conflict, specifically.  However, we refer readers to our previous point that some evidence, gathered from surveys fielded in the midst of high-salience media coverage of police-civilian conflicts, does show that exposure to images of police can affect some types of people’s expressed opinions, which means that media treatments can have an effect above-and-beyond routine media consumption.  We also remind readers that we tested whether image exposure was moderated by a person’s typical volume of news consumption, and we found no significant differences.  If the pretreatment effect were applicable in this case, we would have expected image exposure to affect the opinions of people who infrequently pay attention to the news, since they are the least likely to have been previously exposed to a critical mass of stories and images about police-civilian conflicts.

Our findings indicate that brief exposure to static pictures of police-civilian interactions disseminated through the media may be less likely to influence public opinion about law enforcement than dynamic video clips.  This may be a function of the medium itself, given the inherent appeal that moving images have in drawing cognitive and emotive attention with attendant effects on information recall and use (Calfano & Kruse, 2016; Detenber et al., 1998).  At the same time, it may be that images must be accompanied by text that gives additional information about the circumstances pictured or presents a particular interpretation (i.e., a frame) about the image in order to impact public opinion. There is support for this combined approach in the literature on media format effects (e.g., Adelaar et al., 2010), and our findings might mean this approach is particularly effective when it comes to law enforcement, use of force, police/community relations, and similar topics.  

We argue that our findings point toward avenues for improving scholars’ use of cultivation theory in the study of media effects on public opinion about the police.  Our null treatment effects stand alongside other studies which found that exposure to videos of police-civilian interactions only influenced people’s assessment of the specific, narrow circumstances depicted in the particular video they viewed (Johnson et al., 2017, 2017; Maguire et al., 2017).  Collectively, these experimental findings suggest that media coverage of law enforcement cannot alter people’s global assessments of the police and policing practices, in general.  However, our nonexperimental analyses tell a different story.  There we found that frequency of routine news consumption was significantly related to our respondents’ evaluation of the police and perceptions of the frequency of police misconduct at both the local and national levels, which are global attitudes about the police.

Perhaps this discrepancy can be explained by the fact that experimental media treatments are almost always short and discrete in nature whereas measures of routine news consumption capture cultivation theory’s emphasis on accumulated media exposure that shapes viewers’ worldviews and corpus of accessible cognitive facts and considerations about public issues like policing.  Correlational tests of routine media consumption cannot clearly address the selection bias problem, but they do suggest that it takes a particular volume of media consumption over time to shape (or reinforce) people’s global attitudes about the police.  People seem not to generalize the conclusions they form after viewing depictions of particular policing incidents beyond the officers and civilians involved in that discrete event unless, perhaps, they perceive those incidents to be examples of a wider pattern of police practices they observe in numerous news stories.  We encourage scholars to further test our interpretation with additional mixed-method studies that incorporate both experimental exposure to particular media stimuli and correlational measures of routine media consumption.  In particular, future research should explore the content and types of media coverage about police, crime, and criminal justice in greater depth and specificity to better understand the kinds of stories on those topics that are consumed by people who frequently watch or read the news.10

Another finding also merits further inquiry.  Though we consider our null experimental results to be the predominant finding, we did find one single treatment effect: respondents who had been stopped by a police officer within the past 12 months who saw the community policing image reported that they believed misconduct among both their local police force and police, in general, was more common than respondents who saw the same image but had not been recently stopped.  This result were unexpected.  We hypothesize that these respondents felt that the officer(s) who stopped them behaved in a procedurally unjust manner, and when they saw the positive, community policing image, they reacted with cynicism.11 Because their personal experience was negative, they dismissed the image of a friendly police-civilian interaction as disingenuous and instead asserted that the police misbehave.  Unfortunately, due to data limitations, we cannot empirically test or confirm this hypothesis (see footnote 7); rather, we must leave it as an avenue for future research.   This is particularly true since the only prior experimental studies of which we are aware that depicted friendly or “overaccommodating” officer behavior produced conflicting results (Lowrey et al., 2016; Parry et al., 2017).  Future studies should devote more attention to the manner in which the public perceives and reacts to officers’ attempts to be not just professional and respectful, but to be truly warm and friendly.12  The fact that media exposure generated a cynical reaction toward police among respondents who had personal contact with officers suggests that policing scholars who are interested in media effects should more carefully consider the intersection of cultivation theory and procedural justice theory.  In other words, contact with the police (and civilians’ assessment of the nature of that contact) could be a key “audience characteristic” that conditions how people interpret news coverage of the police.


