Entertainment Education and Condom Use among Emerging Adults:
A Study on the Prototype-Willingness Model and Differences between Narrative and Nonnarrative Entertainment Education
Alma van den Berg 6048501 Master’s Thesis
Graduate School of Communication
Research Master’s Programme Communication Science Dr. J.M.F. van Oosten
Abstract
Previous studies have demonstrated that entertainment education can be an effective strategy to positively influence behavior-related outcomes. Research on whether entertainment education can be effective in relation to condom use is however limited. The present study investigated effects of exposure to entertainment education, containing a conversation about condom use, on emerging adults’ intention and willingness to use condoms. Moreover, on the basis of the prototype-willingness model, it was examined whether exposure to entertainment education could affect the favorability of the prototype that uses condoms. Furthermore, differences between effects of narrative and nonnarrative entertainment education were investigated. An online experiment among 134 college students in the age of 18 to 25 was conducted. Results showed that exposure to entertainment education directly influenced the intention to use condoms, but did not directly influence the willingness to use condoms. The favorability of the prototype that uses condoms was not influenced by the exposure to entertainment education. No differences on the outcomes were found between narrative and nonnarrative entertainment education.
Keywords: entertainment education, prototype-willingness model, condom use, emerging adults
Entertainment Education and Condom Use among Emerging Adults
During adolescence young people start to engage in different types of risk behavior. The prevalence of engagement in these risk behaviors, and especially sexual risk behavior, peaks during emerging adulthood, the age period between 18 and 25. In this phase between adolescence and adulthood people start to develop romantic feelings for others and discover their sexuality. This process of sexual discovery can be accompanied by having sexual intercourse without protection or hooking up with casual partners (Arnett, 2014). A study in the Netherlands on youth’s sexual behavior demonstrated that just 22% of the boys and 34% of the girls in the sample used condoms at the beginning of their relationship with their last sex partner and that 13% stopped the use of condoms after a week (De Graaf, Kruijer, Van Acker, & Meijer, 2012). Another research among Dutch men and women aged 15 to 71 reported that 48% of the men and 60% of the women did not use a condom during the last time they had casual (heterosexual) sex (Goenee, Kedde, & Picavet, 2012). Not using a condom during sexual intercourse can have negative consequences, such as the spread sexual transmitted diseases (STDs) of which some might even cause infertility (De Graaf et al., 2012).
Concerns regarding low condom use among youth and its consequences have led researchers to investigate whether media might play a role in youth engaging in sexual intercourse without using a condom. It has been found that exposure to sexual media content can be related to negative attitudes towards condom use and engagement in unprotected sex (e.g., O’Hara, Gibbons, Gerrard, Li, & Sargent, 2012; Peter & Valkenburg, 2011; Træen et al., 2014; Wingwood et al., 2001). However, not much is known about whether sex-related media content can also influence young people’s engagement in protected sex. A predictor of actual condom use is one’s intention to use condoms, as these were found strongly correlated (Sheeran & Orbell, 1998). Although studies have found that exposure to sexual media content
can lower intention to use condoms (e.g., Wingwood et al., 2001), studies on whether specific media exposure can increase intention to use condoms are limited. However, this would be interesting to investigate, because of intention’s predictive value of actual condom use.
Furthermore, another predictor of performing a certain type of behavior is willingness (Gerrard, Gibbons, Houlihan, Stock, & Pomery, 2008). Willingness indicates the likeliness of engaging in a certain type of behavior when a situation facilitates this behavior (Gerrard et al., 2008). Few studies have examined effects of media exposure on willingness. It was found that exposure to risk behavior in media content can affect the willingness to engage in that risk behavior (Gibbons et al., 2010). Effects of media exposure on willingness to engage in health behavior have never been investigated. As media exposure can affect willingness to engage in risk behavior, it may also be interesting to investigate this relationship in the context of health behavior, such as the willingness to use condoms.
One type of media content that is likely to affect health behavior related outcomes is entertainment education, as this is a form of media content that contains health messages. The health message in entertainment education is, intended or unintended, incorporated in
entertainment media content and can intentionally or incidentally promote health behavior (Moyer-Gusé, 2008). Effects of exposure to entertainment education have been investigated on different topics, such as organ donation (Bae, 2008; Bae & Kang, 2008) and breast cancer (Hether, Huang, Beck, Murphy, & Valente, 2008). These studies demonstrated that
entertainment education may be an effective way of positively influence attitudes and behaviors. Studies on effects of exposure to entertainment education on condom use are scarce. This relation however would be interesting to investigate, because it may potentially promote and increase condom use among young people. The first aim of the present study is therefore to investigate whether entertainment education can influence, two predictors of actual behavior, intention to use condoms and willingness to use condoms.
Both intention and willingness are incorporated in the prototype-willingness model and this may be a particularly relevant model to investigate the effects of entertainment education. The prototype-willingness model is based on the assumption that one’s willingness to engage in a certain type of behavior is associated with the social image (prototype) one has of a person engaging in that behavior. The more favorable the prototype, the more likely the individual is to engage in that behavior (Gibbons & Gerrard, 1995; Gerrard et al., 2008). Although it is assumed that the construction of these prototypes and their favorability can be influenced by media content, research investigating media effects with the
prototype-willingness model as underlying mechanism is limited. Moreover, entertainment education has never been linked to the prototype-willingness model. The second aim of the present study is therefore to examine effects of exposure to entertainment education with the use of the prototype-willingness model and especially investigate whether exposure to entertainment education can directly influence the favorability of the prototype that uses condoms.
Research on entertainment education has always focused on entertainment content with a narrative. By embedding a health message in a narrative, it is assumed to arouse less reactance to the message and thus be more effective than for example a health campaign with a message intended to persuade (Moyer-Gusé, 2008). However, effectiveness of a health message incorporated in entertainment content without a narrative has never been
investigated. The present study’s third aim is therefore to investigate differences in effects of narrative entertainment education and nonnarrative entertainment education, both containing a health message regarding condom use.
Entertainment Education and Young People’s Health Behavior
To prevent young people from engaging in harmful risk behaviors, promoting health behaviors through media content may be an effective way of influencing their behavior. As entertainment education contains health messages, it can be a strategy to establish this
positive influence on youth’s behavior. Moyer-Gusé (2008) defines entertainment education as “prosocial messages that are embedded into popular entertainment media content” (p. 408). The health message in entertainment education can be intended or unintended by the content’s creator and therefore intentionally or incidentally promote health behaviors. This promotion of health behavior can be established by portraying or discussing the type of health behavior in a positive way or portraying or discussing the risk behavior in a negative way (Moyer-Gusé, 2008).
