Abstract
In this study, associations between overall amount of television viewing as well as viewing of reality programs featuring adults in romantic, friendship-oriented, or familial settings and the approval of physical and verbal aggression are examined. Consistent with genre-specific cultivation theory, findings among 248 U.S. adult survey respondents show the ability of exposure to docusoap reality television as well as its perceived reality to predict normative beliefs about aggression, even under multiple controls. Additional analyses explore differences between men and women in the sample and explore approval of female-perpetrated compared with male-perpetrated aggression.
As the primary storyteller of the modern environment, television has the ability to help establish among its viewers a sense of what lies within as well as what lies outside the boundaries of cultural norms and expectations. Television has long been shown to shape both gender-role conceptions (see Oppliger, 2007, for a meta-analysis) and aggression (see Bushman & Huesmann, 2012, for a review), two themes prevalent in the stories told through the medium. Yet, relatively few studies combine the concepts of gender and aggression to ask whether television use might be associated with perceptions of the normativity or acceptability of aggression in general and as perpetrated by men and women, respectively. With cultural norms that support aggression a key contributor to violence (World Health Organization [WHO], 2010), and indications of a closing of the gender gap in physical violence perpetration rates (Lauritsen, Heimer, & Lynch, 2009), the topic is of strong social significance.
Reality television is a staple in contemporary television and is a site of a considerable amount of aggression enacted by its male and female stars (Coyne, Robinson, & Nelson, 2010; Lauzen & Dozier, 2008; Wilson, Robinson, & Callister, 2012). Docusoaps—a subgenre featuring adults in romantic, friendship, and familial relationships in everyday domestic settings—are a much critiqued yet highly popular program type known especially for dramatic displays of conflict that are part of an ostensibly true-to-life account of individuals’ daily interactions (Edwards, 2013; Rifon, Smreker, & Kim, 2014). The current survey study examines both overall television use and the viewing of docusoap reality TV as predictors of viewers’ normative beliefs about the appropriateness of physical and verbal forms of aggression. In doing so, the study further tests cultivation theory, a theory that has seen a great deal of transformation across decades of inquiry and critique (Morgan, Shanahan, & Signorielli, 2015; Potter, 2014).
Literature Review
Gender Differences in Aggression on Television and in the Real World
In real-world populations, males tend to report performing physical and verbal aggression more frequently than females (Bailey & Ostrov, 2008; Card, Stucky, Sawalani, & Little, 2008; Dodge, Coie, & Lynam, 2006) and are more likely to perceive physical aggression to be a normative or routine pattern of behavior (Bailey & Ostrov, 2008; Krahé & Busching, 2014). Physical aggression perpetrated by males can also be seen as more acceptable than physical aggression perpetrated by females (Harris, 1994). Attitudes are often precursors to aggressive actions in that “normative beliefs serve to regulate corresponding actions by prescribing the range of allowable and prohibited behaviors” (Huesmann & Guerra, 1997, p. 409).
On television, male characters have consistently been found to be more frequent perpetrators of physical aggression compared with female characters (Glascock, 2001; Signorielli, 2003; Smith et al., 1998). Hetsroni (2007) performed a meta-analysis on 57 content analyses of (physical) violence on U.S. primetime network television and found the proportion of females ranged only between 10% and 26% of those characters performing violence over time. The evidence is more mixed for rates of engagement in verbal aggression by gender on television (Glascock, 2001, 2003; Greenberg, Edison, Korzenny, Fernandez-Collado, & Atkin, 1980).
Although limited in number, content analyses suggest that reality television is marked by a substantial amount of aggressive content in both physical and verbal forms (Baruh, 2009; Coyne et al., 2010; Smith, Nathanson, & Wilson, 2002; Wilson et al., 2012). Some content analyses have examined gender differences in the display of aggression on reality television, with mixed evidence for such a difference. Among sitcoms, dramas, and reality programs on the six broadcast networks in the 2004-2005 season, Lauzen and Dozier (2008) found male characters to be more likely to use physical responses to conflict situations compared with female characters, and female characters more likely to use verbal responses. Male characters perpetrated slightly more of the antisocial acts found in a sample of Survivor episodes (Wilson et al., 2012), but female characters were no more or less likely to enact physical or verbal aggression than male characters in a sample of British reality shows (Coyne et al., 2010). In the Smith and colleagues (2002) study, reality TV had a higher proportion of male perpetrators of violence than all other program types examined.
Gender, Aggression, and Conflict on Docusoaps
Yet “reality television” is a term that encompasses many quite different types of programs and has undergone substantial change over time (Edwards, 2013). Murray and Ouellette (2009) specified eight subgenres of reality TV: gamedocs, dating programs, makeover programs, docusoaps, talent contests, court programs, reality sitcoms, and celebrity variations of other programs. In the study at hand, the focus is on docusoaps, defined as programs that “portray the everyday experiences of work, family, and relationships in the lives of their ‘real’ subjects” (Coles, 2000, p. 27) and that “[combine] observational documentary techniques with serial narrative techniques of soap operas” (Hill, 2005, p. 23). The docusoap genre of reality TV was chosen for the present study for three reasons: to achieve the cohesiveness necessary to test genre-specific cultivation theory (Bilandzic & Busselle, 2012); because docusoaps are quite popular, especially among viewers aged 18 to 49 years (Coles, 2000; Edwards, 2013; Gardner, 2013); and given that docusoaps are known in large part for melodrama and the headline-grabbing “bad behavior” of their stars within a context that is supposed to depict reality (Bruzzi, 2015).
