(S)Ex on the beach.
An experiment on sexualized reality
television, identification, and permissive sexual attitudes of late
adolescents.
Kim van de Steeg 12545937
Master’s Thesis, Graduate school of Communication Entertainment Communication
Supervised by dr. R.J. Kühne 30-1-2020
Abstract
Highly sexualized reality television has raised concerns, because of its possible negative effect. This experiment looked into the effects of the popular reality television show Ex on the Beach in the Netherlands. The effects of sexualized reality television were examined on the permissive sexual attitudes. Identification was treated as a moderator. The interaction effect of sexual content in reality television and identification was also examined. A sample of 128 Dutch adolescents in the age of 16 to 19 years old was randomly assigned to video and identification conditions. With either a video containing sexual depictions or a neutral video, and either an instruction that told the participant to identify with the characters in the video or to look critically. The results indicated that the manipulation of identification did not work, and that sexual content did not have an effect on permissive sexual attitudes. In contrary to what was expected, identification did not work as moderator in this study, but as a mediator. Participants reported higher permissive sexual attitudes when they identified with the
characters in the video content. And, participants identified more with the characters in the neutral video than in the sexualized video. The implications of sexual content on permissive sexual attitudes, and identification levels are discussed.
Keywords: Late adolescents, reality TV, permissive sexual attitudes, identification,
Introduction
MTV’s series Ex on the Beach is a big hit in the Netherlands. It is a reality television show for 16 years and older and contains a lot of sexual depictions and explicit language
(Kijkwijzer, n.d.). The show is about young people who stay in a house in a nice and warm resort. Each episode, another ex-boy-/girlfriend of one of the participators walks out of the sea. The show contains a lot of drama and a lot of sexual depictions. Over the past years, reality television in the Netherlands became more and more graphic (Jansen & Tienhooven, 2019; Takken & Nieber, 2019). Twenty years ago, it did not even come to mind that it was possible to show sex on Dutch television. Right now, it sells better than ever (Jansen & Tienhooven, 2019).
In 2019, sexual content on Dutch television went even further. The show ‘De Villa’, a Dutch reality television show, stopped broadcasting after an incident had happen. In one of the episodes a participant of the show was raped, and everyone in the Netherlands could see it. Social media exploded with reactions on this episode, it was very bad publicity for the network RTL and eventually they decided to stop further broadcasting the show. Also, other ‘guilty pleasure’ reality television shows were stopped from broadcasting (Jansen &
Tienhooven, 2019; Takken & Nieber, 2019). This can raise questions on what should be normal to show on television.
A content analysis showed that two out of ten scenes in reality shows contain
sexualized messages (Vandenbosch, Vervloessem, & Eggermont, 2013). For Ex on the Beach that is even more, the whole show contains sexual depictions. It is important to study the effects of highly sexualized content in reality television on late adolescents. Especially on this vulnerable age group, because late adolescents are developing their sexual identity and look to media as an example (Valkenburg & Piotrowski, 2017). The large amount of sexual content in reality television has raised concerns. This is because there is a fear that the lessons learned
from television can be inaccurate, or more suitable for parents to teach their children (Collins, Elliott, Berry, Kanouse, & Hunter, 2003).
There has not been a lot of research done on reality television shows and the effect on late adolescents and their attitude towards sex. Especially sexual permissiveness. According to research, identification with the characters in the show can increase the effect (Moyer- Gusé, Chung, & Jain, 2011; Ward & Friedman, 2006).
Reality television can create bigger identification than fictional television, it is more realistic (Beck, Hellmueller, & Aeschbacher, 2012). Entertainment media have the power to influence attitudes and behavior in a more effective way than traditional messages. There is less resistance because the consumer is engaging in a narrative (Moyer-Gusé, 2008). Socially responsible messages in entertainment media can be a powerful way of affecting sexual behavior. The marketing of the certain behavior isn’t very obvious, compared to traditional advertising, therefore the viewer is not as likely to resist the message (Brown, Steele & Walsh-Childers, 2001).
Since the number of sexual depictions on Dutch reality television seem to grow it is important to look at its effect. Especially on the most vulnerable age group who is forming a sexual identity: late adolescents (16-19-year-old). The level of identification with the people in reality television shows seems to have an effect and is therefore also implemented in this study. The most important part of this study is whether sexual content in reality television has an effect on the permissive sexual attitudes of adolescents. This study is the first to research the effects of a reality television show on sexual permissive attitudes of late adolescents with an experiment. Also treating identification in this study as a moderator is new. There has been looked at wishful identification as a mediator, but not at identification as a moderator in studies before (Bond & Drogos, 2014). This study is the first to try to manipulate the level of identification in an experiment. The choice was made to focus on identification instead of
wishful identification, because identification is part of social cognitive theory (Bandura, 1986). And social cognitive theory can be used to explain the effects of watching reality television on permissive sexual attitudes (Moyer-Gusé, 2008). This is tested with an
experiment with an episode of the popular television show Ex on the Beach Double Dutch All Stars. This leads to the following research question: Does viewing sexualized reality television
have an effect on permissive sexual attitudes of adolescents?
Content and characteristics of sexualized reality television
Reality television programs are defined by scholars as unscripted television shows that portray ordinary people instead of actors (Riddle & De Simone, 2013; Nabi, Biely, Morgan, & Stitt, 2003; Reiss & Wiltz, 2004). Reality television shows seem to be highly popular among adolescents (Nielsen, 2009), especially MTV reality shows, which target emerging adults and adolescents (Franko & Krieger, 2011). Reality programs about dating, sexual attraction, seduction, and intimacy dominated the reality genre over the last decade
(Zurbriggen & Morgan, 2006). According to Ward (2002) sexual content on television has been increasing over time. Sexual content that is depicted in reality television shows consist of sexual talk and sexual behavior (Bond & Drogos, 2014).