Adelaar, T., Chang, S., Lancendorfer, K.M., Byoungkwan, L., & Morimoto, M. (2010). Effects of media formats on emotions and impulse buying intent.” Journal of Information Technology,  18(4), 247-266.

Arguel, A., & Jamet, E. (2009). Using video and static pictures to improve learning of procedural contents. Computers in Human Behavior, 25(2), 354–359.

Balko, R. (2013). Rise of the Warrior Cop: The Militarization of America’s Police Forces. PublicAffairs.

Boivin, R., Gendron, A., Faubert, C., & Poulin, B. (2017). The malleability of attitudes toward the police: Immediate effects of the viewing of police use of force videos. Police Practice and Research, 18(4), 366–375.

Bosman, J., & Apuzzo, M. (2014, August 14). In Wake of Clashes, Calls to Demilitarize Police. The New York Times.

Bowman, Q. (2014, September 9). Post Ferguson, senators question military-to-police weapons programs. PBS NewsHour.

Braga, A. A., Brunson, R. K., & Drakulich, K. M. (2019). Race, Place, and Effective Policing. Annual Review of Sociology, 45(1), 535–555.

Braga, A. A., Winship, C., Tyler, T. R., Fagan, J., & Meares, T. L. (2014). The Salience of Social Contextual Factors in Appraisals of Police Interactions with Citizens: A Randomized Factorial Experiment. Journal of Quantitative Criminology, 30(4), 599–627.

Brandl, S. G., Frank, J., Worden, R. E., & Bynum, T. S. (1994). Global and specific attitudes toward the police: Disentangling the relationship. Justice Quarterly, 11(1), 119–134.

Brown, B., & Benedict, W. R. (2002). Perceptions of the police. Policing: An International Journal of Police Strategies & Management, 25(3), 542–580.

Calfano, B. R., & Kruse, A. (2016). Beyond Surveillance: The Effects of Issue Ad Vividness and Anxiety on Information Use. American Politics Research, 44(6), 1098–1122.

Callanan, V. J., & Rosenberger, J. S. (2011). Media and public perceptions of the police: Examining the impact of race and personal experience. Policing and Society, 21(2), 167–189.

Chermak, S., McGarrell, E., & Gruenewald, J. (2006). Media coverage of police misconduct and attitudes toward police. Policing: An International Journal of Police Strategies & Management.

Chermak, S., & Weiss, A. (2006). Community Policing in the News Media. Police Quarterly, 9(2), 135–160.

Cobbina, J. E. (2019). Hands Up, Don’t Shoot: Why the Protests in Ferguson and Baltimore Matter, and How They Changed America. New York University Press.

Cowart, H. S., Saunders, L. M., & Blackstone, G. E. (2016). Picture a Protest: Analyzing Media Images Tweeted From Ferguson: Social Media + Society.

Culhane, S. E., Boman, J. H., & Schweitzer, K. (2016). Public Perceptions of the Justifiability of Police Shootings: The Role of Body Cameras in a Pre- and Post-Ferguson Experiment. Police Quarterly, 19(3), 251–274.

Davis, E., Whyde, A., & Langton, L. (2018). Contacts between Police and the Public, 2015. (NCJ 251145). Office of Justice Programs.

Desmond, M., Papachristos, A. V., & Kirk, D. S. (2016). Police Violence and Citizen Crime Reporting in the Black Community. American Sociological Review, 81(5), 857–876.