Previous studies have found that exposure to entertainment education can in fact be effective in positively affecting behavior-related outcomes. Entertainment education can increase one’s knowledge on a certain health topic and induce searching for additional information on the topic (Brodie et al., 2001; Kennedy, O’Leary, Beck, Pollard, & Simpson, 2004). Although effects on one’s knowledge were found to be slightly stronger than on attitudes, intentions and behavior (Shen & Han, 2014), several studies have shown that exposure to entertainment education can indeed positively influence behavior-related outcomes. Attitude towards organ donation (Khalil & Rintamaki, 2013) and intention to donate organs (Bae, 2008; Bae & Kang, 2008) were increased by exposure to entertainment education on the topic of organ donation. Breast cancer related entertainment education has also positively affected attitudes and behaviors towards breast cancer (Hether et al., 2008). Attitudes and self-reported behaviors regarding obesity, hypertension and healthy eating were positively influenced by exposure to entertainment education on these topics (Valente et al., 2007). A study on exposure to entertainment education containing messages about HIV / AIDS showed that the exposure did not only result in higher self-efficacy with respect to prevention of HIV / AIDS, but also in reduction of the number of sexual partners and increased condom use (Vaughan, Rogers, Singhal, & Swalehe, 2000).
Other research on the relation between entertainment education and condom use is limited and has in most cases focused on memorizing the incorporated health message (Brodie et al., 2001; Kennedy et al., 2004). Collins, Elliott, Berry, Kanouse, and Hunter (2003) investigated whether viewers of the television series Friends were able to recall a message about condom failure in one of the episodes and found that two-third of the sample did. Of the ones who could recall the message, perceptions of condom efficacy were reduced. Farrar (2006) examined effects of portrayal of condom use by characters in a dramatic
television show and found that the exposure led to more positive attitudes towards condom use, as also found by Moyer-Gusé and Nabi (2010), but intention to use condoms was not affected. Because research on effects of exposure to entertainment education on predictors of sexual health behavior is scarce, these relationships should be further investigated.
Entertainment education’s effectiveness. The effectiveness of entertainment education is assumed to depend on the narrative element. Embedding a health message in a narrative ascertains that the persuasiveness of the message will be overcome (Moyer-Gusé, 2008). People experience a certain kind of persuasiveness when someone or something tries to persuade them to conform to a certain behavior. Tainting one’s need for independence, in the case of persuading them in a certain direction, will arouse reactance. This reactance results in resisting the persuasive message (Moyer-Gusé & Nabi, 2010).
Reactance can be overcome by incorporating the health message in a narrative. Moyer-Gusé (2008) distinguishes two aspects of a narrative that will lead to viewers not experiencing persuasiveness of the message: involvement with the narrative and involvement with the characters. Involvement with the narrative implies that a viewer becomes engaged in a certain storyline and focuses on all the events in that storyline. In this way the health
message is processed as part of the story and will therefore not be perceived as a message intended to persuade. The other aspect, involvement with characters, is made up of five
concepts. The first concept is identification. This means that a viewer takes on the role of the character in the storyline and takes on his or her perspective. Second, and related to
identification, is wishful identification. In this case the viewer does not only take on the character’s perspective, but also wants to become the character. Similarity of the viewer to the character also contributes to more involvement with the character. A viewer can also feel like he or she has relation with the character, this is defined as parasocial interaction. The last concept that contributes to the involvement with a character, is liking of the character. The more the viewer likes the character, the more involved he or she is with the character.
Thus, becoming involved in the narrative and involved with the character(s) of entertainment education is assumed to reduce resistance to the persuasiveness of the health message. Brown and Walsh-Childers (2002) mention that entertainment education is especially powerful for messages related to sexual health behavior, just because of not so obviously prescribing the behavior, but incorporating them in a storyline.
The Prototype-Willingness Model
Sexual health behavior thus may be increased by exposure to entertainment education. A relevant model for explaining effects on actual behavior is the prototype-willingness model. This model incorporates two predictors of actual behavior: intention and willingness. The intention to perform a certain behavior involves a reasoned process in which motivational factors are deliberated. In contrast, willingness, the likeliness to engage in a certain type of behavior when the situation facilitates it, involves less deliberation of the behavior and its possible consequences. Willingness adds to the predictive validity of intention (Pomery, Gibbons, Reis-Bergan, & Gerrard, 2009), hence including both factors will result in better prediction for actual behavior.
Intention and willingness share, according to the prototype-willingness model, the same antecedents, which are based on the theory of reasoned action (Fishbein & Ajzen,
1975). The first one is attitude towards the behavior. Secondly, subjective norms on the behavior: to what extent do others perform the behavior. The perception of the risks performing the behavior (or translated to health behavior, perception of the benefits) is the third antecedent (Gerrard et al., 2008).
Prototypes. Another important factor, especially in predicting the willingness to engage in a certain type of behavior, are the social images one has of a typical person engaging in that behavior. These social images are called prototypes. The more favorable a particular prototype, the more willing one is to engage in the behavior of that prototype (Gerrard et al., 2008; Gibbons & Gerrard, 1995). The prototypes people have, are constructed by several influences of their environment. Exposure to media content can contribute to forming prototypes by the portrayal of certain types of behaviors (Gerrard et al., 2008) and especially by showing what is considered as normal behavior in a society (Brown, 2010). The more certain behaviors are portrayed, the more normative they become, and the more
favorable they become (Gerrard et al., 2008).
For example, young people who are exposed to media content that portrays young people consuming alcohol may think this is normal behavior for someone of their age. The prototype of someone that consumes alcohol will therefore become more favorable and the willingness to drink alcohol will increase. A study by Gibbons et al. (2010) found evidence for this assumption: Exposure to alcohol consumption in movies increased participants’ alcohol consumption. This relation was fully mediated by the favorability of drinking prototypes and the willingness to drink alcohol.
Priming. The influence of media exposure on prototypes happens through the process of priming (Roberts, Gibbons, Kingsbury, & Gerrard, 2014). Priming is defined as “the incidental activation of knowledge structures, such as trait concepts and stereotypes, by the current situational context” (Bargh, Chen, & Burrows, 1996, p.230). By this means, when
thinking of a certain prototype, images of which the prototype is constructed can be activated and will increase the prototype’s favorability and subsequently increase the willingness to engage in the prototype’s behavior. The study by Gibbons et al. (2010) has shown that media exposure portraying risk behavior can increase the favorability of the prototype engaging in this risk behavior. However, whether this is also the case for health behavior is unknown and therefore needs to be investigated.
Prototype similarity. Besides the favorability of a prototype, another strong predictor of willingness as well as intention is the extent one is similar to the prototype in question. The more similar one is to the prototype, the more willing one is to perform the behavior of the prototype (Gerrard et al., 2008; Gibbons & Gerrard, 1995). In a meta-analysis of the prototype-willingness model Todd, Kothe, Mullan, and Monds (2014) found that prototype similarity was even a stronger predictor of intention than for willingness. They therefore suggested that research on the prototype-willingness model should always include prototype favorability and prototype similarity as two different constructs, because their influence on willingness and intention might differentiate.
Moderation of social comparison tendencies. Effects of prototypes on willingness may not be the same for every person. The extent to one is used to compare him or her self with others may moderate this effect (Gibbons & Gerrard, 1995). For people with high social comparison tendencies prototypes will have a stronger influence on their willingness to
engage in the behavior of the prototype than for people with low social comparison tendencies (Ouellette, Hessling, Gibbons, Reis-Bergan, & Gerrard, 2005). Therefore, social comparison tendencies should be included as a moderator in the relation between prototype favorability and willingness.