Many docusoap series have, indeed, been ratings hits. Jersey Shore, for instance, delivered the then-largest audience (8.8 million viewers) for an MTV series at its peak (Levine, 2012); The Hills drew over 4 million viewers in its first episodes (Cox & Zeitchik, 2007); and Keeping Up With the Kardashians was the highest rated program among ad-supported cable channels among women 18 to 34 in 2007 (Mediaweek, 2007). Although no formal content analysis of aggression in docusoaps exists, series of this style have garnered a reputation for reliance on ostensibly authentic examples of aggression in verbal and physical forms, often played out in exaggerated or melodramatic ways that reflect the soap opera element from which the hybrid genre derives (Rifon et al., 2014). Jersey Shore, for example, has been discussed as “likely to be the consensus choice for the most appalling show of 2009” in the New York Times (Genzlinger, 2010), and the “catfights” displayed among the female stars of the Real Housewives series have captured news headlines, as well (Farber, 2009; Heffernan, 2009).
The interventions of producers in stirring drama in these programs have been the subject of critique, from The Real World “casting young adults in a manner intended to ignite conflict and dramatic narrative development” (Murray & Ouellette, 2004, p. 3), to the fights of the women on Real Housewives of New York “ . . . produced and edited for the delight of audiences” (Lee & Moscowitz, 2013, p. 72). Producer heavy handedness in shaping the “reality” of reality TV was discussed over a decade ago in the Robins Report in Broadcasting & Cable. Robins (2005) noted, “ . . . it’s not unusual to start shooting an unscripted show with a 100-page ‘outline’” and “scenarios are designed for humor and conflict and lines are often fed to contestants and stars” (p. 3). More recently, Woods (2014) argued for the use of the term structured reality to account for the ways in which the production process constructs casts and frames narratives: “producers structure cast meetings and plan events that prompt conflict from cast members’ existing interpersonal problems, heightening the everyday into melodrama” (p. 201).
Liebler, Hatef, and Munno (2016) drew mixed methods data that indicate that viewers have particular expectations regarding the enactment of aggression by male and female characters within docusoap reality television (DRTV) programs. After seeing multiple clips from Teen Mom, Real World, and Jersey Shore, male and female college student participants rated the aggression perpetrated by male characters as significantly more acceptable than that perpetrated by female characters. For female aggressors, verbal aggression was seen as more acceptable than physical aggression, whereas for male aggressors, that distinction was less strong. The difference between male and female research participants’ views of the acceptability of verbal aggression approached statistical significance (p = .06) in favor of greater acceptance among males. The qualitative results found that male college students tended to justify the aggression demonstrated in the docusoap clips more often than female college students.
Cultivation Theory in the Modern Television Environment
Cultivation theory is based on Gerbner’s (1969) prediction that television exerts a small but consistent and cumulative influence on the ways in which viewers conceive of the world around them. A central premise of the theory is that the messages sent through television are so prevalent that the total amount of viewing, rather than exposure to any particular genre or type of programming, can predict views congruent with the distortions found throughout television content (Morgan et al., 2015). Violence and aggression have been among the most persistent themes in television’s “message system” (Gerbner, Gross, Morgan, & Signorielli, 1980). The majority of violence-related studies has focused on the “mean world syndrome,” the fostering of the view that the world is a dangerous and scary place (Morgan et al., 2015).
Television has also been found in prior research to cultivate norms and beliefs regarding a variety of gender roles, serving an acculturation and socialization role. Meta-analyses by Herrett-Skjellum and Allen (1996), Morgan and Shanahan (1997), and Oppliger (2007) all find a small to moderate but consistently positive association, suggesting television contributes to more stereotypical gender-related outlooks. The question of whether gender-related cultivation (as well as other forms of cultivation) is experienced differently by male and female audience members has not been the subject of much scholarly inquiry. In a meta-analysis, Morgan and Shanahan (1997) found effects sizes were similar for male and female research participants, although for women, cultivation appeared to be more variable based on factors such as genre and context.
The current study investigates the intersection of these two lines of research, asking whether exposure to television’s messages about violence might shape the degree to which individuals view that behavior as an acceptable and normative aspect of social reality, and whether those perceptions differ by gender. As the primary storyteller of the modern environment, television has the ability to help establish cultural norms and expectations (Morgan et al., 2015), including, it is argued here, beliefs about the acceptability of aggression as a social behavior in general as well as beliefs about whether aggression is seen as differentially acceptable for men and for women. The novel question of whether the consistent messages viewers receive about violence and aggression generally through the medium of television might cultivate greater acceptability of those actions has been addressed in prior cultivation research, but not for at least two decades and solely among adolescents. Gerbner, Gross, Jackson-Beeck, Jeffries-Fox, and Signorielli (1978) asked, “How often is it all right to hit someone if you are mad at them? Is it almost always alright, or almost never alright?” and found heavy viewers more likely to agree that it is almost always acceptable to hit someone, especially among girls in sixth through ninth grade. The pattern in which girls, in particular, exhibited a cultivated outlook toward more acceptance of aggression was found once again in Morgan and Shanahan’s (1995) study of Argentinian adolescents, in which heavy viewing female teens were almost twice as likely to agree that hitting is acceptable compared with lighter viewing female teens. The current study updates these relatively dated cultivation theory inquiries by posing the same type of questions about normativity or acceptability of aggression within a sample of adult respondents.