According to a study by Bond and Drogos (2014) television that shows sexual content can influence people’s attitudes and perception on sexual behavior. Bond and Drogos (2014) used a survey to examine the link between emerging adults viewing an unscripted television program and permissive sexual attitudes. They found a positive correlation. In extension, the few studies that examined the relationship between exposure to sexual content in the media and attitudes and beliefs about sex suggest that teens learn about sex from the media they attend to (Brown, Steele & Walsh-Childers, 2001). Experimental studies showed that college students were more likely to endorse the stereotypical sexuality according to television viewing when being exposed to sexual scenes from prime-time television. This stereotypical
sexuality means dating is a game, men having trouble being faithful and being sex-driven and that women are sexual objects (Ward, 1995; Ward, 2002). According to Ward (2002)
adolescents’ permissive sexual attitudes are not correlated with television exposure in general. Specific types of programming showed stronger relationships with permissive sexual
attitudes. Greeson and Williams (1986) found with their experiment that adolescents were more accepting premarital sex after watching MTV shows, than adolescents who did not watch such shows. Significant relationships of permissive sexual attitudes have also been found between unscripted television programs (Riddle & De Simone, 2013) and exposure to one favorite television program (Collins et al., 2003; Wright, Randall, & Arroyo, 2013). Sexual messages on television and its influence on sexual development of youth made Brown, Halpern, and L’Engle (2005) even name television the ‘Sexual super Peer’. They did a study on female adolescents and found that mass media serve as a sexual super peer. Girls rely on media for sexual advice. This raises concerns since there is a lack of sexual health messages in these media (Brown, Halpern, & L’ Engle, 2005). According to Ward and Rivadeneyra (1999) exposure to sexual content is positively correlated with more sexual experiences and negative attitudes towards self-restraint, at least for the age group emerging adults. These were results of their project that also showed that results differed by sex, the associations were stronger for females than males (Ward & Rivadeneyra, 1999). According to the study of Bond and Drogos (2014), unscripted reality television that contains a lot of sexual content has a relationship with sexual permissiveness among emerging adults. All of these studies examined different age groups. Sexual content in media and television has an effect or relation according to the studies discussed above. But what is this effect on late adolescents?
Late adolescents and sexualized reality television
The age group that is examined in this study is late adolescents. This age group is the focus because Ex on the Beach (the show that is used for this study) is for 16 year and older according to Kijkwijzer (n.d.). Late adolescents (16-19 years old) are interested in reality television that places sexuality at the forefront (Valkenburg & Piotrowski, 2017). It can have a big effect on this age group, especially because they are forming an identity by watching media. Adolescents learn about sex, relationships and much more by consuming media (Valkenburg & Piotrowski, 2017). A report shows that exposure to television containing sexual content may have a negative effect on the development of the sexuality of adolescents (Huston, Wartelle, & Donnerstein, 1998). Huston and colleagues (1998) wrote that a few experimental studies found that the attitude and knowledge of viewers was changed by television exposure. In extension, correlational studies found some links between television viewing and sexual behavior of adolescents (Huston, Wartelle, & Donnerstein, 1998). The age group of 16-19-year-old is more vulnerable than others because they have fewer real-world comparative experiences, less reasoning abilities, and idolize media figures (Gruber & Grube, 2000). That is what makes them more susceptible to particular media content. Adolescents are particularly open to new information regarding romance and relationships (Ward & Friedman, 2006). This raises concerns since the sexual content in reality television is not always
accurate, and this age group is still forming a sexual identity and therefore more vulnerable to its effects (Collins et al., 2003; Valkenburg & Piotrowski, 2017). This leads to the following hypothesis in this experimental study:
H1 Adolescents have stronger permissive sexual attitudes when viewing reality shows
The effect of sexualized television and the mechanisms of identification
According to Bond and Drogos (2014) the influence of sexual content is bigger when people watch reality television more often, these people are also known as heavy viewers.
Cultivation theory can be used to explain the effect of repeated media exposure and sexual outcomes (Morgan & Shanahan, 2010; Ward, 2002; Morgan & Signorielli, 1990; Gerbner et al., 2002). Traditionally, the cultivation hypothesis explains beliefs and attitudes of heavy viewers that are created by overall exposure, instead of exposure to individual programs (Bond & Drogos, 2014). More recent studies have shown that the cultivation theory is not only applicable to general watching, but also to repeated exposure to certain programs (Morgan & Shanahan, 2010). The cognitive mechanism, the availability heuristic, is used to explain relationships between overall television and exposure and viewers beliefs. This mechanism can justify relationships that have been found between program exposure and beliefs (Morgan & Shanahan, 2010; Bond & Drogos, 2014). Heavy viewers rely on the messages that are easily accessible when processing information, that are the messages from the shows they frequently watch. When sex is shown a lot in a show, it could also serve as a cultivation for the viewer (Bond & Drogos, 2014). Riddle and De Simone (2013) even found a relation between viewing a certain behavior on television and beliefs of the viewers.
Viewers thought the behavior that was shown is normal. For Ex on the Beach, sex is depicted in almost every scene, therefore it can have a cultivation effect on the viewer. Cultivation theory can be used to explain the effect in general, but it was criticized for being too broad and not taking contextual factors of attitude formation into account ((Rubin, Perse, & Taylor, 1988; Shanahan, Shanahan & Morgan, 1999). Bandura’s (1986) social cognitive theory integrates social contextual factors into the effects model and is therefore relevant to use to try to explain the effects.
