Graduate School of Communication
Master’s programme Communication Science
MASTER’S THESIS
Anti-health promotion at the example of pro-anorexia
Effects of identification and resistance-increasing strategies on
self-esteem and dieting intentions
Written by Natalie Ojvan
11106581
Supervisor: Simon Zebregs 03.02.2017
Abstract
This study investigated anti-health promotion at the example of the pro-anorexia community. Specifically this research concentrated on the effects of identification with pro-anorexic sources on self-esteem and dieting intentions in the context of Instagram. Resistance-increasing strategies aimed to reduce persuasiveness and thus counter the effects of identification. 312 female students participated in an online experiment. The participants viewed three Instagram profiles including newly created anorexic profile. The pro-anorexic profile incorporated a high and low identification manipulation. Half of the
participants received a resistance-increasing treatment based on inoculation theory: The first part informed the participants about the manipulative intent of pro-anorexic profiles; the second part gave an educational message on how to identify pro-anorexic profiles. As expected, identification significantly decreased self-esteem and increased dieting intentions. However, the treatment was not successful. The resistance-increasing treatment reduced perceived persuasiveness but did not influence the effects of identification on self-esteem nor on dieting intentions. Even though participants perceived the pro-anorexic message less persuasive, the effects of identification on self-esteem and dieting intentions were persistent. This study was the first to present identification as a specific influence factor in anti-health promotion. Future research should aim at finding measures that successfully counter anti-health promotion and investigate identification processes further.
Key words: anti-health promotion, pro-anorexia, identification, social media, self-esteem, dieting, persuasiveness, resistance-increasing strategies
Anti-Health Promotion at the example of Pro-Anorexia
Effects of Identification and Resistance-Increasing Strategies on Self-Esteem and Dieting Intentions
“[Pro-anorexia] websites offer constant social support for extreme weight loss behavior. In that way, pro-[anorexia] websites portray extreme thinness not only as normative, but they associate it with a signal of success as well.” (Custers, 2015, p. 432)
As a result, the pro-anorexia community not only practices unhealthy behavior but also glorifies and promotes it. By means of online communication, the community engages in disordered eating behavior and reaches out to a broad audience to encourage unhealthy weight loss behavior (Yom-Tov & Boyd, 2014). However, until now research mainly focused on the promotion of healthy behaviors. A legacy of literature ranges from prevention measures that focus on avoiding undesirable outcomes to promotional measures that emphasize obtaining a favorable state of health (Rothman, Bartels, Wlaschin, & Salovey, 2006). In fact, health promotion describes “the process of enabling people to increase control over their health and its determinants, and thereby improve their health” (World Health Organization, 2005, p. 1). For the pro-anorexia community, the opposite seems to be the goal. The group aims at promoting behaviors that lead people to impair their health (Harshbarger, Ahlers-Schmidt, Mayans, Mayans, & Hawkins, 2009; Wittchen & Hoyer, 2011).
The study at hand will investigate anti-health promotion at the example of the pro-anorexia community on Instagram. Recent literature suggests a focus of pro-pro-anorexia research on social media (Custers, 2015; Fardouly, Diedrichs, Vartanian, & Halliwell, 2015; Sidani, Shensa, Hoffman, Hanmer, & Primack, 2016). As Instagram focuses on appearance, the pro-anorexia community is widely present and engages in new forms of communication that are not sufficiently examined yet (Custers, 2015; Sidani et al., 2016) Therefore, this study will focus on pro-anorexia promotion on Instagram. First, it is important to understand the effects that attempts of promoting unhealthy behavior might have. Therefore, this study will explore
the effects of pro-anorexia promotion on self-esteem and dieting intentions, which have been shown to be affected immensely (Bardone-Cone & Cass, 2008; Custers, 2015; Custers & Van den Bulck, 2009). Secondly, it is crucial to comprehend which factors influence the
effectiveness of these attempts. The factors identification and message persuasiveness will serve to examine the mechanisms of pro-anorexia promotion further. Lastly, measures to counter anti-health promotion need to be found. By means of a resistance-increasing strategy, this study will test whether reducing message persuasiveness diminishes potential effects.
RQ: How does identification with pro-anorexic sources on Instagram affect self-esteem and dieting intentions, to what extent is this mediated by message persuasiveness and can it be moderated by resistance-increasing strategies?