Strong critiques have challenged cultivation research across multiple decades and many recent applications of the theory have veered away from some of the original premises (Morgan et al., 2015; Potter, 2014). Perhaps the largest departure from the original premises put forth by Gerbner is an exploration of the role of genre viewing rather than (or in addition to) overall amount of viewing in predicting cultivation outcomes. Potter (2014) characterized this change as a form of “boundary crossing” that moves the focus of cultivation analysis from the macrolevel (the television message system discussed by Gerbner and colleagues) to a microlevel analysis. Indeed, the assumption of cultivation theory that television presents a unified set of messages regardless of genre or program type was met with both early and continuing critique (Hawkins & Pingree, 1990; Potter, 2014). By the early 1990s, Potter (1993) observed that sufficient data had accrued to suggest “the cultivation effect may be even stronger with nonglobal measures” of television consumption (p. 575). In more recent research, the viewing of specific genres or program types has continued to predict a number of cultivation outcomes (e.g., Bilandzic & Rössler, 2004; Cohen & Weimann, 2000; Grabe & Drew, 2007; Segrin & Nabi, 2002).
Bilandzic and Busselle (2012) explained that a genre-based explanation for cultivation rests in large part on whether audiences expect a particular experience from the genre and, relatedly, whether the content patterns within the genre reflect a relatively cohesive set of themes. According to Cohen and Weimann (2000), another reason why genre matters in cultivation phenomena is that, in addition to other distinctions, “genres also differ in their focus on social reality” (p. 102). Social realism, defined as “the extent to which television content, whether real or fictional, is similar to life in the real world” (Busselle, Ryabovolova, & Wilson, 2004, p. 369), is of particular relevance to the sorts of judgments that occur in cultivation effects, particularly those made “off line” (Van den Bulck, 2003). Conceiving of television content as closely approximating real-world situations should facilitate a cultivation effect by reducing discrepancies between messages received from viewing and beliefs about the social world.
Only a handful of prior studies have investigated perceived realism’s role in the effects of reality television, despite the obvious claims to reality within this subset of programs. The general pattern within the existing studies is that perceived realism of reality television operates as either a main effect predictor of gender- or aggression-related beliefs (Behm-Morawitz, Lewallen, & Miller, 2016; Ferris, Smith, Greenberg, & Smith, 2007), or it strengthens the relationship between reality TV viewing and holding such beliefs (Martins & Jensen, 2014; Ward & Carlson, 2013), although some exceptions to these patterns occur (e.g., Segrin & Nabi, 2002).
Acknowledging that “whatever contributions specific genres may make to conceptions and beliefs take place in a broader cultural environment” (Morgan et al., 2015, p. 691), in the study at hand, docusoap viewing is examined alongside overall amount of viewing to test their relative contributions to views of the normativity and acceptability of aggression. Docusoaps have considerable cohesiveness in content (Murray & Ouellette, 2009) and therefore meet the condition put forth by Bilandzic and Busselle (2012) for genre-specific cultivation theory. By definition, they claim to represent social reality, which addresses the argument of Cohen and Weimann (2000). Yet prior research suggests that audience members vary in the degree to which they view reality television as realistic (Hall, 2006). Docusoaps, in particular, with their blending of the production features of documentary and the melodrama of soap operas (Coles, 2000; Murray & Ouellette, 2009) as well as the considerable notoriety of some of the series within the genre (Farber, 2009; Genzlinger, 2010; Heffernan, 2009; Rifon et al., 2014), are likely to produce variation in audience perceptions of their social reality. This complexity as well as the prominence of docusoaps in the television landscape (Coles, 2000; Edwards, 2013; Gardner, 2013) make them a unique feature of the contemporary cultural environment through which to explore cultivation theory.
Hypotheses and Research Questions
Consistent with (a) the larger body of cultivation research in which television exerts a socialization or enculturation influence (Morgan et al., 2015), (b) the cultivation of acceptance of aggression among heavy viewing female adolescents (Gerbner et al., 1978; Morgan & Shanahan, 1995), and (c) the large and enduring presence of violence as a meta-theme in television programming (Gerbner, Morgan, & Signorielli, 1994; Glascock, 2001, 2003; Greenberg et al., 1980; Hetsroni, 2007; Signorielli, 2003; Smith et al., 1998), we begin our analyses with a test of the original premise of cultivation theory by exploring the ability of overall television viewing to positively correlate with approval of both verbal and physical forms of aggression.