Social cognitive theory (Bandura, 1977, 1986; Bushman & Huesmann, 2001; Huesmann, 1998) claims that: human beings observe role models, make inferences and attributions, and accumulate scripts, schemas, and normative beliefs that then give directions to their subsequent behavior. According to this theoretical perspective, adolescents learn sexual behaviors and the consequences of this behavior by watching television. Especially from reality television because it is closer to reality, the effect is therefore bigger (Ward & Friedman, 2006). Social cognitive theory has four subprocesses that control observational learning: attention, retention, production, and motivation (Bandura, 1986). Motivation is key in these processes, because people do not always make the decision to engage in certain behavior they learn from media (Bandura, 1986). Motivation is linked to identification. It depends on how motivated the people are to make them act in the way they have observed. People vary in how motivated they are, this is influenced by outcome expectancies and self-efficacy (Moyer-Gusé, 2008). The outcome expectancies are the perceptions of the
consequences of a certain behavior (positive and/or negative) by the viewer. Attractive and similar characters on television are more likely to be imitated, the motivation here is higher. Also, if a certain behavior is rewarded and not punished, the viewer is positively motivated to imitate the behavior (Bandura, 2004). Reality television shows like Ex on the Beach contain real, ordinary people, therefore viewers may find themselves similar and more attracted to these people, instead of actors in scripted television (Murray & Ouellette, 2009). The viewers identify more with the characters. According to a study, a motive for watching unscripted television is that people are curious about other people’s lives and like to watch interpersonal interaction (Nabi, Biely, Morgan & Stitt, 2003). Another survey study found that people who watch unscripted television for relaxation and entertainment, thought that these television programs are highly realistic (Papacharissi & Mendelson, 2007). Identification is seen as a motive to watch unscripted television programs. It is closely linked to social cognitive theory,
because viewers identify with the characters, find themselves similar, find the content realistic, and are therefore motivated to watch and eventually adapt the behavior.
According to Cohen (2001), identification is a mechanism of interpretation and experience of reception where audience members feel like the event was happening to them. The viewer takes on the perspective of the character and forgets about his or her own world (Moyer-Gusé, 2008). When a viewer identifies with the character, he or she imagines being the character and the identity will be replaced by the one from the character in the show (Cohen, 2001). According to research, identification with the characters in the show can increase the effect (Moyer- Gusé, Chung, & Jain, 2011; Ward & Friedman, 2006). Reality television can create bigger identification than fictional television, it is more realistic and contains ordinary people (Beck, Hellmueller, & Aeschbacher, 2012).
The process of identification has four dimensions: empathic (feel with the character); cognitive (share the perspective of the character); motivational (internalize the goals of the character); and absorption (loose self-awareness during watching; Cohen, 2001).
Identification is very close to the social effects of media (Cohen, 2001; Basil, 1996; Macoby & Wilson, 1957). Appearance, attitudes, values, and activities were adjusted from viewers to become more like the media characters they identified with (Caughey, 1986; Boon & Lomore, 2001). When audience members identify with the characters on a television show they feel like they have something in common (Tian & Yoo, 2015). Viewers self-identity is redefined, reinforced or elevated through identification, by watching reality television shows (Tian & Yoo, 2015). Identification can increase the effect of television as stated above. In this study the interest lays in the effect of sexual content in reality television on permissive sexual attitudes, and if this is moderated with the level of identification of adolescents with
characters in the show. This could be an interaction effect of sexual content and identification level.
This leads to the hypothesis:
H2 The effect of the absence or presence of sexual content in reality television shows
on adolescents’ permissive sexual attitudes is moderated by the level of identification with the characters. When adolescents identify with characters in reality television the effect is
stronger (higher permissive sexual attitudes), than when adolescents don’t identify with the cast.
Method
Design and participants
To test the hypotheses, an online experiment was conducted. The study employed a 2x2 experimental design in which a reality television episode and an instruction about identification were manipulated. The study received ethical approval. For this study
adolescents were contacted via social media and social networks. A gift card of choice was offered as an incentive, the incentive could be won by the participants if they filled in their email at the end of the experiment. Lots of adolescents participated in the experiment (N = 165). After data cleaning 27 participants were removed. These 27 records were removed because the participantsstopped the experiment before seeing the stimulus material. The final number of participants was 128, without any missing values. The mean age of the participants was M = 17.46 (SD = 1.18). 35% (n = 45) of the participants were male and 65% (n = 83) were female.
Procedure
The data was collected via an online experiment, using Qualtrics. The participants were asked their consent and their age. If the age was below 16, the participant was directed to the end of the survey. First, the participants were exposed to a brief instruction that either told the
participants to identify with the characters in the video or watch the video critically. After the instruction the participants were exposed to a video of an episode of the series Ex on the beach Double Dutch All Stars with either sexual content or neutral content. The participants were randomly assigned to the conditions by Qualtrics. The experiment had four conditions: manipulated instruction with sexual content (n = 33), manipulated instruction with neutral content (n = 35), neutral instruction with sexual content (n = 30), and neutral instruction with neutral content (n = 30). After the participant had watched the video, he or she had to answer several questions. These questions contained the treatment checks and the measurement of permissive sexual attitudes, and level of identification with the characters. After these
measurements the participant was asked whether he or she had seen the episode before and if and how often he or she had watched Ex on the Beach.
Experimental manipulations
In the current study, a 2x2 experimental design, the two factors had each two conditions. The instruction was manipulated, and the video was manipulated. The instruction was manipulated to create low or high identification with the characters on the show and the video had either sexual content or neutral content.
Stimulus and manipulation of sexual content
The sexual content in the stimulus material was either present or absent. Two versions of a short-edited video of an episode of Ex on the Beach Double Dutch All Stars were made. This study examined the effects of the sexual depictions in reality television by using the popular show Ex on the Beach. The stimulus material is in line with the material that Ward (2002) used in her study. She used multiple conditions with sexual aspects and one neutral condition. In this particular study the interest is in overall sexual depictions, not multiple categories as
Ward (2002) used, therefore there are only two conditions. One condition included a lot of sexual depictions like sexual behavior and sexual talk, the other condition showed more platonic scenes, without any sexual content. Both clips were made from the same episode and with the same people that were in the television program. Only different parts of the episode were used. One video (sexual condition) was only about partying, drinking alcohol, showing sexual behavior and sexual talk. The other video (neutral condition) showed people eating breakfast and going on dates without any sexual content, the people in the show were only doing normal activities. The video in each condition was around 15 minutes. The original episodes were shortened because the study was conducted online and was focused on
adolescents. These are two factors which have been shown to reduce participants willingness to participate in long studies (Guo, Kim & Rubin, 2014; Valkenburg & Piotrowski, 2014).