Detenber, B. H., Simons, R. F., & Jr, G. G. B. (1998). Roll ‘em!: The effects of picture motion on emotional responses. Journal of Broadcasting & Electronic Media, 42(1), 113–127.

Donovan, K. M., & Klahm, C. F. (2015). The Role of Entertainment Media in Perceptions of Police Use of Force. Criminal Justice and Behavior, 42(12), 1261–1281.

Dowler, K. (2002). Media Influence on Citizen Attitudes Toward Police Effectiveness. Policing and Society, 12(3), 227–238.

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.

Dowler, K., & Zawilski, V. (2007). Public perceptions of police misconduct and discrimination: Examining the impact of media consumption. Journal of Criminal Justice, 35(2), 193–203.

Du, Y. R., Zhu, L., & Cheng, B. K. L. (2019). Are Numbers Not Trusted in a “Post-Truth” Era? An Experiment on the Impact of Data on News Credibility: Electronic News.

Eschholz, S., Blackwell, B. S., Gertz, M., & Chiricos, T. (2002). Race and attitudes toward the police: Assessing the effects of watching “reality” police programs. Journal of Criminal Justice, 30(4), 327–341.

Fine, A. D., Rowan, Z., & Simmons, C. (2019). Do politics Trump race in determining America’s youths’ perceptions of law enforcement? Journal of Criminal Justice, 61, 48–57.

Fox, J. (1991). Regression Diagnostics: An Introduction. SAGE.

Gabbidon, S. L., & Higgins, G. E. (2009). The Role of Race/Ethnicity and Race Relations on Public Opinion Related to the Treatment of Blacks by the Police. Police Quarterly, 12(1), 102–115.

Gaines, B. J., Kuklinski, J. H., & Quirk, P. J. (2007). The logic of the survey experiment reexamined. Political Analysis, 15(1), 1–20.

Gately, G., & Stolberg, S. G. (2015, November 16). Baltimore Police Assailed for Response After Freddie Gray’s Death. The New York Times.

Gauthier, J. F., & Graziano, L. M. (2018). News media consumption and attitudes about police: In search of theoretical orientation and advancement. Journal of Crime and Justice, 41(5), 504–520.

Gerbner, G. (1970). Cultural Indicators: The Case of Violence in Television Drama. The ANNALS of the American Academy of Political and Social Science, 388(1), 69–81.

Graziano, L. M., & Gauthier, J. F. (2019). Examining the racial-ethnic continuum and perceptions of police misconduct. Policing and Society, 29(6), 657–672.

Graziano, L., Schuck, A., & Martin, C. (2010). Police Misconduct, Media Coverage, and Public Perceptions of Racial Profiling: An Experiment. Justice Quarterly, 27(1), 52–76.

Hirschfield, P. J., & Simon, D. (2010). Legitimating police violence: Newspaper narratives of deadly force. Theoretical Criminology.

Holmes, M. D., Painter, M. A., & Smith, B. W. (2019). Race, Place, and Police-Caused Homicide in U.S. Municipalities. Justice Quarterly, 36(5), 751–786.

Intravia, J., Wolff, K. T., & Piquero, A. R. (2018). Investigating the Effects of Media Consumption on Attitudes Toward Police Legitimacy. Deviant Behavior, 39(8), 963–980.

Iyengar, S., & Hahn, K. S. (2009). Red Media, Blue Media: Evidence of Ideological Selectivity in Media Use. Journal of Communication, 59(1), 19–39.

Jackson, J., & Bradford, B. (2009). Crime, policing and social order: On the expressive nature of public confidence in policing. The British Journal of Sociology, 60(3), 493–521.

Jackson, J., & Sunshine, J. (2007). Public Confidence in Policing: A Neo-Durkheimian Perspective. The British Journal of Criminology, 47(2), 214–233.

Jefferis, E. S., Kaminski, R. J., Holmes, S., & Hanley, D. E. (1997). The effect of a videotaped arrest on public perceptions of police use of force. Journal of Criminal Justice, 25(5), 381–395.