Next, we look specifically at whether the viewing of docusoaps, in particular, is linked with the approval of aggression in physical and verbal forms. Prior studies generally show a higher incidence of nonphysical forms of aggression being modeled by characters in reality TV (Coyne et al., 2010; Wilson et al., 2012). The docusoap genre features aggression in an ostensibly realistic fashion (Rifon et al., 2014) and has captured headlines for its emphasis on discord among its male and female stars (Farber, 2009; Genzlinger, 2010; Heffernan, 2009) often fueled by the interventions of producers (Lee & Moscowitz, 2013; Murray & Ouellette, 2004; Robins, 2005; Woods, 2014). Based on these content patterns as well as the evidence that specific genres of television can cultivate viewer outlooks (Bilandzic & Rössler, 2004; Potter & Chang, 1990), especially when those genres have a claim on social reality (Cohen & Weimann, 2000) and have consistent content features within (Bilandzic & Busselle, 2012), we predict the following hypothesis:
Because physical forms of aggression are likely to be present (Smith et al., 2002) but relatively infrequent in the reality genre compared with verbal forms (Coyne et al., 2010), we ask the following research question:
Given the potential for the perceived realism of reality TV to strengthen the association between viewing those programs and holding gender- and aggression-related views consistent with their depictions (Behm-Morawitz et al., 2016; Ferris et al., 2007; Martins & Jensen, 2014; Ward & Carlson, 2013), we predict the following hypothesis:
Once again, given the relative infrequency of physical aggression in reality TV programming compared with verbal forms (Coyne et al., 2010; Wilson et al., 2012), we ask the following research question:
Based on the mixed evidence that reality TV portrays gender differences in aggression (Coyne et al., 2010; Lauzen & Dozier, 2008; Wilson et al., 2012) and the finding that viewer perceptions of the acceptability of aggression within docusoap reality TV can differ by gender of the perpetrator (Liebler et al., 2016), associations between DRTV viewing, perceived realism, and approval of aggression might differ when the gender of the perpetrator of aggression varies.
Finally, based on the potential for differential cultivation patterns to occur by gender of the respondent (Gerbner et al., 1978; Morgan & Shanahan, 1995, 1997) as well as the real-world gendered patterns of aggressive behavior and attitudes (Bailey & Ostrov, 2008; Card et al., 2008; Dodge et al., 2006), we ask,
Method
Sample
A survey was distributed to adults across the United States using Amazon’s Mechanical Turk (Mturk), an online crowdsourcing system in which people can elect to complete tasks for a small financial reward. MTurk allows for more variation in age of respondents compared with college student samples, an important quality given recent data that show median ages of docusoap viewers ranging from 27 to 44 years (Dehnart, 2015). MTurk allows for respondents from various geographic areas and tends to produce samples with more demographic diversity than college student samples (Buhrmester, Kwang, & Gosling, 2011; Paolacci & Chandler, 2014).
A call for participants was posted to MTurk, asking for volunteers to participate in a research study on “your TV viewing habits, your opinions about some shows, and some general questions,” a prompt designed to avoid self-selection based on docusoap interest or fandom. Using filters provided by MTurk, respondents for the current study were restricted to adults in the United States. A total of 258 people took the survey. Ten participants failed an attention check near the end of the survey, leaving a final N of 248. In addition to the items used to measure the concepts central to the current study, the questionnaire contained a number of distractor items intended to mask the purpose of the study.
The mean age of the sample was 31.42 years (SD = 10.76); 56.8% of the sample identified as male and 43.2% as female. The most common race/ethnicity was White/Caucasian (81.9%), followed by Asian/Asian American (8.9%), Black/African American (4.4%), Hispanic/Latino (3.2%), and Biracial/Multiracial (1.6%). Political orientation was measured on a 7-point scale ranging from very liberal (1) to very conservative (7). The sample skewed slightly liberal (M = 3.08, SD = 1.5). The median household income range in the sample was US$30,000 to US$39,000.
Measures
Overall TV viewing
To measure overall TV viewing, participants reported how many hours of TV, including from sources like DVDs, DVRs, and streaming video, they watched on an average day (M = 2.48, SD = 2.03). This global measure of television use has been found to strongly correlate with estimates of exposure produced by diary data (Fikkers, Piotrowski, & Valkenburg, 2017) and fits assumptions of cultivation theory in which absolute exposure is less important than relative exposure compared with others (Shanahan & Morgan, 1999).
DRTV viewing
To measure DRTV viewing, participants were asked to indicate on a 5-point scale (1 = never, 5 = all of the time) how often they watched 13 current shows that meet the definition of the docusoap, an approach modeled after prior studies on the current topic (e.g., Behm-Morawitz et al., 2016; Ferris et al., 2007). 1 This list was compiled by visiting TV.com in the month prior to distributing the survey and selecting all programs that met the definition of DRTV programs from their list of the 80 most popular reality shows. It included Keeping Up With the Kardashians and two of its spin-off series, three Real Housewives series (Atlanta, Beverly Hills, and Orange County), three MTV series (The Real World, The Hills, and Jersey Shore), and four miscellaneous series (Bad Girls Club, Duck Dynasty, LA Ink, and The Real L Word). Participants’ scores for relative exposure to each of these 13 series were summed to create a measure of frequency of viewing DRTV programs (M = 16.32, SD = 5.14, α = .84).