Manipulation of identification
To manipulate the level of identification of the participant while watching the video, two brief instructions were created. One instruction told the participant to really feel the emotions with the characters in the video, to feel like they were their friends and they can feel the same emotions. The other instruction told the participant to watch the video with a critical lens. From a perspective of a film director, not focusing on the emotions at all. The goal was that the neutral instruction would not create identification with the characters on Ex on the Beach and the manipulated instruction would create identification. This instruction is created with two conditions based on the mechanisms of identification (Cohen, 2001). People identify more with the character when they feel with the character (empathic), think like the character (cognitive) internalize the goals (motivation), and loose self-awareness (absorption; Cohen, 2001). In one instruction the participants were told to feel with the characters, to think like them, to understand them, and to feel connected. In the other condition of the instruction the
opposite perspective of identification was chosen, to not feel, understand, think like, and connect with the characters. The instruction (manipulation of identification) is treated as a moderator in this study.
Measures
The survey measured the treatment check of sexualization, the treatment check of identification, the manipulation check of identification, permissive sexual attitudes and variables for randomization check. All measures are based on 5-point rating scales (1 = strongly disagree, 5 = strongly agree). All items for all assessments were Dutch.
Treatment check of sexualization
In order to check if the sexualization exposure (video containing sexual content) worked, one items was used. Participants were asked to indicate on a scale of 1 (not at all) to 5 (absolutely) to what extent they saw sexual content in the video.
Treatment check of identification
In order to check if the identification exposure (instruction to identify) worked, two items were used. Participants were asked to indicate on a scale of 1 (not at all) to 5 (absolutely) to what extent the instruction told them to watch critically and to what extent the instruction told them to recognize themselves in the characters. These two items were computed into a new variable. One of the items was reverse coded.
Manipulation check of identification
Identification was measured with an existing scale from Godlewski and Perse (2010). They based their scale on writings on identification by Cohen (2001) and Eyal and Rubin (2003). The questions on the scale were answered on a 5-point Likert scale. An example of a
measurement item is: “When I watched the video, I felt the emotions of the participants”. There was chosen to measure identification because it was interesting to see if not only a treatment would be recognized, but if the manipulation would also work as intended. This measure served as a manipulation check. The study was the first on this topic to manipulate identification with an instruction. It was therefore relevant to measure if the manipulation worked. The scale consisted of six items that were computed into a new variable by
calculating the mean. The 6-item scale also proved to be reliable as indicated by a Cronbach’s Alpha of .874. The new item is identification levels (M = 2.60, SD = .88). The whole scale can be found in Appendix A.
Permissive sexual attitudes
Permissive sexual attitudes were measured with the brief sexual attitudes scale by Hendrick, Hendrick & Reich (2006). Their scale showed reliability and validity in previous research (Hendrick, Hendrick & Reich, 2006; Bond & Drogos, 2014). An example of a measurement item is “Casual sex is acceptable”, or “It is okay to have ongoing sexual relationships with more than one person at a time”. The scale consisted of ten items that were computed into a new variable by calculating the mean. The 10-item scale also proved to be reliable as indicated by a Cronbach’s Alpha of .851. The new item is permissive sexual attitudes (M = 2.80, SD = .74). The whole scale can be found in Appendix B.
Variables for randomization check
In this experiment the following variables for randomization check and general information were implemented: age, gender, current level of education, if the participant had already seen the (edited) episode, and how often the participants watched the series Ex on the Beach. The mean age of the participants was 17.5 (SD = 1.18), 35% were males and 65% were females.
Almost a third of the participants are in the education level MBO (29,7%). 18,8% is currently in Havo and 17,2% is in Vwo/ Atheneum. Almost half of the participants 46% (n = 59) had seen the whole season of Ex on the Beach Double Dutch All Stars and about one third 31% (n = 40) had not seen the show but had heard about it. Only 3 participants had never heard of the show. Almost half of the participants had seen the episode (46%) and almost half of them had not seen the episode before (45%). 11 participants could not remember if they had seen the episode before.
Results
Randomization check
In order to check if participants’ age was comparable over the conditions (two conditions of the video that was shown and two conditions of the instruction before the video was shown), a two-way ANOVA was conducted with sexual content (video) and identification (instruction) as independent variables and age as dependent variable. In total there are four conditions (2x2 design). The ANOVA showed that there was no difference in participants’ age across the experimental conditions: the sexual content and manipulated instruction group (M = 17.30,
SD = 1.21), the sexual content and neutral instruction group (M = 17.33, SD = 1.06), the
neutral content and neutral instruction group (M = 17.43, SD = 1.31) and the neutral content and manipulated instruction group (M = 17.74, SD = 1.15), F(3, 124) = .979, p = .405, Eta2 = .023. The age is well randomized across the conditions. In order to check if gender was also comparable over the conditions, a Chi-square test was conducted as well. The Chi Square showed that participants’ gender was not significantly different across the conditions. The Chi Square is .77 with a significance of p = .86 which is higher than p = .05. The randomization according to gender is also well done.
Treatment check of sexual content
In order to check if the participants of the study recognized the video conditions as sexual and neutral a two-way ANOVA was conducted. With the video conditions as independent
variable and treatment check questions as dependent variable. The ANOVA showed that there was a significant difference between conditions. The participants who saw the sexual content (M = 4.25, SD = .92) reported significantly different scores than the participants who saw the neutral content (M = 3.34, SD = 1.23), F(1,126) = 22.753, p < .001, Eta2 = .153 on the
question whether the participant had seen sexual content in the video. According to the ANOVA’s the manipulation worked as planned and was recognized.
Treatment and manipulation check of identification
In order to check if the participants of the study recognized the instruction as one that told them to identify with the characters in the video or to watch the video critically a two-way ANOVA was conducted. With the instruction conditions as independent variable and treatment check questions as dependent variable. The ANOVA showed that there was a significant difference between conditions. The identification condition (M = 3.04, SD = .82) was seen significantly different from the neutral condition (M = 2.50, SD = .75), F(1,126) = 15.173, p < .001, Eta2 = .107, the participants recognized the identification condition and neutral condition and answered the questions significantly different. The treatment was thus recognized as intended.