In sum, the prototype-willingness model assumes that both intention and willingness are predictors of performing a specific type of behavior. One important antecedent, especially
for willingness, is the prototype one has someone engaging in a certain type of behavior. Exposure to media content portraying a certain type of behavior can contribute in constructing prototypes. Exposure to media content containing sexual health related images is therefore likely to affect the favorability of the prototype that engages in this sexual health behavior. A powerful way of promoting sexual health behavior, without arousing reactance to the
persuasiveness of the health message, is to incorporate the messages in entertainment media (entertainment education). Therefore it is likely that exposure to entertainment education, containing a conversation about condom use, may affect the favorability of the prototype uses condoms and subsequently the intention and willingness to use condoms.
Narrative versus Nonnarrative Entertainment Education
As discussed, entertainment education’s effectiveness is assigned to its narrative. Although it has been stated that entertainment education can differ in media type, genre, message length and strategy (Greenberg et al., 2004), research has only focused on the
incorporation of a health message in entertainment content with a narrative. Nothing is known about effects of health messages incorporated in entertainment content without a narrative. However, Moyer-Gusé and Nabi (2010) investigated differences in effects between
entertainment education, a sexual health message incorporated in dramatic narrative
entertainment content, and a sexual health message in an informational news program. They found, as suggested, that entertainment education increased safe sex intentions, whereas the news program did not. But, for those who perceived the entertainment education content as intended to persuade, reactance was aroused and the message did not increase the intention to engage in safe sex. In contrast, reactance was not aroused for those who perceived the news program as intended to persuade.
Thus, entertainment education may be more effective in positively influencing intentions than a news program. However, effects of exposure to a health message
incorporated in entertainment content without a narrative are still unknown and therefore interesting to examine.
The Present Study
The present study aimed at investigating the influence of exposure to entertainment education on emerging adults’ intention and willingness use to condoms. Although research on effects of entertainment education on intention and willingness to use condoms is scarce, other studies on entertainment education have found to be effective in positively changing attitudes, intentions or actual behavior. Therefore, it was expected that entertainment
education, containing a conversation about condom use, would positively influence emerging adults’ intention to use condoms (H1a) and willingness (H1b) to use condoms.
Based on the prototype-willingness model it was expected that the relation between exposure to entertainment education and willingness and intention would be mediated by attitude, perceived benefit, subjective norm, prototype favorability, and prototype similarity. The following hypothesis was therefore conducted: exposure to entertainment education, containing a conversation about condom use, will positively influence attitude, perceived benefit, subjective norm, prototype favorability and prototype similarity (H2a) and these factors will mediate the relationship between exposure to entertainment education, containing a conversation about condom use, and intention to use condoms and willingness to use condoms (H2b). Furthermore it was hypothesized that social comparison tendencies would moderate the relation between prototype favorability and willingness: For those with high social comparison tendencies, as prototype favorability increases, the willingness to use condoms increases (H2c).
The third aim of the present study was to examine differences between narrative and nonnarrative entertainment education. Because a narrative is suggested to overcome the arousal of reactance to the incorporated health message in entertainment education, it was
expected that narrative entertainment education, containing a conversation about condom use, would have a more positive influence on all factors related to condoms use of the prototype-willignness model than nonnarrative entertainment education (H3).
Method Sample and Experimental Design
Participants were recruited through the online research facility LAB of the Faculty of Social and Behavioral Sciences of the University of Amsterdam. The sample consisted of 134 bachelor’s students Communication Science, 76.1% females, with a mean age of 20.8 years old (SD = 1.70). Of the sample 85.8% was heterosexual; 6.5% preferred the opposite sex, but fancied also sometimes a person of the same sex; 4.5% was homosexual; 1.5% preferred their own sex, but fancied also sometimes the opposite sex; 0.8% preferred both men and women as much. About half (47.8%) of the participants was involved in a relationship.
An online experiment with a factorial between-subjects design with four conditions, two control conditions (i.e., one condition without a stimulus and one condition with a stimulus very subtly portraying condom use) and two experimental conditions (i.e., one condition with a stimulus with a conversation about condom use in a narrative context and one condition with a stimulus with a conversation about condom use in a nonnarrative context), was conducted.
Procedure
For the present study, ethical approval was obtained from the Ethical Review Board of the Amsterdam School of Communication Research. Participants had to give their consent before the start of the experiment. After giving their consent, they were automatically assigned to one of the four conditions. In the two experimental conditions participants were asked to watch an entertainment education video clip (clip of either Girls, narrative, or
fill out the questionnaire and in the other condition the participants also had to watch a video clip, a scene of the movie Fifty Shades of Grey showing a sexual scene with a subtle
suggestion of condom use. In the conditions containing a video clip the participants were asked to answer questions about their demographics, followed by the questions for social comparison orientation. After answering these questions they were instructed to watch the video clip from beginning to end and with full attention. After watching the clip, they were asked to answer the questions for the manipulation check. Then, they received the questions related to the dependent variables. Lastly they were asked to answer questions regarding their sexual behavior and depictions of sex in the media. The participants who completed the experiment gained 0.50 research credit.
Stimulus Materials
Both experimental conditions contained an entertainment education video clip of a few minutes. These two video clips contained a conversation about condom use in relation to STD prevention. One of the two videos was a scene of the American television series Girls. In the scene three of the female main characters are talking about using condoms during sexual intercourse. The health message, using condoms, in this case is incorporated in a narrative. In the other video clip, the health message was included in a nonnarrative way: This was a scene of the Dutch television program Spuiten en Slikken, a talk-show format, targeted at young people, about sex and drugs-related topics. In this fragment the program’s host interviewed guests about STDs as a consequence of not using condoms.
In one of the control conditions there was also a short video clip of a few minutes included. This video clip contained a scene of the movie Fifty Shades of Grey portraying sexual intercourse (no showing of genitals) between a man and a woman with subtle suggesting the use of a condom. This clip did not contain any conversation, but can be
considered an entertainment fragment with a narrative. This way, an alternative explanation of an effect of a narrative only, versus narrative entertainment education, could be excluded. Measures
Intention to use condoms. To measure the intention to use condoms the scale of Prati, Mazzoni and Zani (2014) was used with a 7-point scale from extremely unlikely to extremely likely. The scale consisted of two items: “I plan to use a condom during the next time I have sexual intercourse” and “I intend to use a condom during the next time I have sexual intercourse”. These two items were significantly very strongly correlated (r = 0.96, p < 0.001). Therefore, these items were combined in the scale for intention to use condoms (M = 3.92, SD = 2.52). The constructed scale was not normal distributed: The skewness was -0.048 and the kurtosis -1.770. In attempt to create a more normal distribution first a logarithmic transformation and then a square-root transformation was applied. However, this did not improve the distribution and therefore there was decided to keep the initial scale for the analyses.