To date, the active promotion of unhealthy behavior lacks extensive research.
Nevertheless, a number of recent studies deal with specific anti-health groups. Pro-smoking, pro-self-harm, pro-anorexia and excessive fitness represent anti-health groups that were focus of latest literature (Bardone-Cone & Cass, 2008; Boyd, Yan, & Eavitt, 2010; Setodji, Martino, Scharf, & Shadel, 2014; Tiggemann & Zaccardo, 2015).
On social media identification processes play a great role for the effectiveness of content (Schmidt, 2009). According to Bandura’s social cognitive theory (1986), people adapt to behaviors of people they identify with. Especially identification with peers proved to affect users on social media (Fardouly et al., 2015; Fardouly & Vartanian, 2015). As models from anti-health promotion present risk behavior, social media users might adopt unhealthy
behavior. Furthermore, studies revealed that identification with peers on social media affected users’ self-esteem (Fardouly & Vartanian, 2015). In the case of pro-anorexia, studies
indicated that promoted negative self-perception is transferred to others (Bardone-Cone & Cass, 2008; Custers, 2015; Custers & Van den Bulck, 2009).
Moreover, studies found that exposure to pro-anorexic content greatly affects people without eating disorder tendencies (Bardone-Cone & Cass, 2008; Custers & Van den Bulck, 2009). Hence, research on anti-health communities suggests to find measures minimizing effects on uninvolved people (Bardone-Cone & Cass, 2008; Fardouly & Vartanian, 2015; Martijn, Smeets, Jansen, Hoeymans, & Schoemaker, 2009). Resistance-increasing strategies help people to withstand such persuasive attacks (Banas & Rains, 2010). When faced with illegitimate content, these strategies decrease persuasion (Sagarin, Cialdini, Rice, & Serna, 2002). By educating people and enhancing their skills to recognize persuasion, their resistance towards attacks increases and the persuasion fails (Banas & Rains, 2010; Epstein & Botvin, 2008; Lee, 2010; Sagarin et al., 2002). In addition to anti-health promotion, resistance-increasing strategies might serve to reduce negative outcomes.
Theoretical Background Anti-Health Promotion of Pro-Anorexia
Anti-health promotion encourages people to start and keep unhealthy mindsets and behavior patterns, which consequently worsen the quality of life. In contrast, the overall goal of health promotion is to improve the quality of life (Gielen & Mcdonald, 2002). Health promotion aims either at preventing the initiation of behaviors that contribute to a worsened state of health or at intervening in unhealthy behavior patterns (Rothman et al., 2006). In the case of pro-anorexia, the most prominent unhealthy features that pro-anorexics encourage online are self-loathing and extreme dieting behavior (Custers, 2015; Haas, Irr, Jennings, & Wagner, 2011; Harshbarger et al., 2009; Norris, Boydell, Pinhas, & Katzmann, 2006; Rodgers, Skowron, & Chabrol, 2012).
Dieting intentions. Research found that people adapt extreme dieting patterns and
increase their drive for thinness when they are exposed to pro-anorexic content (Bardone-Cone & Cass, 2008; Custers & Van den Bulck, 2009). Bardone-Cone and Cass (2008) showed that participants were more willing to exercise and engage in weight-reducing behaviors after
viewing pro-anorexic websites. Few recent studies stress the importance of examining social media as the content is more extreme and graphic than on blogs (Custers, 2015; Tiggemann & Zaccardo, 2015; Tiggemann & Zaccardo, 2016).
The current study will examine the behavioral consequences that the pro-anorexic community causes via social media in form of dieting intentions. Dieting is defined as “efforts to restrict food intake with the goal of weight loss” (Cruwys et al., 2013, p. 265). The theory of planned behavior by Ajzen (1991) gives reason to believe that intentions are the strongest predictor of following behavior change. Hence, dieting intentions will be the focus of this research and mean a “person’s intentions to diet in the immediate future” (Cruwys et al., 2013, p.275).