In order to test if the manipulation of identification (instruction) changed identification levels, a two-way ANOVA was conducted to check whether there was a significant effect between the people who were manipulated to identify with the characters and the people who were told to watch the video neutrally. With the ANOVA identification levels was used as a dependent variable and the instruction (manipulated vs. neutral) and the video (sexual vs.
neutral) were used as independent variables. Regarding the assumptions, the identification levels among the four conditions did almost all deviate significantly from normal, sexual video D(63) = .120, p = .025, neutral video D(65) = .211, p < .001, manipulated instruction
D(68) = .093, p = .200, neutral instruction D(60) = .228, p < .001. Levene’s test showed that
the variances for identification levels were equal for the video conditions F(1, 126) = 1.898, p = .171 and instruction conditions F(1, 126) = .155, p = .695. Homogeneous variance can be assumed, the assumption is tenable.
Regarding to test the difference between the participants in the identification conditions who reported identification levels, no significant effect was found between the manipulated group (M = 2.61, SD = .89) and the neutral group (M = 2.60, SD = .87), F(1, 124) = .004, p = .95, Eta2 = .00. What is unexpected here is that the exposure did seem to work, participants did recall the instructions, but the instructions did not affect their level of identification. In extension the same ANOVA was looked at to find out if there was an interaction effect between the instruction and the video on the identification levels from the participants. No significant interaction effect was found F(1, 124) = .569, p = .452, Eta2 = .005. However, there was a significant effect found F(1, 124) = 3.09, p = .049, Eta2 =.032 for the dependent variable identification levels with the independent variable video conditions with either sexual content (M = 2.45, SD = .96) and neutral content (M = 2.75, SD = .77). This means that participants could identify themselves more with the characters when the content was neutral in the video than when the content was sexual.
Tests of hypotheses
To test all the hypotheses at once a two-way ANOVA was conducted, with sexual permissive attitudes as a dependent variable and the video conditions (sexual vs. neutral) and instruction conditions (identification vs. neutral) as independent variables. The test of assumptions showed that the permissive sexual attitudes among the four conditions did deviate
significantly from normal, sexual video D(63) = .145, p = .002, neutral video D(65) = .182, p < .001, manipulated instruction D(68) = .111, p = .036, neutral instruction D(60) = .138, p = .006. Levene’s test showed that the variances for permissive sexual attitudes were roughly equal for the video conditions F(1, 126) = 2.404, p = .124 and instruction conditions F(1, 126) = .382, p = .538. Homogeneous variance can be assumed.
To check the effects of sexual content (hypothesis 1) and the interaction between sexual content and identification (hypothesis 2) permissive sexual attitudes, a two-way ANOVA was conducted. When looking at hypothesis 1, the effect of sexual content on permissive sexual attitudes, there was no significant effect found, F(1, 124) = 2.18, p = .142, Eta2 = .017. Participants in the condition of the sexual content (M = 2.70, SD = .79) did not score significantly different from the participants in the condition where the neutral video was shown (M = 2.90, SD = .67). Thus, hypothesis 1 is rejected. In contrary to what was expected, participants who saw the neutral video (without any sexual content), scored even a bit higher on permissive sexual attitudes than participants who saw the video with sexual content. This means that they report more permissive sexual attitudes when viewing the neutral video versus the sexualized video.
Table 1. Descriptive statistics on experimental conditions. Mean and standard deviations.
Sexual content Neutral content
Manipulated instruction 2.68 (.67) 2.88 (.77)
Neutral instruction 2.73 (.93) 2.92 (.55)
Hypothesis 2 posits that adolescents have stronger permissive sexual attitudes when viewing sexualized content in reality television, and that this effect will be stronger for high levels of identification with the characters than low levels of identification. Identification did not have a main effect on sexual permissive attitudes. There was no significant difference found between the conditions manipulated instruction (M = 2.78, SD = .72) and neutral instruction (M = 2.83, SD = .76), F(1,124) = .112, p = .738, Eta2 = .001. There was also no interaction effect found for the factors manipulated instruction with sexual content (M = 2.68,
SD = .67), manipulated instruction with neutral content (M = 2.88, SD = .77), neutral
instruction with sexual content (M = 2.73, SD = .93) and neutral instruction with neutral content (M = 2.92, SD = .55), F(1, 124) = .001, p = .975, Eta2 = .000. The instruction did not
have an influence on how the video content was perceived and what the participant reported in the questions about permissive sexual attitudes. Hypothesis 2 is rejected as well.
Post hoc analysis
For post hoc analysis a mediation test was performed. The mediation test was run because no significant effects and interaction were found on the dependent variable permissive sexual attitudes. Sexualized content did not seem to increase permissive sexual attitudes of the participants. This could be a result of identification being determined by the content of the video, instead of the instruction which was manipulated. Sexualized content seemed to reduce identification levels. This implies that identification could be a mediator of the effect of the video on permissive sexual attitudes. Therefore, a mediation analysis is required.
Bootstrapping analysis was used to test for mediation using the SPSS process macro that Hayes (2013) developed. The estimated model is depicted in Figure 1. No residual direct effect was found for the video (sexual content vs. neutral content) on permissive sexual attitudes B = .071, t = .606, p = .546. A significant effect of video on identification was found (B = .304, t = 1.984, p < .05). Also, a significant effect of identification was found on
permissive sexual attitudes (B = .404, t = 6.051, p < .01). Thus, an indirect effect was found: the video exerts an indirect effect through levels of identification on permissive sexual attitudes. The video predicts identification which then predicts permissive sexual attitudes. Accordingly, the indirect effect of sexual content on permissive sexual attitudes through identification is significant. The bootstrapped 95%-confidence interval (CI) (k = 5000) ranges from .004 to .274 and does thus not include zero. This means that when participants saw the neutral video they identified more and then reported higher permissive sexual attitudes.