Johnson, D., Wilson, D. B., Maguire, E. R., & Lowrey-Kinberg, B. V. (2017). Race and Perceptions of Police: Experimental Results on the Impact of Procedural (In)Justice. Justice Quarterly, 34(7), 1184–1212.

Kaminski, R. J., & Jefferis, E. S. (1998). The effect of a violent televised arrest on public perceptions of the police. Policing: An International Journal of Police Strategies & Management.

Kraska, P. B. (2007). Militarization and Policing—Its Relevance to 21st Century Police. Policing: A Journal of Policy and Practice, 1(4), 501–513.

Kraska, P. B., & Kappeler, V. E. (1997). Militarizing American Police: The Rise and Normalization of Paramilitary Units. Social Problems, 44(1), 1–18.

Lasley, J. R. (1994). The impact of the Rodney king incident on citizen attitudes toward police. Policing and Society, 3(4), 245–255.

Lowery, W. (2016). They Can’t Kill Us All: Ferguson, Baltimore, and a New Era in America’s Racial Justice Movement. Little, Brown.

Lowrey, B. V., Maguire, E. R., & Bennett, R. R. (2016). Testing the Effects of Procedural Justice and Overaccommodation in Traffic Stops: A Randomized Experiment. Criminal Justice and Behavior, 43(10), 1430–1449.

Maguire, E. R., Lowrey, B. V., & Johnson, D. (2017). Evaluating the relative impact of positive and negative encounters with police: A randomized experiment. Journal of Experimental Criminology, 13(3), 367–391.

Miethe, T. D., Venger, O., & Lieberman, J. D. (2019). Police use of force and its video coverage: An experimental study of the impact of media source and content on public perceptions. Journal of Criminal Justice, 60, 35–46.

Mirko, M. (2013, July 3). Pictures: New Haven Community Policing. Courant.Com.

Morgan, M., & Shanahan, J. (2010). The State of Cultivation. Journal of Broadcasting & Electronic Media, 54(2), 337–355.

Mullinix, K. J., Leeper, T. J., Druckman, J. N., & Freese, J. (2015). The Generalizability of Survey Experiments. Journal of Experimental Political Science, 2(2), 109–138.

Mutz, D. C. (2011). Population-Based Survey Experiments. Princeton University Press.

Osborne, J. W. (2013). Best Practices in Data Cleaning: A Complete Guide to Everything You Need to Do Before and After Collecting Your Data. SAGE Publications.

Parry, M. M., Moule, R. K., & Dario, L. M. (2017). Technology-Mediated Exposure to Police–Citizen Encounters: A Quasi-Experimental Assessment of Consequences for Citizen Perceptions. Justice Quarterly, 0(0), 1–25.

Peck, J. H. (2015). Minority perceptions of the police: A state-of-the-art review. Policing: An International Journal of Police Strategies & Management.

Peffley, M., & Hurwitz, J. (2010). Justice in America: The Separate Realities of Blacks and Whites. Cambridge University Press.

Pew Research Center. (2014). Stark Racial Divisions in Reactions to Ferguson Police Shooting. Author.

Phillips, S. W. (2018). Eyes are not Cameras: The Importance of Integrating Perceptual Distortions, Misinformation, and False Memories into the Police Body Camera Debate. Policing: A Journal of Policy and Practice, 12(1), 91–99.

Photo essay: Police and protesters in Ferguson. (2014, August 14). Stltoday.Com.

Post Editorial Board. (2015, June 1). How many New Yorkers must die before the mayor brings back stop-and-frisk?

Potterf, J. E., & Pohl, J. R. (2018). A Black Teen, a White Cop, and a City in Turmoil: Analyzing Newspaper Reports on Ferguson, Missouri and the Death of Michael Brown: Journal of Contemporary Criminal Justice.

Prowse, G., Weaver, V. M., & Meares, T. L. (2019). The State from Below: Distorted Responsiveness in Policed Communities. Urban Affairs Review, 1078087419844831.