Self-esteem. Pro-anorexics display self-loathing openly online and present it as
desirable (Custers, 2015; Haas et al., 2011). In this regard, several studies showed that exposure to pro-anorexic content especially affects self-esteem negatively (Bardone-Cone & Cass, 2008; Custers, 2015; Custers & Van den Bulck, 2009). Bardone-Cone and Cass (2008) revealed that the self-esteem of students significantly decreased after exposure to pro-anorexia weblogs. These effects were not only found for people with eating disorder tendencies, but as well for people without tendencies. Another study by Custers and Van den Bulck (2009) showed similar results amongst teenagers between 13 and 16 years. For the purpose of this study, self-esteem will be investigated and considered as the evaluation of one’s self and self-worth (Jordan et al., 2015).
Investigation of exposure. The body of recent research on pro-anorexia focused on
effects caused by exposure to pro-anorexic content and suggests that self-esteem and dieting intentions are highly affected. However, no further factors that influence the strength of these effects were examined until now. Sidani and colleagues (2016) studied the associations of social media use and eating concerns among young adults. Their findings suggest that there is a high association and that frequent users of social media struggle increasingly with eating concerns.
They mention repeatedly that the pro-anorexic community seems to be a factor in this relationship. According to them, the examination of specific social media characteristics would be useful both contextually and content-related (Sidani et al., 2016). This study will examine the concept of identification as a specific influence factor on self-esteem and dieting intentions in the context of anti-health promotion on social media.
Identification Processes
Bardone-Cone and Cass (2008) tested the effects of exposure to pro-anorexic content but did not test whether specific aspects alter these effects. However, they assumed that identification with pro-anorexics acts as an enhancing factor (Bardone-Cone & Cass, 2008).
The social cognitive theory by Bandura (1988) holds that observing a model can influence subsequent behavior of the observer. When the observer identifies with the model the likeliness of engaging in the performed behavior increases. Thus, behavioral effects will be stronger the more the observer identifies with the model (Bandura, 1995). The study at hand will investigate whether identification with pro-anorexic sources influences self-esteem and dieting intentions. Relying on Bandura’s definition, identification refers to the extent to which an observer perceives one-to-one similarities with the model and is influenced by this perception (Bandura, 1988). Furthermore, the social comparison theory (Festinger, 1954) states that people compare themselves to similar others and evaluate their self based on these comparisons. When someone identifies with a person and this persons fails or succeeds, it can affect self-efficacy, self-esteem and behavior accordingly (Fardouly et al., 2015; Fardouly & Vartanian, 2015; Groothof, Siero, & Buunk, 2007; Manago, Monique Ward, Lemm, Reed, & Seabrook, 2015).
Wilcox and Laird (2000) pointed out the differences of engaging in social comparison and identification in the context of the portrayal of thin women in the media and women’s self-esteem. According to them, the main difference is whether personal or situational cues aid to evaluate unusually thin models. Wilcox and Laird (2000) found that situational cues
activate identification processes. Identification with a slim model led to a temporary good feeling because women enjoyed identifying with the positive portrayal of the thin model. Identification as Wilcox and Laird (2000) propose it implies that identification will lead an observer to feel accordingly to how the person is represented. As pro-anorexics prominently present negative self-evaluations (Haas et al., 2011), identification would suggest negative feelings for the observer. As people are likely to adapt behaviors of models they identify with (Bandura, 1988) identification with pro-anorexics would mean the adaptation of destructive thought and behavior patterns. For the present study, this suggests that identification with a pro-anorexic source influences the receiver’s self-esteem and dieting intentions in line with the content the source promotes.
H1a) People that identify highly with a pro-anorexic source will have lower self-esteem than
people that identify low with a pro-anorexic source.
H1b) People that identify highly with a pro-anorexic source will have higher dieting
intentions than people that identify low with a pro-anorexic source.
Mediating Role of Message Persuasiveness
When people are exposed to anti-health promotion, identification might not solely affect mindsets and behaviors. Identification with a source of anti-health promotion might serve as an enhancing factor of persuading people into unhealthy behavior. In fact,
identification with a source might increase the persuasiveness of an unhealthy message. The effectiveness of a message depends on whether or not it can persuade people (Pounders, Lee, & Mackert, 2015). When promoting health the aim is to present the message in the most convincing way (Kok, Schaalma, Ruiter, van Empelen, & Brug, 2004). Success can be attributed to consequent change in attitudes, intentions and behaviors (Pounders et al., 2015). This study makes use of the definition of message persuasiveness as the perception of
how effective of a message appears to be (Chandran & Menon, 2004; Pounders et al., 2015). Message persuasiveness entails that a message is perceived to be convincing, useful,
interesting and/or informative (Chandran & Menon, 2004; Pounders et al., 2015).