Perceived realism
The perceived realism of DRTV shows was adapted from the Busselle (2001) measure, using a set of items designed to emphasize perceived similarity between a mediated depiction of reality and the real world, modified to reflect DRTV. Participants were asked to indicate level of agreement (1 = strongly disagree, 5 = strongly agree) with 12 such statements after being told that DRTV shows are domestic reality TV shows, those shows that deal with people in adult romantic, friendship, and family relationships in the places they live and work. Some popular examples of this kind of reality television include The Real World, Keeping Up With the Kardashians, and The Real Housewives series.
The phrase “domestic reality TV shows” was used in these items rather than “docusoap” given that the term docusoap may carry a negative connotation (Bruzzi, 2015) and therefore may have suppressed reports of exposure. Sample items include “Domestic reality TV shows portray the same kind of friendships you see in real life” and “Characters in domestic reality TV are very similar to people in the real world.” Results were summed to create an index for Perceived Realism of DRTV (M = 25.95, SD = 8.14, α = .87).
Approval of physical and verbal aggression
The approval of physical and verbal aggression was measured using the Normative Beliefs About Aggression Scale (NOBAGS). The original scale, designed for children, was modified for adults by replacing all mentions of “boy” and “girl” with “man” and “woman,” as has been the case in prior research (e.g., Bailey & Ostrov, 2008). Participants were presented with a series of prompts describing aggressive behaviors and asked to evaluate the appropriateness of verbally aggressive (e.g., yelling) and physically aggressive (e.g., hitting) responses, using a 5-point scale ranging from it’s really wrong (1) to it’s perfectly okay (5). Examples include “Suppose a man says something bad to another man, John.” “Do you think it’s OK for John to scream at him?” followed by, in a separate question, “Do you think it’s OK for John to hit him?” The responses to the four verbal prompts (e.g., screaming as a response) were summed to create a measure for approval of verbal aggression (M = 10.11, SD = 3.6, α = .92). Likewise, the responses to eight physical prompts (e.g., “Suppose a woman hits another woman, Mary. Do you think it’s OK for Mary to hit her back?” which indicates hitting as a response) were summed to create a measure for approval of physical aggression (M = 18.42, SD = 5.63, α = .81).
Gendered aggression
The NOBAGS contains a gendered element, wherein the gender of both the aggressor and the target are varied to allow for four possible gender configurations for each aggression scenario. This breakdown was used to create four additional subscales based on the gender of the perpetrator of aggression: approval of male-perpetrated verbal aggression (two items, M = 4.89, SD = 1.84, α = .82), approval of female-perpetrated verbal aggression (two items, M = 5.21, SD = 1.94, α = .91), approval of male-perpetrated physical aggression (four items, M = 8.64, SD = 2.77, α = .56), and approval of female-perpetrated physical aggression (four items, M = 9.80, SD = 3.16, α = .67). It is apparent from the means reported here that there was more approval of female-perpetrated than male-perpetrated aggression of the same form. (The gender of the target of the aggression was considered outside the current scope.)
Results
A series of independent t tests were run to examine whether male and female respondents differed in the main variables of the study before exploring hypotheses and research questions (see Table 1). There were no gender differences in total television viewing or in perceived social realism of DRTV, but women in the sample did report a higher frequency of viewing DRTV programs compared with men. There were no significant differences in DRTV viewing by race, nor was age of respondent correlated with DRTV viewing.
Independent t Tests Exploring Gender Differences in Approval of Aggression, Total Television Exposure, Docusoap Viewing Frequency, and Perceived Realism of Docusoaps.
Approval of physical aggression (regardless of the gender of the perpetrator) was significantly higher among men in the sample than among women (see Table 1). For both physical aggression perpetrated by males and physical aggression perpetrated by females, male respondents were more accepting than female respondents. Compared with views on physical aggression, approval of verbal aggression among male and female respondents was much more closely aligned (see Table 1). There was no significant gender difference for approval of verbal aggression in general, and the difference in views of the acceptability of verbal aggression performed by males, in particular, approached but did not reach statistical significance, with the trend toward greater approval among male respondents. There was no gender difference in perception of the acceptability of verbal aggression performed by females.
Hierarchical regression was used to test the relationships between television viewing, DRTV viewing, and approval of aggression, and to examine the potential moderating effect of perceived realism of DRTV. Two regression analyses were initially conducted, one for each form of aggression (physical and verbal), followed by regression analyses for each of the gender-of-perpetrator subscales. Prior to conducting the analyses, all variables were checked for normal distribution. The DRTV viewing index was found to be slightly skewed and therefore a logarithmically transformed version was used in subsequent analyses (M = 2.74, SE = .24). DRTV viewing and the measure for perceived realism of DRTV were both mean-centered to reduce issues of multicolinearity in their use as interaction terms. Finally, given the relatively low number of respondents of color, race/ethnicity was recoded to White/respondent of color.