.304* .404*
.071
Fig 2. Mediation model. *p < .05; The model is significant, F(2, 125) = 19.724, p < .01, explaining about 24% of
the variance.
Discussion
The goal of this study was to discover if sexual content in reality television has an effect on permissive sexual attitudes of late adolescents. The role of identification as a possible
moderator was also taken into account. An online experiment was conducted with adolescents as participants, the participants red an instruction (manipulated to identify with characters in the video vs. neutral) and watched a video (sexual content vs. neutral content).
The analyses did not confirm with the hypotheses. Hypothesis 1 that tested the effect of sexual content on permissive sexual attitudes was rejected. According to the results, no significant effect was found. Participants did not report higher permissive sexual attitudes after watching a video with sexual content. This is in contrast with what was expected based
Video
Identification
on several studies (e.g. Bond & Drogos, 2014; Ward & Friedman, 2006). Despite the lack of a significant effect, participants who were classified to the neutral condition reported higher permissive sexual attitudes than participants who were exposed to the sexual content condition. One speculation is that the participants in the sexual condition could feel disgust when watching the video which could have an opposite influence. This can be linked to cognitive dissonance theory (Festinger & Carlsmith, 1959), there was an imbalance between what was shown and what was expected to be showed. This dissonance is resolved by answering the questions in an opposite way (Coyne, Robinson & Nelson, 2010). Another reason for the rejection of the hypothesis could be that there is only an effect when cultivation occurs. According to cultivation theory, repeated exposure to the same content can have an effect (Morgan & Shanahan, 2010; Ward, 2003). Vandenbosch and Eggermont (2011) claim that the influence of the stimulus is not sufficient enough because of the short exposure. Hypothesis 2 that tested the interaction effect between the instruction and the video on permissive sexual attitudes was also rejected. The instruction that was manipulated to make the participants identify with the characters in the video did not have an effect on permissive sexual attitudes. According to several studies (Moyer-Gusé, Chung, & Jain, 2011; Ward & Friedman, 2006) identification with the characters in the reality show could have a bigger effect. The results of the study did not support that identification had a bigger effect, but it also did not provide evidence against the importance of identification, because the
manipulation did not work properly. The participants seemed to remember the instruction, but the instructions did apparently not elicit differential levels of identification.
Post hoc analysis showed that the level of identification is actually a mediator in this study. The level of identification someone has with the characters in the reality television content mediates the permissive sexual attitudes. This means that when a participant watched the reality television content and identified with the characters, he or she reported higher
permissive sexual attitudes. This result is in line with what Hoffner and Buchanan (2005) also found: identification mediates the audience responses when watching television. Identification as a mediator also supports a growing body of research that suggest that media exposure and sexual attitudes are influenced by identification by characters (Bond & Drogos, 2014; Moyer-Gusé et al., 2011; Ward & Friedman, 2006). The remarkable result in this study is that the participants identified more with the characters in the neutral video than with the characters in the sexual video, and that the participants who identified more reported higher permissive sexual attitudes. This is confirmed by Bandura (2004) and Moyer-Gusé (2008): people
identify more with similar people. The characters in the neutral version could be seen as more similar to the participants of the study. And therefore, they identified with these characters instead of characters in the sexualized version. Apparently, strong sexualization in reality television does not seem to promote social learning, but it offends people. This in contrast to social cognitive theory by Bandura (1986). Maybe the sample did identify more with the neutral version because the mean age was 17, and the legal age to go clubbing and drinking alcohol is 18 in the Netherlands. In the sexualized version there was partying and drinking depicted. Most of the characters were drunk when they engaged in sexual behavior. Since the participants (according to the mean age) weren’t old enough to do this, maybe they could not identify with the characters. Instead they identified more with the characters in the neutral version who were doing neutral activities.
The answer to the research question Does viewing sexualized reality television have an
effect on permissive sexual attitudes of adolescents? is the current study does not replicate
earlier findings, but that does not mean that the effect does not exist. This study provides evidence for identification levels to be a mediator which is very interesting. The reason why the results are not consistent with earlier findings could be that this is an experiment and other studies were correlational (Bond & Drogos, 2014). Which could mean that the relation is the
opposite, people who report permissive sexual attitudes, watch sexual unscripted television more often. Other studies examined normal television shows, which is a different genre from reality television and could therefore have a different effect (Ward & Friedman, 2006). Vandenbosch and Eggermont (2013) also did not find a long-term effect of sexualized reality television on the sexual attitudes of adolescents with their longitudinal study. They even question the results of cross-sectional studies that implied an effect based on correlations. Ex on the beach has never been studied before, which can also be a reason that the results do not replicate earlier findings.
As with every study, there are limitations that needs to be acknowledged. First of all, the instruction manipulation did not work, as mentioned above. Identification was not correctly manipulated in this study. However, this study was the first to try to manipulate identification with this topic. And still, identification was measured to make sure it did or did not play a role in the study. Eventually it did, as a mediator. Thus, identification was still a valuable item in this study, but the manipulation should be done differently next time. In future studies, the identification should be manipulated in a different way, or maybe not at all. A better way to manipulate the identification level is to use different characters. For high identification, use characters in the video that look like participants and have the same nationality. For low identification levels, use characters from another nationality and
appearance. This would make the study very difficult because the internal validity needs to be high as well. And all conditions should be connected to the answers in the survey, to make sure the participant does identify with the characters.
Another limitation of this study could be that almost everyone in the study already knew about the series Ex on the Beach and almost half of the participants had already seen the episode and almost half of the participants had already seen the whole season. It is therefore questionable if the video conditions even had an effect. The neutral version could have no
effect because participants have already seen the (edited) episode. Nevertheless, the
immediate effect was measured here and not the cultivated effect. Also, the series aired six months before the time of the study, so the immediate effect made sense. According to Ward and Friedman (2006), the fact that this show is still (not the season that was used for this experiment, but other seasons) on air and making new seasons, can contribute to schemas of viewers. And each time they watch this show, their schema grows. This is connected with cultivation theory (Ward & Friedman, 2006).