Further, perceived persuasive intent can influence the persuasiveness of a message. If people perceive a message as manipulative, they will easily feel threatened in their freedom and raise their resistance towards the message (Burgoon, Alvaro, Grandpre, & Voulodakis, 2002). The less the message seems as an attempt to persuade, the more persuasive the message will be (van Noort & Voorveld, 2014). Identifying with a source lowers resistance and consequently makes the message more persuasive (Belch & Belch, 2015). Therefore, identification could mask the persuasive intent and increase persuasiveness.
H2a) People that identify highly with the pro-anorexic source will perceive the message as
more persuasive than people that identify low with the pro-anorexic source.
Moreover, message persuasiveness positively affects people’s willingness to perform the promoted behavior (Pounders et al., 2015). The more people perceive something as convincing, the more likely they are to follow up on the convincing message content (Pounders et al., 2015). In consequence, the effects of identification on self-esteem and dieting intentions might be influenced indirectly by how persuasive the message is perceived. Figure A1 presents an overview of all hypotheses.
H2b) People that identify highly with a pro-anorexic source will perceive the content as more
persuasive than people that identify low with the pro-anorexic source and the more a person perceives the pro-anorexic content as persuasive, the lower the self-esteem will be.
H2c) People that identify highly with a pro-anorexic source will perceive the content as more
persuasive than people that identify low with the pro-anorexic source and the more a person perceives the pro-anorexic content as persuasive, the higher the dieting intentions will be.
Resistance-Increasing Strategies
In order to prevent people from initiating unhealthy behavior, research needs to investigate the most effective approach to counter anti-health promotion. First, the source can be addressed (Conner & Norman, 2005). This can be administered by persuading the source that promoting unhealthy behavior is harmful not only to them but to others as well. However, reaching anti-health sources might prove to be difficult in the ever-changing and scattered world of social media where anti-health promotion seems to be most prominent (Holland & Tiggemann, 2016; Santarossa, Coyne, Lisinski, & Woodruff, 2016). Instagram tried to address sources of anti-health promotion by deleting profiles that promote harmful behavior and forbidding self-harm promotion in the guidelines (Instagram, 2016a). However, this attempt failed and did not stop harmful groups from appearing on Instagram and avoiding detection (Sidani et al., 2016).
Secondly, receivers of anti-health messages can be targeted. Health promotion oftentimes relies on interventions that aim at enhancing self-efficacy (Conner & Norman, 2005). Providing people with efficient skills increases their ability to perform a certain health behavior (Linn, van Weert, Smit, Perry, & van Dijk, 2013). Furthermore, it equips people with the necessary tools to resist persuasive attacks (Compton, Jackson, & Dimmock, 2016). The study at hand will focus on receivers of anti-health messages and test strategies that increase their resistance towards anti-health promotion.
Sagarin et al. (2002) investigated how people can be more resistant towards a message with the use of illegitimate and legitimate expert sources in advertisements. The results showed that people are more resistant towards a message when confronted with their own vulnerability. When people found out that they have been manipulated, their perception of the
persuasive intent got activated. In consequence, this led to more resistance and less persuasion (Reijmersdal et al., 2016; Sagarin et al., 2002). Although this strategy was successful in the controlled experimental setting, the feasibility and implementation of a procedure that bases its success on explicitly misleading people seems difficult (Sagarin et al., 2002). In order for a resistance-increasing strategy to be applicable and successful in real life, it requires an
educational message (Gielen & Mcdonald, 2002).
Therefore, this study will focus on the resistance-increasing strategy of inoculation. Developed by McGuire (1961), the inoculation theory explains how attitudes and beliefs change and how original attitudes and beliefs are strengthened when faced with persuasion attempts. In the case of anti-health promotion, the latter will be important. When faced with the promotion of unhealthy behavior people should be enabled to refute the messages and stick to healthy or rather neutral beliefs.