For each analysis, age, gender (1 = male, 0 = female), race (White = 1, Respondents of Color = 0), household income, and political orientation (liberal/conservative) were entered in the first block as control variables. Overall television viewing was entered in the second block. Frequency of DRTV viewing and perceived realism of DRTV was entered in the third block. To test the proposed moderating variable, a multiplicative interaction term between perceived realism of DRTV and DRTV viewing was entered in the fourth block. The full reported results of all regressions can be found in Table 2. Among the demographics included in the first block, only the variables that emerged as significant predictors are displayed.
Regression Coefficients in Predicting Approval of Verbal and Physical Forms of Aggression (N = 248).
Note. DRTV = docusoap reality television.
p < .05. **p < .01. ***p < .001.
The first block indicated that none of the demographic variables predicted approval of verbal aggression. Overall TV viewing was entered in the second block, and was not significant. In the third block, ΔR2 = .06, F(2, 214) = 6.71, p < .001, DRTV viewing was a positive predictor of verbal aggression approval, β = .16, t(214) = 2.19, p < .05—supporting
Repeating the regression with the gender-of-perpetrator subscales as the dependent variables, as called for in
For approval of physical aggression, the first block, ΔR2 = .10, F(5, 217) = 4.68, p < .001, indicated that age was a negative predictor, β = −.20, t(217) = −2.91, p < .01, and men scored significantly higher than women, β = −.21, t(217) = 3.13, p < .01. In the second block, overall television viewing had no significant effect. The third block, ΔR2 = .03, F(2, 214) = 3.03, p = .05, indicated that while DRTV viewing itself was not a significant predictor (in association with
Again, the repetition of the regression using the gender-of-perpetrator subscales revealed differences in the role of DRTV viewing, lending further data to explore
To investigate potential gender differences among respondents as called for in

Graphic representation of perceived realism of docusoap’s moderating role in the relationship between docusoap viewing and approval of female-perpetrated verbal aggression among women respondents only (n = 108).
The regression analysis for the male respondents in the sample revealed an inverse pattern. The analyses for approval of verbal aggression—both the full measure and the gender-of-perpetrator subscales—showed no significant associations between approval of verbal aggression and DRTV viewing or perceived realism of DRTV. However, several differences emerge when examining the approval of physical aggression. In the demographic block, ΔR2 = .05, F(4, 121) = 1.54, ns, age fell out as a significant predictor, while increased political conservatism was shown to correlate with approval of physical aggression, β = .21, t(121) = 2.33, p < .05. The association between perceived realism of DRTV and approval for physical aggression was no longer statistically significant within the smaller sample of male-only participants, β = .17, t(118) = 1.82, p = .07, and the negative association observed between DRTV viewing and approval of male-perpetrated physical aggression was also no longer significant. However, in the final block, ΔR2 = .07, F(1, 117) = 6.38, p < .05, the moderating effect of perceived realism of DRTV on the association between DRTV viewing and approval of physical aggression was significant for male respondents, β = −.24, t(117) = −2.52, p < .05. This was also the case when examining the gender-of-perpetrator subscales of approval of male-perpetrated, β = −.22, t(121) = −2.39, p < .05, and female-perpetrated, β = −.21, t(118) = −2.3, p < .05, physical aggression. Graphic representations of these interaction effects (see Figure 2) indicated a mainstreaming effect of DRTV viewing. Men who viewed DRTV as more realistic tended to approve of physical aggression more than men who viewed it as unrealistic. However, this distinction diminished with increased DRTV viewing.

Graphic representation of perceived realism of docusoap’s moderating role in the relationship between docusoap viewing and approval of physical aggression among men respondents only, n = 140.
Discussion
There are key gender differences in perceptions of the normativity of physically and verbally aggressive responses to provocation or conflict, according to both the present study and prior research. First, the means were higher for the approval of female-perpetrated compared with male-perpetrated physical and verbal aggression items in the current study, indicating more acceptance of females engaging in these forms than of males doing so. Yet the current results also find that male respondents tended to accept aggression as a normative aspect of behavior more so than females in the sample did, as has been the case in prior research (Bailey & Ostrov, 2008; Card et al., 2008; Dodge et al., 2006; Harris, 1994; Krahé & Busching, 2014). There is indication, then, that men tend to approve of aggression enacted by both members of their own gender and by women more so than women do, an important finding for its potential implications for the enactment of aggression (Huesmann & Guerra, 1997). The question of whether television exposure—either overall viewing of the medium or exposure to the docusoap subset of reality TV programs—might contribute to these differential outlooks cannot be fully answered by the present study, given its reliance on cross-sectional survey design. Yet, the study did reveal a number of significant associations among exposure to docusoaps, perceptions of the social realism of docusoaps, and approval of physical and verbal aggression, pointing speculatively to cultivation theory’s explanation of television’s storytelling function enculturating viewers by helping define what is be considered “normal” and what is not (Morgan et al., 2015).