Willingness to use condoms. Based on Gerrard et al.’s (2008) scale for willingness, a scale to measure the willingness to use condoms was conducted. The scale consisted of four items followed by the question how likely it is that the participant would use a condom during sexual intercourse. The items were: the person you are in a relationship with, the person you are dating (but with whom you are not in relationship), a friend, and someone you do not or barely know. Each of these items contained a response scale from very unlikely (1) to very likely (7). The items loaded on one factor with an explained variance of 58% and Guttman’s lambda of 0.77. These items were combined into one scale for willingness to use condoms (M = 4.83, SD = 1.22).
Attitude towards condom use. To measure attitude towards condom use two
the use of a condom during the next time I have sexual intercourse is…” The first response category ranged from extremely worthless (1) to extremely valuable (7) and the second from extremely unpleasant (1) to extremely pleasant (7). The two items were significantly strongly correlated (r = 0.55, p < 0.001) and were hence combined into the variable attitude towards condom use (M = 3.79, SD = 1.61).
Perceived benefit of condom use. Based on the example of measuring perceived risk (in relation to risk behavior) in the article of Gerrard et al. (2008), a question for measuring perceived benefit (because the present study focused on health behavior) of condom use was conducted: “How likely is it that using a condom during sexual intercourse protects you from a STD?” with a 7-point Likert scale from very unlikely (1) to very likely (7) (M = 6.09, SD = 1.17).
Subjective norm on condom use. As Rivis, Sheeran and Armitage (2006) mentioned, the prototype-willingness model “conceptualizes social norms in terms of what significant others actually do (i.e., descriptive norm) rather than what significant others think the person ought to do (i.e., injunctive norm)” (p. 485). Based on this, subjective norm on condom use was measured by asking the respondents how likely it is that their peers use a condom during sexual intercourse. The response category ranged from very unlikely (1) to very likely (7) (M = 4.69, SD = 1.47).
Favorability of the prototype that uses condoms. To measure the favorability of the prototype that uses a condom during sexual intercourse, the twelve-item scale of Gibbons and Gerrard (1995) was used. The scale consisted of twelve adjectives: smart, confused, popular, immature, cool, confident, independent, careless, unattractive, dull, considerate and self-centered, each item with a 7-point scale that ranged from not at all (1) to extremely (7). The twelve items, with the negative adjectives recoded, resulted in an unreliable scale. Therefore, two separate scales were created: one with the positive adjectives (smart, popular, cool,
self-confident, independent, considerate) and one with the (unrecoded) negative adjectives
(confused, immature, careless, unattractive, dull, self-centered). The positive items loaded on two components with 45% of the explained variance on the first component. The scale was reasonable (Cronbach’s α = 0.74) and therefore these items were combined into one scale for prototype favorability (M = 4.83, SD = 0.85). The negative items loaded on two components with 50% of the explained variance on the first component. The reliability of the scale was very reasonable and therefore these items were also combined into one scale, prototype unfavorability (M = 2.55, SD = 1.00).
Similarity to prototype that uses condoms. The participant’s similarity to the prototype that uses condoms was measured with the question: “In general, how similar are you to the type of young adult that uses a condom during sexual intercourse?” with the response categories ranging from not at all (1) to very (7) (M = 4.72, SD = 1.50).
Moderator. According to Gibbons and Gerrard (1995) the effect of prototype favorability on willingness is moderated by social comparison tendencies. To measure these tendencies the scale, constructed by Gibbons and Buunk (1999), was used. The scale
consisted of eleven statements, such as “I often compare how I am doing socially with other people” and “I always like to know what others in a similar situation would do”, each item with a 7-point response scale from strongly disagree (1) to strongly agree (7). Because the eleven items did not result in a reliable scale, two items were deleted. These items were deleted, because they had factor loadings below 0.4 (respectively 0.18 and 0.38). The nine remained items loaded on two components with 50% of the explained variance on the first component. The scale was reliable (Cronbach’s α = 0.87) and therefore the items were combined to measure social comparison tendencies (M = 4.75, SD = 0.92).
Demographics. The participants were asked about their gender, age, education (study and bachelor’s or master’s phase), their sexual preference and their relationship status.
Covariates. Besides some demographics, some measures were included as well that might covariate with the other measures. The respondents were asked about the type of partner they had sexual intercourse with in the last six months in general. Response options were: the person I am / was in a relation with, the person I am / was dating, a friend, a person did not or barely knew, I rather do not tell and I have not have sexual intercourse in the last six months. Also, the participants were asked how often they had used a condom of all the times they had sexual intercourse, with a scale from never (1) to always (5) (M = 2.63, SD = 1.28). Respondents who never had sexual intercourse were asked to skip this question.
Based on skipping this question and answering that they had not had sexual
intercourse in the last six months on the question with what type partner they had had sexual intercourse, a variable was created for whether the respondent ever had sex, with no (0) and yes (1). Lastly, two questions about condom use in the media were included, both with a scale from never (1) to often (5). One question posited how often condom use was portrayed when coming across sexual intercourse in the media (M = 2.16, SD = 0.88) and the other question posited how often condom use was mentioned when coming across a conversation about sexual intercourse in the media (M = 2.34, SD = 0.82).
Manipulation check. To check for differences in the perception of the different stimulus materials, three questions were asked. First, to what extent the respondent thought the video clip was educational, not at all (1) to very (7). The second question posited whether they thought the video clip was intended more to entertain (1) or to persuade (7). Lastly, they were asked to what extent the emphasis was on condom use in the video clip, not at all (1) to very (7).
Results Manipulation check and covariate checks
To check whether the stimuli were perceived as intended, it was analyzed to what extent the stimuli were experienced as educative, whether the stimuli were more intended to entertain or to persuade and to what extent the emphasis was on condom use in the stimuli. Cross-tabulation showed that the respondents thought there was more emphasis on condom use in the stimuli of the experimental conditions (i.e., with conversations about condom use) (M = 4.14, SD = 1.25), compared to the stimulus of the control condition, with only subtle suggestion of condom use (M = 2.06, SD = 1.48; χ2 = 57.00, p < 0.001).
Between the narrative stimulus (M = 4.09, SD = 1.33) and the nonnarrative stimulus (M = 4.19, SD = 1.18), both of the experimental group, there was no significant difference in emphasis on condom use (χ2
= 5.59, p = 0.349). However, the nonnarrative stimulus (M = 5.25, SD = 1.08) was evaluated as more educative than the narrative stimulus (M = 2.72, SD = 1.30; χ2
= 36.20, p < 0.001). This stimulus was also perceived as intending more to persuade (M = 4.81, SD = 1.31) and the narrative stimulus as more to entertain (M = 3.22, SD = 1.60; χ2
Experimental stimuli Control stimulus
Measure M SD M SD χ2 p
Emphasis on condom use 4.14 1.25 2.06 1.48 50.41 < 0.001 Extent of educativeness 3.98 1.74 2.23 1.26 23.63 0.001
Entertain - Persuade 4.02 1.66 2.45 1.18 20.22 0.003
Narrative stimulus Nonnarrative stimulus
Emphasis on condom use 4.09 1.33 4.19 1.18 5.59 0.349 Extent of educativeness 2.72 1.30 5.25 1.08 36.20 < 0.001
Entertain - Persuade 3.22 1.60 4.81 1.31 21.16 0.002
Table 1
= 21.16, p = 0.002). In comparison, the mean score for the control condition with the stimulus for this scale indicated that the respondents perceived this stimulus (M = 2.45, SD = 1.18) as more entertaining than the stimuli of the experimental group (M = 4.02, SD = 1.66; χ2 = 20.22, p = 0.002). Mean scores and significance levels are shown in Table 1.