Although this study was the first to examine the effects of Ex on the beach, future research is needed to explicate the effects and address the limitations. In this study, only the attitudes towards sex were measured. For future research it could be very interesting to study the actual behavior of adolescents after repeatedly being exposed to highly sexualized content on television. With a longitudinal study for instance. Because there is a lack of studies on this topic, especially with the shift of showing more and more sexual depictions on television, it is very interesting to study this effect on different age groups. Looking at identification levels as a mediator could explain more effects and would be very interesting to examine, especially in longitudinal studies.
Conclusion
Adolescents and the effect of highly sexualized reality television. In this study, a mediation effect was found of identification levels when watching the reality television show. The sexualized content had no effect on the permissive sexual attitudes, but adolescents reported higher permissive sexual attitudes when they identified with neutral reality content. This study has some limitations. The topic is very new and there has not been a lot of research done on this topic. To study the actual effects on adolescents, a longitudinal study on their sexual behavior with treating identification levels as a mediator is a great idea for future
research. Future research is needed because there are still concerns that the media do not portray sexual behavior in a healthy way.
References
Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice Hall.
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.
Englewood Cliffs, NJ, 1986.
Bandura, A. (2004). Social cognitive theory for personal and social change by enabling media. In A. Singhal, M. J. Cody, E. M. Rogers, & M. Sabido (Eds.),
Entertainment-education and social change: History, research, and practice (pp. 75–96). Mahwah, NJ:
Lawrence Erlbaum.
Basil, M. D. (1996). Identification as a mediator of celebrity effects. Journal of Broad-
casting and Electronic Media, 40(4), 478–495.
Beck, D., Hellmueller, L. C., & Aeschbacher, N. (2012). Factual entertainment and reality TV. Communication Research Trends, 31, 4-27.
Bond, B. J., & Drogos, K. L. (2014). Sex on the shore: Wishful identification and parasocial relationships as mediators in the relationship between Jersey Shore exposure and
emerging adults' sexual attitudes and behaviors. Media Psychology, 17(1), 102-126. Boon, S. D., & Lomore, C. D. (2001). Admirer-celebrity relationships among young adults:
Explaining perceptions of celebrity influence on identity. Human Communication
Research, 27, 432–465.
Brown, J. D., Halpern, C. T., & L’Engle, K. L. (2005). Mass media as a sexual super peer for early maturing girls. Journal of Adolescent Health, 36, 420–427.
Brown, J. D., Steele, J. R., & Walsh-Childers, K. (2001). Sexual teens, sexual media:
Bushman, B. J., & Huesmann, L. R. (2001). Effects of televised violence on aggression. Handbook of children and the media, 223-254.
Caughey, J. L. (1986). Social relations with media figures. Inter/media: Interpersonal
communication in a media world, 3, 219–252.
Cohen, J. (2001). Defining identification: A theoretical look at the identification of audiences with media characters. Mass Communication & Society, 4, 245–264.
Collins, R. L., Elliott, M. N., Berry, S. H., Kanouse, D. E., & Hunter, S. B. (2003). Entertainment television as a healthy sex educator: The impact of condom-efficacy information in an episode of Friends. Pediatrics, 112, 1115–1121.
Coyne, S. M., Robinson, S. L., & Nelson, D. A. (2010). Does reality backbite? Physical, verbal, and relational aggression in reality television programs. Journal of Broadcasting
& Electronic Media, 54(2), 282-298.
Eyal, K., & Rubin, A. M. (2003). Viewer aggression and homophily, identification, and parasocial relationships with TV characters. Journal of Broadcasting & Electronic
Media, 47, 77–98.
Festinger, L. and Carlsmith, J. M. (1959). Cognitive consequences of forced compliance. Journal of Abnormal and Social Psychology, 58(2): 203–210
Franko, M., & Krieger, L. (2011). Reality on MTV. Gender portrayals on MTV reality
programming. Los Angeles, CA: Parents Television Council.
Gerbner, G., Gross, L., Morgan, M., Signorielli, N., & Shanahan, J. (2002). Growing up with television: Cultivation processes. In Media effects: Advances in theory and research (pp. 43–95). Mahwah, NJ: Erlbaum.
Godlewski, L. R., & Perse, E. M. (2010). Audience activity and reality television:
Identification, online activity, and satisfaction. Communication Quarterly, 58(2), 148-169.
Greeson, L. E., & Williams, R. A. (1986). Social implications of music videos for youth: An analysis of the content and effects of MTV. Youth and Society, 18, 177–189.
Gruber, E., & Grube, J. W. (2000). Adolescent sexuality and the media: A review of current knowledge and implications. Western Journal of Medicine, 172, 210–214.
Guo, P. J., Kim, J., & Rubin, R. (2014). How video production affects student engagement: An empirical study of MOOC videos. In Proceedings of the first ACM conference on
Learning@ scale conference (pp. 41-50).
Hayes, A. F. (2017). Introduction to mediation, moderation, and conditional process
analysis: A regression-based approach. Guilford publications.
Hendrick, C., Hendrick, S. S., & Reich, D. A. (2006). The brief sexual attitudes scale. Journal
of Sex Research, 43, 76–86.
Hoffner, C., & Buchanan, M. (2005). Young adults’ wishful identification with televi- sion characters: The role of perceived similarity and character attributes. Media Psychology, 7, 325–351.
Huesmann, L. R. (1998). The role of social information processing and cognitive schema in the acquisition and maintenance of habitual aggressive behavior. In Human
aggression (pp. 73-109). Academic Press.
Huston, A. C., Wartella, E., & Donnerstein, E. (1998). Measuring the effects of sexual content in the media: A report to the Kaiser Family Foundation. Menlo Park, CA: Kaiser Family Foundation.