In the current study’s findings, overall television viewing consistently failed to predict any of the approval of aggression measures, including approval of verbal as well as physical forms within the entire sample as well as within male and female respondents examined separately. This result is surprising given the presence of aggression as a meta-theme in the “message system” of television (Gerbner et al., 1994; Glascock, 2001, 2003; Greenberg et al., 1980; Hetsroni, 2007; Signorielli, 2003; Smith et al., 1998) and the results among female adolescent participants in earlier cultivation studies using a very similar approval of aggression measure (Gerbner et al., 1978; Morgan & Shanahan, 1995). Yet this pattern is consistent with the evolution of cultivation theory toward genre-specific rather than global television exposure predicting key outcomes (Bilandzic & Rössler, 2004; Potter, 2014; Potter & Chang, 1990).
Indeed, in the study at hand, docusoap viewing was a significant, positive predictor of approval of verbal aggression (but not physical aggression) among the entire sample, with standardized beta coefficients indicating predictive power on par with that found for age or gender of respondent. Thus, the observed findings were consistent with the form of aggression more commonly present in these programs—verbal rather than physical—according to content analyses of reality television (Baruh, 2009; Coyne et al., 2010; Smith et al., 2002; Wilson et al., 2012). The magnitude of the docusoap viewing beta coefficient as a predictor of approval of verbal aggression is quite striking, and it suggests that docusoaps do, indeed, stand out in the television landscape as a unique source of information about verbal aggression in the social world among viewers of the genre. The inclusion of highly dramatized arguing, yelling, name calling, and other forms of verbal aggression in the docusoap (Farber, 2009; Genzlinger, 2010; Heffernan, 2009; Rifon et al., 2014) may socialize viewers of the genre to accept such aggression as normal. This calls into question the ethics of the casting of individuals who will dislike and disagree with each other, the scripting of an extensive “outline” for the narrative, and/or the suggestion of dialogue lines or character responses from those involved in the production of the docusoap (Lee & Moscowitz, 2013; Murray & Ouellette, 2004; Robins, 2005; Woods, 2014). That process appears to result in a substantial amount of verbal aggression in the programs which, in turn, can socialize viewers of the genre into perceiving verbal forms of aggression to be normative and acceptable. To confirm and further explore this key pattern that emerged from the current, cross-sectional data, a content analysis of the docusoap genre should be conducted and an experiment or longitudinal study should follow to examine causal directions. From the study at hand, it appears that docusoaps—with their blending of the true-to-life filming of day-to-day interactions of “ordinary” people and the exciting appeal of soap opera melodrama (Bruzzi, 2015)— are one feature of the cultural environment that can contribute to the degree to which verbal aggression is seen as socially acceptable and culturally normative. The macrolevel influence of television as originally argued in cultivation theory is therefore played out more specifically as a more microlevel phenomenon (Potter, 2014), but the basic premise of television functioning as a socialization agent remains the same (Morgan et al., 2015).
As has been found in past research on related topics (Behm-Morawitz et al., 2016; Ferris et al., 2007; Martins & Jensen, 2014; Ward & Carlson, 2013), perceived social realism appears to play an important role. In the present results, the degree to which respondents viewed docusoap programs as realistic was a main effects predictor of approval of aggression (both physical and verbal) independent of the frequency with which they viewed the genre. The pattern in which perceptions of the social realism of docusoaps appear to operate independently from viewing the content in relation to approval of aggression in the sample as a whole may indicate that assumptions about docusoaps rather than actual viewing experiences are used as a heuristic. If the programs are known in large part for melodrama and the “bad behavior” of their stars as covered in the news media (Farber, 2009; Genzlinger, 2010; Heffernan, 2009; Rifon et al., 2014), perhaps our respondents were relying on that reputation when responding to the social realism of docusoap items on the survey and that, in turn, predicted approval of aggression. Logically, considering what critics and journalists have dubbed melodrama and bad behavior to be realistic, typical features of daily life would be likely to correspond to views of physical and verbal aggression as normative and acceptable. Where some see such examples of conflict, argumentation, and aggression as remarkable, others may see them as routine or even mundane.
The data also reveal interesting patterns in the views respondents hold about what forms of aggression are acceptable for women to enact as compared with for men, extending prior work on television use and the socialization of views of gender roles and norms (Herrett-Skjellum & Allen, 1996; Morgan & Shanahan, 1997; Oppliger, 2007). Among the entire sample, docusoap viewing was a significant positive predictor of approval of female-perpetrated verbal aggression and a significant negative predictor of approval of male-perpetrated physical aggression. The viewing of the genre, therefore, appears to correspond with support for women using yelling, arguing, and name calling when aggressed against as an acceptable practice (but not men doing so) as well as lack of support for men using hitting and other physical forms when aggressed against (but not women doing so). This pattern is reminiscent of the findings of Liebler and colleagues (2016) who measured views of aggression after screening sample clips from popular docusoaps, and may be explained by content patterns in docusoaps, although the lack of prior content analyses of the genre make this explanation speculative at this point. Anecdotally, there do appear to be prominent female stars who engage frequently in verbal aggression in the list of programs used to measure frequency of viewing docusoaps in the present study, with the cast members of the Real Housewives series (Lee & Moscowitz, 2013) as well as Snooki and other female characters in Jersey Shore (Genzlinger, 2010) prime examples. These characterizations could have played a role in the ability of docusoap viewing to predict approval of female-perpetrated verbal aggression. It is also possible that the programs show physical aggression enacted by male stars to be accompanied by negative consequences, which could explain the association between viewing docusoaps and lack of approval of male-perpetrated physical aggression. These theories should be tested with follow-up studies, and cultivation—which does not always account for identification with particular characters or observation of positive or negative outcomes associated with characters’ actions—should be updated accordingly.