Covariate checks using a MANOVA showed that having a relationship was
significantly related to the two dependent variables intention to use condoms and willingness to use condoms, F (2, 128) = 107.74, p < 0.001. However, the univariate analyses with relationship on these two variables revealed that having a relationship only was significantly related to intention (r = 0.65, p < 0.001) and not to willingness (p = 0.171). Because the multivariate analysis was significant and having a relationship was strongly correlated with intention, this variable was included as a covariate in the further analyses. The type of partner that the respondent had sexual intercourse with in the last six months was also included as a covariate. The type of partner was significantly related to willingness and intention, F (2, 128) = 36.70, p < 0.001. Two one-way ANOVAs also showed that the type of partner was
significantly related to willingness (F = 4.21, p = 0.042) and to intention (F = 73.95, p < 0.001).
A third variable that was included as a covariate was past condom use, which was also significantly related to willingness and intention, F (2,128) = 4.09, p = 0.019. The two one-way ANOVAs revealed that past condom use also significantly correlated with willingness (r = 0.20, p = 0.038) and intention (r = 0.25, p = 0.015). Lastly, a MANOVA showed that whether the respondent ever had sexual intercourse was significantly related to willingness and intention, F (2, 128) = 9.35, p < 0.001, and also for the two one-way ANOVAs with willingness (r = 0.25, p = 0.002) and intention (r = 0.29, p < 0.001). These four variables were thus included in the further analyses as covariates.
Effects of Exposure to Entertainment Education on Intention and Willingness to Use Condoms
For the following analyses the two conditions with the entertainment education clip containing a conversation about condom use (narrative and nonnarrative) were taken together as one experimental group (n = 64). The two control conditions, one with no stimulus and the other with a stimulus subtly suggesting condom use, were taken together as well (n = 70).
The first hypothesis, stating that exposure to entertainment education, containing a conversation about condom use, would positively influence the intention and willingness to use condoms, was analyzed with a MANOVA. The MANOVA showed a significant result, F (2, 127) = 3.75, p = 0.026. A one-way ANOVA showed no significant effect on willingness to use condoms (p = 0.599). Because willingness is a hierarchical scale, it was a possibility that there were some differences to be found when the items were taken separately. However, a MANOVA with the four items of the willingness scale as dependent variables had no significant effect (p = 0.970). A one-way ANOVA did show a significant effect on intention to use condoms, F (1, 128) = 7.14, p = 0.009. The intention to use condoms was higher for the experimental group, with an entertainment education clip containing a conversation about condom use, than for the control group, with no stimulus or a clip with a subtle suggestion of condom use (Mdifference = 0.74). So exposure to entertainment education did positively
influence intention to use condoms (Hypothesis 1a), but did not positively influence willingness to use condoms (Hypothesis 1b). Hypothesis 1a is therefore supported and Hypothesis 1b rejected.
Mediation of Attitude, Perceived Benefit, Subjective Norm, Prototype Favorability, and Prototype Similarity
To test Hypothesis 2a six one-way ANOVAs, with respectively attitude towards condom use, perceived benefit of condom use, subjective norm on condom use, favorability
of the prototype that uses condoms, unfavorability of the prototype that uses condoms, and similarity to the prototype that uses a condom as dependent variables, were executed. Neither attitude (p = 0.599), nor perceived benefit (p = 0.722) showed significant differences between the experimental group and the control group. There was a significant effect of condition on subjective norms, F (1, 128) = 4.07, p = 0.046. However, the effect differentiated from what was expected, because the control group (M = 4.93, SD = 1.42) scored significantly higher than the experimental group (M = 4.42, SD = 1.50). This effect was further analyzed by a one-way ANOVA with the independent variable condition as the initial four different conditions. This analysis revealed that the control condition with the stimulus (M = 5.19, SD = 1.35) significantly differed from the condition with the narrative entertainment education clip (M = 4.41, SD = 1.43, Mdifference = -0.810, p = 0.027) and also from the condition with the nonnarrative entertainment education clip (M = 4.44, SD = 1.59, Mdifference = -0.876, p = 0.020). Exposure to the video clip portraying sexual intercourse with a subtle suggestion of condom use thus positively influenced the extent to which the respondents thought how likely it was that their friends use condoms, b = -0.51, t = -2.02, p = 0.046.
Experimental group Control group
Measure M SD M SD F (1, 128) p Intention 4.32 2.50 3.55 2.50 7.14 0.009 Willingness 4.96 1.03 4.72 1.36 0.28 0.599 Attitude 3.88 1.57 3.70 1.65 0.28 0.599 Perceived benefit 6.02 1.03 6.16 1.28 0.13 0.722 Subjective norm 4.42 1.50 4.93 1.41 4.07 0.046 Prototype favorability 4.73 0.89 4.91 0.80 1.97 0.163 Prototype unfavorability 2.69 1.00 2.42 1.00 2.86 0.093 Prototype similarity 4.70 1.56 4.74 1.45 0.49 0.485 Table 2
The favorability of the prototype that uses condoms (p = 0.163), the unfavorability of the prototype that uses condoms (p = 0.093) and similarity to the prototype that uses condoms (p = 0.750) revealed no significant effects of exposure to entertainment education. Because exposure to entertainment education did not positively influence attitude towards condom use, perceived benefit of condom use, subjective norm on condom use, the favorability and
unfavorability of the prototype that uses condoms and the similarity to the prototype that uses condoms, Hypothesis 2a was rejected. Effects corresponding with Hypothesis 1a, Hypothesis 1b and Hypothesis 2a are shown in Table 2.
Hypothesis 2b posited that attitude towards condoms use, perceived benefit of condom use, subjective norm on condom use, favorability of the prototype that uses condoms, and similarity to the prototype that uses condoms would mediate the relationship between exposure to entertainment education and intention and willingness to use condoms. Because there was no effect found for exposure to entertainment education on willingness to use condoms, the mediators were only tested for relation between exposure to entertainment education and intention to use condoms. Each mediation effect was tested with PROCESS (model 4). The effect of exposure to entertainment education on intention to use condoms was mediated by attitude towards condom use, b = 0.66, t = 2.81, p = 0.006. Perceived benefit of condom use also mediated this relation, b = 0.75, t = 2.70, p = 0.008. Exposure to
entertainment education also significantly predicted intention to use condoms with subjective norm on condom use, b = 0.76, t = 2.68, p = 0.008.
For both prototype favorability and prototype unfavorability there were found
mediation effects of exposure to entertainment education on intention, b = 0.75, t = 2.68, p = 0.008 and b = 0.70, t = 2.50, p = 0.014. Also prototype similarity was tested as mediator and resulted in a significant effect, b = 0.78, t = 2.86, p = 0.005. Thus, Hypothesis 2b was partially supported, because attitude towards condom use, perceived benefit of condom use, subjective
norm on condom use, favorability of the prototype that uses condoms, and similarity to the prototype that uses condoms did mediate the effect of exposure to entertainment education on intention to use condoms, but did not mediate the effect of exposure of entertainment
education on willingness to use condoms.