Jansen, D., & Tienhooven, G. (2019, November 27). ‘Het genre guilty pleasure-tv is besmet geraakt’. Retrieved via https://www.ad.nl/show/het-genre-guilty-pleasure-tv-is-besmet-geraakt~aaaac38b/?referrer=https://www.google.com/
Kijkwijzer (n.d.). Ex on the Beach Double Dutch - All Stars. Retrieved via https://www.kijkwijzer.nl/ex-on-the-beach-double-dutch-all-stars/page26-0-1433783.html
Maccoby, E. E., & Wilson, W. C. (1957). Identification and observational learning from films. The Journal of abnormal and social psychology, 55(1), 76.
Morgan, M., & Shanahan, J. (2010). The state of cultivation. Journal of Broadcasting and
Electronic Media, 54, 337–355.
Morgan, M., & Signorielli, N. (1990). Cultivation analysis: New directions in media effects
research (Vol.108). Sage Publications, Inc.
Moyer-Gusé, E. (2008). Toward a theory of entertainment persuasion: Explaining the ´ persuasive effects of entertainment-education messages. Communication Theory, 18, 407–425.
Moyer-Gusé, E., Chung, A. H., & Jain, P. (2011). Identification with characters and discussion of taboo topics after exposure to an entertainment narrative about sexual health. Journal of Communication, 61, 387–406.
Murray, S., & Ouellette, L. (2009). Reality TV: Remaking television culture. New York, NY: New York University Press.
Nabi, R. L., Biely, E. N., Morgan, S. J., & Stitt, C. R. (2003). Reality-based television programming and the psychology of its appeal. Media Psychology, 5, 303–330.
Nielsen. (2009). Nielsen 2009 report on television. New York, NY: Nielsen Media Research. Papacharissi, Z., & Mendelson, A. (2007). An exploratory study of reality appeal: Uses and
gratifications of reality TV shows. Journal of Broadcasting and Electronic Media, 51, 355–370.
Riddle, K., & De Simone, J. J. (2013). A Snooki effect? An exploration of the surveil- lance subgenre of reality TV and viewers’ beliefs about the ‘‘real’’ real world. Psychology of
Popular Media Culture, 2, 237–250.
Rubin, A. M., Perse, E. M., & Taylor, D. S. (1988). A methodological examination of cultivation. Communication Research, 15(2), 107–134.
Shanahan, J., Shanahan, J., James, S., & Morgan, M. (1999). Television and its viewers:
Cultivation theory and research. Cambridge university press.
Takken, W., & Nieber, L. (2019, November 29). De camera verwacht seks, drank en tranen. Retrieved via https://www.nrc.nl/nieuws/2019/11/29/de-camera-verwacht-seks-drank-en-tranen-a3982220
Tian, Y., & Yoo, J. H. (2015). Connecting with the biggest loser: An extended model of parasocial interaction and identification in health-related reality TV shows. Health
communication, 30(1), 1-7.
Valkenburg, P. M., & Piotrowski, J. T. (2017). Plugged in: How media attract and affect
youth. Yale University Press.
Vandenbosch, L., & Eggermont, S. (2011). Temptation Island, The Bachelor, Joe Millionaire: A prospective cohort study on the role of romantically themed reality television in
adolescents’ sexual development. Journal of Broadcasting and Electronic Media, 55, 563–580.
Vandenbosch, L., Vervloessem, D., & Eggermont, S. (2013). “I might get your heart racing in my skin-tight jeans”: Sexualization on music entertainment television. Communication
Studies, 64, 178–194.
Ward, L. M. (1995). Talking about sex: Common themes about sexuality in the prime-time television programs children and adolescents view most. Journal of youth and
Ward, L. M. (2002). Does television exposure affect emerging adults' attitudes and assumptions about sexual relationships? Correlational and experimental confirmation. Journal of youth and adolescence, 31(1), 1-15.
Ward, L. M. (2003). Understanding the role of entertainment in the sexual socialization of American youth: A review of empirical research. Developmental Review, 23, 347–388. Ward, L. M., & Friedman, K. (2006). Using TV as a guide: Associations between television
viewing and adolescents’ sexual attitudes and behavior. Journal of Research on
Adolescence, 16, 133–156.
Ward, L. M., & Rivadeneyra, R. (1999). Contributions of entertainment television to
adolescents’ sexual attitudes and expectations: The role of viewing amount versus viewer involvement. Journal of Sex Research, 36, 237–249.
Wright, P. J., Randall, A. K., & Arroyo, A. (2013). Father-daughter communication about sex moderates the association between exposure to MTV’s 16 and Pregnant/Teen Mom and female students’ pregnancy-risk behavior. Sexuality & Culture, 17, 50–66.
Zurbriggen, E. I., & Morgan, E. M. (2006). Who wants to marry a millionaire? Reality dating television programs, attitudes toward sex, and sexual behaviors. Sex Roles, 54, 1–17.
Appendix
Appendix A
Scale to measure identification levels based on the scale by Godlewski and Perse (2010). 1. Ik begrijp de deelnemers van Ex on the Beach
2. Toen ik de video bekeek, begreep ik hoe de deelnemers zich voelen 3. Toen ik de video bekeek, voelde ik met de deelnemers mee
5. Toen ik de video bekeek, had ik het idee dat ik de persoonlijkheid van de deelnemers kon begrijpen
6. Toen ik de video bekeek, kon ik echt in het hoofd van de deelnemers kruipen
Appendix B
Scale to measure permissive sexual attitudes by Hendrick, Hendrick & Reich (2006). 1. Ik hoef niet in een relatie met een persoon te zitten om seks met hem/haar te hebben 2. Casual seks (seks zonder een relatie) is acceptabel
3. Ik zou graag seks hebben met veel mensen 4. One-night stands zijn soms erg plezierig
5. Het is normaal om meerdere seksuele relaties tegelijk te hebben 6. Seks als het doen van gunsten is prima als iedereen akkoord gaat 7. De beste seks is zonder verplichtingen (zonder relatie)
8. Het leven heeft minder problemen als mensen meer vrije seks konden hebben 9. Het is mogelijk om van seks te genieten terwijl je de persoon niet erg leuk vindt 10. Het is goed om seks te hebben voor ontspanning