Reisig, M. D., Mays, R. D., & Telep, C. W. (2018). The effects of procedural injustice during police–citizen encounters: A factorial vignette study. Journal of Experimental Criminology, 14(1), 49–58.

Rios, V. M. (2011). Punished: Policing the lives of black and Latino boys. New York University Press.

Roche, S. P., Pickett, J. T., & Gertz, M. (2016). The Scary World of Online News? Internet News Exposure and Public Attitudes Toward Crime and Justice. Journal of Quantitative Criminology, 32(2), 215–236.

Romer, D., Jamieson, K. H., & Aday, S. (2003). Television News and the Cultivation of Fear of Crime. Journal of Communication, 53(1), 88–104.

Sigelman, L., Welch, S., Bledsoe, T., & Combs, M. (1997). Police Brutality and Public Perceptions of Racial Discrimination: A Tale of Two Beatings. Political Research Quarterly, 50(4), 777–791.

Silver, J. R., & Pickett, J. T. (2015). Toward a Better Understanding of Politicized Policing Attitudes: Conflicted Conservatism and Support for Police Use of Force. Criminology, 53(4), 650–676.

Simpson, R. (2017). The Police Officer Perception Project (POPP): An experimental evaluation of factors that impact perceptions of the police. Journal of Experimental Criminology, 13(3), 393–415.

Simpson, R. (2018). Officer Appearance and Perceptions of Police: Accoutrements as Signals of Intent. Policing: A Journal of Policy and Practice.

Simpson, R. (2019). Police vehicles as symbols of legitimacy. Journal of Experimental Criminology, 15(1), 87–101.

Solomon Zhao, J., & Ren, L. (2015). Exploring the Dimensions of Public Attitudes Toward the Police. Police Quarterly, 18(1), 3–26.

Sundar, S. S. (2000). Multimedia Effects on Processing and Perception of Online News: A Study of Picture, Audio, and Video Downloads. Journalism & Mass Communication Quarterly, 77(3), 480–499.

Testa, P., & Dietrich, B. J. (2017, September). Seeing is believing: How vide of police action affects criminal justice beliefs. Annual Meeting of the American Political Science Association, San Francisco, CA. Available at:

Thompson, A., & Pickett, J. T. (n.d.). Are relational inferences from crowdsourced and opt-in samples generalizable? Comparing criminal justice attitudes in the GSS and five online samples. Journal of Quantitative Criminology.

Topaz, J. (2014, August 14). Critics slam “militarization” of police. POLITICO.

Tran, H. L. (2015). More or Less? Multimedia Effects on Perceptions of News Websites. Electronic News, 9(1), 51–67.

Tyler, T. R. (2006). Why people obey the law (Vol. 2nd). Princeton University Press.

Weinberg, J., Freese, J., & McElhattan, D. (2014). Comparing Data Characteristics and Results of an Online Factorial Survey between a Population-Based and a Crowdsource-Recruited Sample. Sociological Science;  Stanford, 1, 292–310.

Weisburd, D., & Eck, J. E. (2004). What can police do to reduce crime, disorder, and fear? The Annals of the American Academy of Political and Social Science, 593(1), 42–65.

Weitzer, R. (2002). Incidents of police misconduct and public opinion. Journal of Criminal Justice, 30(5), 397–408.

Weitzer, R., & Kubrin, C. E. (2004). Breaking news: How local TV news and real-world conditions affect fear of crime. Justice Quarterly, 21(3), 497–520.

Weitzer, R., & Tuch, S. A. (2004a). Reforming the Police: Racial Differences in Public Support for Change*. Criminology, 42(2), 391–416.

Weitzer, R., & Tuch, S. A. (2004b). Race and Perceptions of Police Misconduct. Social Problems, 51(3), 305–325.

Yeager, D. S., & Krosnick, J. A. (2012). Does Mentioning “Some People” and “Other People” in an Opinion Question Improve Measurement Quality? Public Opinion Quarterly, 76(1), 131–141.