To test Hypothesis 2c, stating that social comparison tendencies moderate the effect of prototype favorability on willingness, a PROCESS analysis (model 1) was conducted. First, the effect was tested for prototype favorability and second for prototype unfavorability. The test with prototype favorability did not result in a significant effect of favorability of the prototype that uses condoms on willingness to use condoms (p = 0.244) and also not in a significant effect with social comparison tendencies as moderator (p = 0.438). Prototype unfavorability did have a significant effect on willingness to use condoms, b = 0.91, t = 2.00, p = 0.048, and this effect was also significant when moderated by social comparison
tendencies, b = -0.22, 95% CI [-0.40, -0.03], t = 2.29, p = 0.024. For respondents with high social comparison tendencies there is a significant negative relationship between prototype unfavorability and willingness to use condoms, b = -0.032, 95% CI [-0.58, -0.07], t = -2.51, p = 0.013. Thus, for those with high social comparison tendencies, as the prototype that uses condoms becomes less unfavorable, the willingness to use condoms increases and therefore this supports Hypothesis 2c.
Differences between Narrative Entertainment Education and Nonnarrative Entertainment Education
To examine whether there were different effects between exposure to entertainment education with a narrative and entertainment education without a narrative, the two
experimental conditions (n = 32; n = 32) were analyzed. Hypothesis 3 posited that
entertainment education with a narrative would have a more positive influence on all factors than entertainment education without a narrative. To test this hypothesis firstly a MANOVA
with intention and willingness was conducted. The analysis showed a nonsignificant effect (p = 0.980) and also both intention to use condoms (p = 0.867) and willingness to use condoms (p = 0.884) separately revealed no significant differences.
Secondly, effects of exposure to entertainment education on the mediators were analyzed. No significant differences between the narrative and the nonnarrative condition were found for attitude towards condom use (p = 0.802), perceived benefit of condom use (p = 0.768), subjective norm on condom use (p = 0.688), favorability of the prototype that uses condoms (p = 0.241), unfavorability of the prototype that uses condoms (p = 0.543) and similarity to the prototype that uses condoms (p = 0.613). Because there were no significant differences between the narrative and the nonnarrative group on all these factors, Hypothesis 3 therefore was rejected. Effects corresponding with Hypothesis are shown in Table 3.
Narrative condition Nonnarrative condition
Measure M SD M SD F (1, 58) p Intention 3.94 2.58 4.70 2.40 0.03 0.867 Willingness 4.85 1.04 5.06 1.04 0.02 0.884 Attitude 3.70 1.68 4.06 1.46 0.06 0.802 Perceived benefit 6.03 0.93 6.00 1.14 0.09 0.768 Subjective norm 4.41 1.43 4.44 1.59 0.16 0.688 Prototype favorability 4.80 0.76 4.67 1.02 1.40 0.241 Prototype unfavorability 2.66 1.01 2.73 1.00 0.37 0.543 Prototype similarity 4.66 1.58 4.75 1.57 0.26 0.613 Table 3
Discussion
Previous studies have shown that entertainment education can be an effective way of positively influencing behavior’s antecedents and actual behavior (e.g., Shen & Han, 2014) and is suggested to be especially powerful in relation to sexual health behavior (Brown & Walsh-Childers, 2002). A relevant model to investigate behavioral predictors is the prototype-willingness model, as this model incorporates two strong predictors of behavior: intention and willingness (Gerrard et al., 2008). However, research on media effects related to the
prototype-willingness model is limited and entertainment education has never been investigated in relation to this model.
The present study therefore investigated, in the first place, whether exposure to entertainment education could influence emerging adults’ intention and willingness to use condoms. Secondly, following the assumptions of the prototype-willingness model, it was investigated whether the relation between exposure to entertainment education and intention and willingness was mediated by attitude towards condom use, perceived benefit of condom use, subjective norm on condom use, favorability of the prototype that uses condoms and similarity to the prototype that uses condoms. Furthermore, it was examined whether social comparison tendencies moderated the relationship between favorability of the prototype that uses condoms and willingness to use condoms. As entertainment education has only been investigated in terms of a health message incorporated in entertainment content with a narrative, but never without a narrative, the third aim of the present study was therefore to examine differences in the condom use related outcomes between narrative and nonnarrative entertainment education.
Exposure to entertainment education, containing a conversation about condom use, directly influenced emerging adults’ intention to use condoms. This influence was mediated by attitude towards condom use, perceived benefit of condom use, subjective norm on
condom use, favorability of the prototype that uses condoms and similarity to the prototype that uses condoms. However, willingness to use condoms was not influenced by exposure to entertainment education and exposure to entertainment education also did not directly affect attitude towards condom use, perceived benefit of condom use, subjective norm on condom use, prototype favorability and prototype similarity. Though, the favorability of a prototype that uses condoms predicted the willingness to use condoms and this relationship was moderated by social comparison tendencies. Moreover, exposure to narrative entertainment education did not result in more positive effects on the condom use related outcomes than nonnarrative entertainment education.
Effects of Exposure to Entertainment Education on Condom Use Related Outcomes In line with studies that found positive effects of exposure to entertainment education on different outcomes (Bae, 2008; Bae & Kang, 2008; Hether et al., 2008; Khalil &
Rintamaki, 2013; Valente et al., 2007; Vaughan et al., 2012) and specifically positive effects of condom use related entertainment education (Collins et al., 2003; Farrar, 2006; Moyer-Gusé & Nabi, 2010), the present study found that exposure to entertainment education, containing a conversation about condom use, positively influenced the intention to use
condoms. In contrast to the study of Farrar (2006) it was found that exposure to entertainment education did influence intention to use condoms, but did not influence the attitude towards condom use. These contradicting findings might be caused by differences in stimulus
materials. Whereas Farrar (2006) exposed the respondents to a video clip containing portrayal of sexual intercourse with depictions of condom use, the present study exposed respondents in the experimental conditions to a video clip containing a conversation about condom use. Because intention is based on motivations (Pomery et al., 2009), discussing a health-related topic might have a bigger influence on intention than only portrayal of the behavior. For future research it might be interesting to examine effects of both ways of targeting health
behavior in entertainment education. Further, it was found that, in line with the assumptions of the prototype-willingness model (Gerrard et al., 2008), attitude towards condom use, perceived benefit of condom use, subjective norm on condom use, favorability of the
prototype that uses condoms, and similarity to the prototype that uses condoms mediated the effect of entertainment education on intention to use condoms.
Although research has shown that media exposure can affect one’s willingness to engage in risk behavior (Gibbons et al., 2010), and it has been suggested that willingness would be a relevant variable to include in health behavior related research (Blanton et al., 2001; Gerrard et al., 2008; Rivis et al., 2006), the present study did not find an effect of entertainment education on willingness to use condoms. One possibility is that the
conversation about condom use might have affected the reasoned process of decision-making and therefore influenced intention, but did not affect willingness, because this is a more image-based process (Gerrard et al., 2008).