The first element of inoculation is the threat (Banas & Rains, 2010). The threat element of inoculation relates to a person’s freedom. When people feel threatened in their ability to make a free choice, they will react against this threat and defend their beliefs and attitudes (Burgoon et al., 2002). People that take part in anti-health communities like pro-anorexia tend to act against social norms (Boyd et al., 2010; Fleming-May & Miller, 2010; Haas et al., 2011; Moreno, Ton, Selkie, & Evans, 2016). Health is no primary goal, which is contrary to the general public belief (Boyd et al., 2010). Thus, in the case of anti-health messages it might be most important to reinforce attitudes and beliefs that counter unhealthy beliefs.
The refutational preemption represents another key element of the inoculation theory (Banas & Rains, 2010). In order to inoculate a person against persuasion attempts, the person needs to be exposed to a weak attack on the favored position. This triggers a process of counter arguing to defend one’s attitude actively (Banas & Rains, 2010). In plain language,
the refutational preemption trains a person on how to react in a situation of an actual persuasion attack.
Prior research integrated the inoculation theory into the field of health promotion. An early study by Pfau and Van Bockern (1994) established that inoculation effectively prevents teenagers from smoking. The use of inoculation succeeded in promoting resistance towards smoking initiation among adolescents. The effectiveness of inoculation in this age group was due to the fact that the behavior of smoking reaches the initiation point shortly after (Pfau & Van Bockern, 1994). Hence, inoculation worked preventative, kept adolescents from initiating an unhealthy behavior, and strengthened their beliefs in non-smoking to be the desirable behavior. In adaption to anti-health promotion, people that are not yet familiar with the promoted unhealthy behavior might profit from inoculation. Strengthening their attitude towards the healthy behavior they are naturally performing could empower them to stay away from the promoted one. Another early study tried to inoculate children by having them role-play a persuasion attack (Perry, Slinkard, Killen, & McAlister, 1980). In consequence, the children would practice the appropriate behavior in a persuasive situation. Learning to react in a certain manner provides people with the ability to handle future persuasive situations and avoid manipulation (Banas & Rains, 2010). The refutational preemption as the weak attack served as a preparation for an actual persuasive attack and became a self-efficacy tool.
In conclusion, inoculation might help people to resist anti-health messages. When people receive the information that anti-health groups will try to manipulate them (threat) and instructions on how this might appear (weak attack), they might be more likely to resist the unhealthy messages. Consequently, they would be aware of the persuasive attack and perceive the message as less convincing.
H3a) People that receive the resistance-increasing treatment of inoculation will perceive the
Since the resistance will serve as a tool to uncover the persuasive intent of the source,
identification should not have great influence on persuasiveness anymore. Since identification was supposed to increase persuasiveness, the resistance-increasing treatment is expected to act as a counter measure and decrease the effect of identification on persuasiveness.
H3b) People that identify highly with the pro-anorexic source and receive the
resistance-increasing treatment will perceive the messages less persuasive than people that did not receive the treatment and also identify highly with the pro-anorexic source.
Relating this to the conclusions from earlier chapters, the persuasiveness could as well have effects of self-esteem and dieting intentions. In line with expectations of the mediation effect of message persuasiveness on identification and self-esteem and dieting intentions, the resistance-increasing treatment should reduce these effects indirectly. Figure A2 includes these hypothesized changes.
H4a) The resistance-increasing treatment will lower the effect of identification on
persuasiveness and consequently lead to increased self-esteem.
H4b) The resistance-increasing treatment will lower the effect of identification on
persuasiveness and consequently lead to decreased dieting intentions.
Method Research Design
In order to answer the research question at hand, an experiment was conducted and a 2 x 2 factorial design was employed. The factor identification with pro-anorexic source (two levels, namely: high identification and low identification) and the moderator
resistance-increasing strategy (two levels, namely: treatment and no treatment) served as between-subjects variables. Further, it was an independent-measures design so that respondents participated in one condition only.
Procedure. Participants were recruited over a three-week timeframe in December
2016 via social media. The experiment took 5 to 8 minutes depending on whether the treatment was presented.
A cover story was used to avoid answering biases in terms of pro-anorexia.