In addition to being a main effects predictor among the sample as a whole, perceived realism of DRTV also moderated the relationship between DRTV viewing and approval of aggression, although this pattern only surfaced when the sample was divided by gender of respondent. For the men in the sample, the interaction between viewing the genre and perceiving the genre as realistic predicted approval of physical aggression but not verbal aggression; and for women, the interaction predicted approval of verbal aggression and not physical aggression. For the women, the data array in a manner that might indicate resonance, a corollary of cultivation theory in which one’s experiences and orientations when aligned with television’s depictions make for a “double dose” shaping of perceptions of social reality (Gerbner et al., 1980; Morgan et al., 2015). Heavy docusoap viewing women who believed the content encountered on docusoaps was similar to what might occur in the real world had the highest approval levels for verbal aggression, so that perceived realism and viewing resonated with one another to predict approval. However, the pattern among the men might be interpreted as mainstreaming, a phenomenon in which individuals with differing background characteristics converge in their outlook as a function of heavy viewing (Gerbner et al., 1980; Morgan et al., 2015; Van den Bulck, 2003). Heavy docusoap viewing men (but not light docusoap viewing men) had similar approval levels of physical aggression regardless of the degree to which they believed the content encountered on docusoaps was similar to what might occur in the real world. It is clear from the present findings that we see different cultivation processes by gender of respondent, and follow-up research should seek to explain how and why these processes occur.
Views of the normativity or acceptability of aggression among the males and the females in the sample become even more complex in the current data when the gender of the perpetrator is taken into consideration. Among the women in the sample, perceived realism of docusoaps interacted with docusoap viewing to positively predict approval of female-perpetrated verbal aggression. Among the men, perceived realism of docusoaps interacted with docusoap viewing to negatively predict approval of both male- and female-perpetrated physical aggression. The significant interactions suggest that using the reputation of the genre as a heuristic when responding to the perceived realism items (as we have speculated was the case) is likely to work in concert with direct knowledge of program content stemming from viewing docusoaps in predicting approval of aggression. Yet the messages viewers receive from this convergence are different by gender of viewer, despite the cohesiveness of the genre (Murray & Ouellette, 2009).
Given that men in the sample had more tolerant views of physical aggression than women did, it is curious that the interaction between viewing of docusoaps and their perceived realism was associated with less approval of physical aggression among the men in the sample. Perhaps the observations gleaned from viewing docusoaps in combination with the reputation of the genre called to mind when considering its perceived realism present a sort of cautionary tale against physical aggression for the male respondents. Prior research has established that some reality television viewers obtain gratifications from feeling superior to the people presented in reality television programs (Hall, 2006) and perhaps male respondents were exhibiting a tendency to distance their own views and values from those exemplified in docusoaps. Conversely, the women in the sample may have taken their cues from the use of verbal aggression observed in viewing docusoap content and in its reputation as a means of empowering women, which could explain the interaction between docusoap viewing and perceived realism in predicting female respondents’ greater approval of verbal aggression as enacted by women, in particular. In other words, perhaps the female stars of these programs use verbal aggression to stand up for themselves or otherwise assert themselves in a manner that women in the current sample found to be positive. Once again, future content analysis research should examine the aggressive behaviors of male and female characters in docusoaps and a future effects study should explore whether differences in identification with docusoap program stars who may aggress differently on the programs might explain these patterns.
There are a number of limitations to the present study that should be considered when interpreting its results. The reward structure of Mechanical Turk can lead to social desirability biases (Goodman, Cryder, & Cheema, 2013) as well as to research fatigue among frequent participants (Stewart et al., 2015). Whereas Mechanical Turk samples tend to be more diverse than student convenience samples (Buhrmester et al., 2011), they can also skew demographically in ways consistent with differences between internet users and nonusers (Paolacci & Chandler, 2014). It is also possible that viewers and nonviewers of docusoaps differ in ways not accounted for in the present study—including in whether they experience a voyeuristic desire to get a glimpse into other people’s lives (Nabi, Stitt, Halford, & Finnerty, 2006). Aspects of viewer personalities—such as extraversion, previously found to predict reality TV viewing (Aubrey et al., 2012)—may also help explain both motivations for viewing and messages received from exposure to docusoap content.
Social statistics show that physical violence is perpetrated most often by men but increasingly by women, and attitudes about aggression are important predictors of the enactment of violence (Lauritsen et al., 2009; WHO, 2010). Therefore, it is important to understand the ways in which individuals come to learn about what forms of aggression are considered acceptable or normative in society. The present results suggest that the docusoap genre of reality television is one such source of socialization, and future research should further examine the processes by which these associations occur.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Notes
Author Biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