The Influence of Prototype Favorability and Prototype Similarity
Based on the assumption that exposure to media content can contribute in the
construction of prototypes (Gerrard et al., 2008), evidence that media exposure can influence prototypes (Gibbons et al., 2010), and the concept of priming (Bargh et al. 1996), it was expected that the health message regarding condoms use in the entertainment education clip would positively influence the favorability of the prototype that uses condoms. However, the present study did not find effects of exposure to entertainment education on prototype
favorability.
An explanation for the absence of an effect might be that respondents were exposed just once and for a short time period to the entertainment education clip, while the effect of Gibbons et al. (2010) was over a longer period of time. This study thus suggests that there is no direct priming effect of media exposure on the favorability of a prototype. As Gerrard et al.
(2008) indicated, when certain types of behavior appear more often in the media, they become more normative and the favorability of the prototype engaging in that behavior will probably increase. Further research should investigate whether frequent exposure to health behavior in media content can also affect the favorability of prototypes engaging in this health behavior.
Another possible reason that might have contributed to the absence of a significant effect can be sought in the concept of involvement with characters. As Moyer-Gusé (2008) posits, involvement with characters might be a factor contributing in entertainment
education’s effectiveness. Because emerging adults are most likely to identify themselves with peers in the same age category and will most likely be affected by prototypes of a typical emerging adult engaging in a certain behavior (Gerrard et al., 2008), the stimulus materials included characters in the same age category. However, other aspects, such as gender or disliking, might have caused that the respondents could not identify with the characters. The favorability of and similarity to the character might influence the prototype engaging in the character’s behavior. Though, this is just speculation and the role of involvement with characters and other possible moderators of the relation between media exposure and prototype favorability should be investigated.
Even though exposure to entertainment education did not influence the favorability of the prototype that uses condoms, an influence of favorability of the prototype willingness to use condoms was found, in line with the prototype-willingness model. This suggests that the favorability of a prototype engaging in health behavior is correlated with the willingness to engage in that behavior. Furthermore, and also in line with previous studies (Gibbons & Gerrard, 1995; Ouellette et al. 2005), social comparison tendencies moderated the effect of prototype favorability on willingness. This therefore suggests that the extent to which one compares his or her self to others influences the relation between prototype favorability and
willingness and should therefore be included as moderator in future research on the prototype-willingness model.
As for prototype similarity, it was found that this factor did mediate the effect of exposure to entertainment education on intention to use condoms, but not the effect on willingness to use condoms. This is in the line with a meta-analysis of studies on the
prototype-willingness model that showed that prototype similarity was a stronger predictor of intention than for willingness (Todd et al., 2014). However an underlying mechanism for this effect has never been investigated and should therefore be examined to make predictions. Narrative versus Nonnarrative Entertainment Education
In contrast with the expectation, the findings did not show differences in effects of narrative and nonnarrative entertainment education for any of the outcome variables.
Entertainment education, as researched in previous studies, is assumed to be effective because of the idea that a narrative overcomes the arousal of reactance and the health message
therefore would affect the viewer (Moyer-Gusé, 2008). However, the present study suggests that incorporating a health message in entertainment content with a narrative is not more effective than a health message incorporated in entertainment content without a narrative. An explanation for the absence of a difference might be found in the perception of the extent of persuasiveness of the message. As Moyer-Gusé and Nabi (2010) found that reactance was aroused when respondents perceived the entertainment education with a narrative as intended to persuade and therefore had not the positive effect that respondents had that did not perceive the message as intended to persuade. However, respondents in the present study perceived the entertainment education clip with a narrative as more entertaining than persuasive, so
reactance was probably not aroused and therefore not influenced the outcomes. To gain more insights in differences between entertainment education with a narrative en without a
Limitations
The present study has some limitations. First of all, the study focused on condom use as prevention for STDs, but not on condom use as contraception. For women who do not let their male sexual partner use a condom (because they know he does not have a STD), but do use another form of contraception, does not mean that they have unsafe sex. People with a regular sex partner might therefore score low on intention to use condoms. This was an implication for this research: The variable intention was very skewed and was still skewed after a log and square root transformation. The skewness, however, can be explained by the fact that 47.8% was involved in a relationship and therefore a great part of the sample scored low on intention to use condoms. However, relationship status was controlled for in the analyses. For future research on condom use respondents might be selected on the notion that they are not in a relationship.
In conclusion, the present study provides insights in effects of entertainment education in relation to the prototype-willingness model and differences between narrative and
nonnarrative entertainment education. This study suggests that single exposure to entertainment education does directly positively affect emerging adults’ intention to use condoms, but does not influence willingness to use condoms and its predictor favorability of a prototype that uses condoms. However, longitudinal research is needed to understand the relation between exposure to entertainment education and prototypes and subsequently willingness. Also, present research indicates that incorporating a health message in entertainment content with a narrative is not more effective than a health message
incorporated in entertainment content without a narrative. Present findings might be relevant for future research in investigating how entertainment education can be effective in increasing condom use among emerging adults.
References
Arnett, J. J. (2014). Emerging adulthood: The winding road from the late teens through the twenties. New York: Oxford University Press.
Bae, H. S. (2008). Entertainment-education and recruitment of cornea donors: The role of emotion and issue involvement. Journal of Health Communication, 13(1), 20-36. doi: 10.1080/10810730701806953
Bae, H. S., & Kang, S. (2008). The influence of viewing an entertainment–education program on cornea donation intention: A test of the theory of planned behavior. Health
Communication, 23(1), 87-95. doi: 10.1080/10410230701808038
Bargh, J. A., Chen, M., & Burrows, L. (1996). Automaticity of social behavior: Direct effects of trait construct and stereotype activation on action. Journal of Personality and Social Psychology, 71(2), 230-244.
Blanton, H., VandenEijnden, R. J., Buunk, B. P., Gibbons, F. X., Gerrard, M., & Bakker, A. (2001). Accentuate the negative: Social images in the prediction and promotion of condom use. Journal of Applied Social Psychology, 31(2), 274-295. doi:
10.1111/j.1559-1816.2001.tb00197.x
Brodie, M., Foehr, U., Rideout, V., Baer, N., Miller, C., Flournoy, R., & Altman, D. (2001). Communicating health information through the entertainment media. Health
Affairs, 20(1), 192-199. doi: 10.1377/hlthaff.20.1.192
Brown, J. D., & Walsh-Childers, K. (2002). Effects of media on personal and public health. In J. Bryant & D. Zillman (Eds.), Media effects: Advances in theory and research (pp. 453-488). New Jersey: Lawrence Erlbaum Associates.
Brown, J. D. (2002). Mass media influences on sexuality. Journal of Sex Research, 39(1), 42-45. doi: 10.1080/00224490209552118
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(5), 1115-1121. doi:
10.1542/peds.112.5.1115
De Graaf, H., Kruijer, H., Van Acker, J., & Meijer, S. (2012). Seks onder je 25e: Seksuele gezondheid van jongeren in Nederland anno 2012. Delft: Eburon.