[1] Some studies jointly examined the effects of news and TV show (both fictional and reality) consumption: Callanan & Rosenberger (2011), Dowler (2002, 2003), Dowler & Zawilski (2007).

[2] Some of the studies that we review here are framed as examinations of procedural justice rather than media effects.  We consider laboratory- or survey-based experimental studies whose treatment stimuli were either videos or pictures of police-civilian interactions to be sufficiently comparable and pertinent to warrant comparison to self-identified studies of media effects.  We do not consider studies of procedural justice that employed field experiments in which the behavior of police officers was manipulated in the context of in-person interactions with civilians (for a review of pertinent studies, see Johnson, Wilson, Maguire, & Lowrey-Kinberg, 2017).

[3] In reality, of course, “community policing” is an umbrella term that is used to refer to a wide variety of quite disparate policies and practices (many of which do not involve positive interactions between police and citizens).  Our intent was not to capture response to community policing broadly but rather response to one idealized and abstract image of a seemingly positive interaction between a police officer and a citizen, given how few prior studies have shown people depictions of officers acting in a friendly or helpful manner.

[4] The treatment images may be viewed on the following websites, which are the sources from which we copied the pictures: militarized policing (Topaz, 2014), community policing (Mirko, 2013), stop and frisk (Post Editorial Board, 2015).  The original stop and frisk image depicted an officer training exercise, and the officers were carrying bright blue model side arms.  We digitally altered the picture in order to color the handgun hilts black so that they would look like regular guns, thereby depicting a typical stop and frisk.

[5] The full factorial ANOVA results are available upon request.

[6] This null result is not for lack of racial diversity in stops.  Twenty percent of white respondents, 13% of black respondents, 19% of Hispanic respondents, and 8% of Asian respondents reported being stopped within the previous 12 months.

[7] We did ask four follow-up questions to measures respondents’ perception of the presence or absence of procedurally-just behavior on the part of the officer(s) who stopped them.  However, while 169 respondents reported being stopped within the past 12 months, unfortunately only eight of those respondents answered the follow-up questions, giving us too little data to use these measures.  We suspect an implementation flaw; we programmed a skip function into the survey so that only respondents who reported being stopped would be asked the follow-up questions, but it looks like only a handful of respondents were presented with those items.

[8] A number of the coefficients generated p values between 0.051 and 0.059, just above the commonly-accepted threshold of p ≤ 0.05.  We report these effects as marginally-significant.  We employed robust standard errors because a Breusch-Pagan test indicated that four of the five models demonstrated heteroscedasticity.  We also estimated variance inflation factor (VIF) scores to test for the presence of multicollinearity; no variable in any model generated a VIF greater than 2.42, which falls below the standard thresholds of concern for multicollinearity (Fox, 1991).

[9] The study was fielded just before the wave of protests in the summer of 2016 but well after the widely-covered protests in Ferguson beginning in 2014 and Baltimore in the spring of 2015. 

[10] To our knowledge, content analyses of news coverage of police are rarely paired with analyses of public opinion; see, for example, Chermak and Weiss (2006), Cowart, Saunders, and Blackstone (2016),  Hirschfield and Simon (2010), and Potterf and Pohl (2018).

[11] As we discussed above, the image represents a necessarily abstracted and idealized version of community policing, meaning our test gauged reactions to this idealized positive image rather than to the full breadth of quite disparate community policing activities implemented in different departments (some of which may in fact produce negative interactions simply by bringing police and civilians into increased contact). 

[12] Simpson (2017) presents a potentially pertinent finding.  His respondents evaluated police officers presented in civilian clothing as less approachable, accountable, and respectful than officers presented in uniform.  Together with our finding that some respondents “backlashed” against the community policing image, this raises the possibility that citizens may, under some circumstances, react negatively if they perceive that police officers are trying to be overly friendly rather than professional and authoritative.  This evidence supports Lowrey, Maguire, and Bennett's (2016) hypotheses regarding “overaccommodation.”


No comments here
Why not start the discussion?