Farrar, K. M. (2006). Sexual intercourse on television: Do safe sex messages matter? Journal of Broadcasting & Electronic Media, 50(4), 635-650. doi:
10.1207/s15506878jobem5004_4
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Boston: Addison-Wesley.
Gerrard, M., Gibbons, F. X., Houlihan, A. E., Stock, M. L., & Pomery, E. A. (2008). A dual-process approach to health risk decision making: The prototype willingness
model. Developmental Review, 28(1), 29-61. doi: 10.1016/j.dr.2007.10.001 Gibbons, F. X., & Buunk, B. P. (1999). Individual differences in social comparison:
Development of a scale of social comparison orientation. Journal of Personality and Social Psychology, 76(1), 129. doi: 10.1037/0022-3514.76.1.129
Gibbons, F. X., & Gerrard, M. (1995). Predicting young adults' health risk behavior. Journal of Personality and Social Psychology, 69(3), 505-517. doi:
10.1037/0022-3514.69.3.505
Gibbons, F. X., Pomery, E. A., Gerrard, M., Sargent, J. D., Weng, C. Y., Wills, T. A., ... & Yeh, H. C. (2010). Media as social influence: Racial differences in the effects of peers and media on adolescent alcohol cognitions and consumption. Psychology of Addictive Behaviors, 24(4), 649-659. doi: 10.1037/a0020768
Goenee, M., Kedde, H., & Picavet, C. (2012). Condoomgebruik en testgedrag in Nederland. Tijdschrift voor Seksuologie, 36(2), 109-120.
Greenberg, B. S., Salmon, C. T., Patel, D., Beck, V., & Cole, G. (2004). Evolution of an EE research agenda. In A. Singhal, M. J. Cody, E. M. Rogers, & M. Sabido (Eds.), Entertainment-education and social change: History, research, and practice (pp. 191-206). New Jersey: Lawrence Erlbaum Associates.
Hether, H. J., Huang, G. C., Beck, V., Murphy, S. T., & Valente, T. W. (2008).
Entertainment-education in a media-saturated environment: Examining the impact of single and multiple exposures to breast cancer storylines on two popular medical dramas. Journal of Health Communication, 13(8), 808-823. doi:
10.1080/10810730802487471
Kennedy, M. G., O'Leary, A., Beck, V., Pollard, K., & Simpson, P. (2004). Increases in calls to the CDC national STD and AIDS hotline following AIDS‐ related episodes in a soap opera. Journal of Communication, 54(2), 287-301. doi: 10.1111/j.1460-2466.2004.tb02629.x
Khalil, G. E., & Rintamaki, L. S. (2014). A televised entertainment-education drama to promote positive discussion about organ donation. Health Education Research, 29(2), 284-296. doi: 10.1093/her/cyt106
Moyer‐ Gusé, E. (2008). Toward a theory of entertainment persuasion: Explaining the persuasive effects of entertainment‐ education messages. Communication Theory, 18(3), 407-425. doi: 10.1111/j.1468-2885.2008.00328.x
Moyer‐ Gusé, E., & Nabi, R. L. (2010). Explaining the effects of narrative in an
entertainment television program: Overcoming resistance to persuasion. Human Communication Research, 36(1), 26-52. doi: 10.1111/j.1468-2958.2009.01367.x
O’Hara, R. E., Gibbons, F. X., Gerrard, M., Li, Z., & Sargent, J. D. (2012). Greater exposure to sexual content in popular movies predicts earlier sexual debut and increased sexual risk taking. Psychological Science, 23(9), 984-993. doi: 10.1177/0956797611435529 Ouellette, J. A., Hessling, R., Gibbons, F. X., Reis-Bergan, M., & Gerrard, M. (2005). Using images to increase exercise behavior: Prototypes versus possible selves. Personality and Social Psychology Bulletin, 31(5), 610-620. doi: 10.1177/0146167204271589 Peter, J., & Valkenburg, P. M. (2011). The influence of sexually explicit internet material on
sexual risk behavior: A comparison of adolescents and adults. Journal of Health Communication, 16(7), 750-765. doi: 10.1080/10810730.2011.551996
Pomery, E. A., Gibbons, F. X., Reis-Bergan, M., & Gerrard, M. (2009). From willingness to intention: Experience moderates the shift from reactive to reasoned
behavior. Personality and Social Psychology Bulletin, 35(7), 894-908. doi: 10.1177/0146167209335166
Prati, G., Mazzoni, D., & Zani, B. (2014). Perceived behavioural control, subjective norms, attitudes and intention to use condom: A longitudinal cross-lagged design. Psychology & Health, 29(10), 1119-1136. doi: 10.1080/08870446.2014.913043
Rivis, A., Sheeran, P., & Armitage, C. J. (2006). Augmenting the theory of planned behaviour with the prototype/willingness model: Predictive validity of actor versus abstainer prototypes for adolescents' health‐ protective and health‐ risk intentions. British Journal of Health Psychology, 11(3), 483-500. doi: :10.1348/135910705X70327 Roberts, M. E., Gibbons, F. X., Kingsbury, J. H., & Gerrard, M. (2014). Not intending but
somewhat willing: The influence of visual primes on risky sex decisions. British Journal of Health Psychology, 19(3), 553-565. doi: 10.1111/bjhp.12055
Sheeran, P., & Orbell, S. (1998). Do intentions predict condom use? Meta-analysis and examination of six moderator variables. British Journal of Social Psychology, 37, 231-252.
Shen, F., & Han, J. (2014). Effectiveness of entertainment education in communicating health information: A systematic review. Asian Journal of Communication, 24(6), 605-616. doi: 10.1080/01292986.2014.927895
Todd, J., Kothe, E., Mullan, B., & Monds, L. (2014). Reasoned versus reactive prediction of behaviour: A meta-analysis of the prototype willingness model. Health Psychology Review, 1-24. doi: 10.1080/17437199.2014.922895
Træen, B., Hald, G. M., Noor, S. W., Iantaffi, A., Grey, J., & Rosser, B. S. (2014). The relationship between use of sexually explicit media and sexual risk behavior in men who have sex with men: Exploring the mediating effects of sexual self-esteem and condom use self-efficacy. International Journal of Sexual Health, 26(1), 13-24. doi: 10.1080/19317611.2013.823900
Valente, T. W., Murphy, S., Huang, G., Gusek, J., Greene, J., & Beck, V. (2007). Evaluating a minor storyline on ER about teen obesity, hypertension, and 5 A Day. Journal of Health Communication, 12(6), 551-566. doi: 10.1080/10810730701508385
Vaughan, P. W., Rogers, E. M., Singhal, A., & Swalehe, R. M. (2000). Entertainment-education and HIV/AIDS prevention: A field experiment in Tanzania. Journal of Health Communication, 5, 81-100. doi: 10.1080/10810730050019573
Wingood, G. M., DiClemente, R. J., Harrington, K., Davies, S., Hook, E. W., & Oh, M. K. (2001). Exposure to X-rated movies and adolescents' sexual and contraceptive-related attitudes and behaviors. Pediatrics, 107(5), 1116-1119. doi: 10.1542/peds.107.5.1116