Participants were introduced to the experiment as a study about Instagram perceptions. After the description, the letter of consent was introduced and the experiment started after
agreement. Half of the participants were presented with the resistance-increasing treatment. Then, three Instagram profiles followed including a pro-anorexic profile being last. Two neutral profiles acted as distractions from the pro-anorexic profile. The first distraction profile displayed fashion lifestyle and the second presented shoes (see Appendix B). After viewing the profiles, participants were asked to answer questions about the persuasiveness of the pro-anorexia profile. Questions concerning self-esteem and dieting intentions followed. Finally, participants filled in the manipulation check and demographical questions and were debriefed about the aim of the study (see Appendix B).
Material
Creating the material emphasis was placed on generating authentic Instagram profiles. This would decrease the feeling of being in an experimental setting for the participants (Bryman, 2012). First, the profile overview was shown. The profile overview gives users an idea about the profile content and the person the profile belongs to (Instagram, 2016b). Secondly, two single postings were presented to the participants. This included the profile name, a picture, a caption describing the picture as well as hashtags describing the content and category of the picture (Instagram, 2016b). Two postings were shown to give more insight into the profile, the person the profile belongs to, the message and the tone.
Pro-anorexic profile. The pro-anorexic profile was newly created on the basis of
content analyses on pro-anorexia in general and on Instagram specifically (Custers, 2015; Haas et al., 2011; Moreno et al., 2016; Tiggemann & Zaccardo, 2016). Furthermore, existing pro-anorexia profiles on Instagram aided in capturing the tone and visuals. The profile
overview contained the name tobethinandbeautiful, a profile picture of a very thin female and a description that focused on weight loss in a community. The description further entailed pro-anorexia specific details and abbreviations indicating the current and goal weights. A selection of pictures of pro-anorexia profiles was included in the overview ranging from self-loathing quotes, weight loss tips, motivational quotes and thinspirations (see Appendix C1). The first posting contained weight loss motivation with thinspiration pictures of thin individuals and the question “hungry?”. This was further described as a motivation not to eat and look like the presented girls (see Appendix C2a). This signifies a common pro-anorexia practice of self-loathing (Haas et al., 2011). The second posting listed several tips on what to do instead of eating and was enforced by a widely used tip of the pro-anorexic community in the description (see Appendix C2b). Likewise, giving dieting tips and providing insider knowledge on the effectiveness of tips presents a characteristic of the community (Csipke & Horne, 2007; Fleming-May & Miller, 2010; Haas et al., 2011).
Identification. In addition, the profile included the identification manipulation.
Identification had two manipulation levels, namely high identification and low identification. High identification was operationalized as “24. Student. Travel is life” in the personal
description of the pro-anorexic profile. In contrast, for low identification the description said “18. Waitress. Work is life”. Since the sample consisted of female students only, it was important to stress aspects they would identify with in the high identification condition and not identify with in the low identification condition. Therefore, an average age of a student (24), the main employment (student) and a typical activity (traveling) was utilized in the high identification condition. A typical manner of writing was used to enhance the authenticity of
the profile (Authenrieth, 2014). All other aspects of the profile were kept constant leaving the identification manipulation as the only difference (see Appendix C2).
Resistance-increasing strategy. The implemented resistance-increasing treatment of
this study was based on the aspects of inoculation. Therefore, two parts were of crucial nature: a threat and a refutational preemption. First, the treatment contained the threat (see Appendix C3). Based on literature that analyzed the motives and content of the pro-anorexia movement (Custers, 2015; Tiggemann & Zaccardo, 2016; Yom-Tov & Boyd, 2014) a text was created that would briefly explain how the pro-anorexia community tries to manipulate people.
Next, the refutational preemption followed informing participants about what they might be faced with on Instagram (see Appendix C4). The threat ends with the statement that pro-anorexic profiles attempt to merge with accepted categories on Instagram. The
refutational preemption consisted of tips on how to recognize pro-anorexia profiles through indicators that are most common to appear on pro-anorexia profiles (Ojvan, 2015; Sidani et al., 2016; Tiggemann & Zaccardo, 2016). Thus, the refutational preemption served as an educational message and enabled people to distinguish between pro-anorexia and non-pro-anorexia profiles. Half of the participants were presented with this treatment before seeing the Instagram profiles.
Participants
Selection. This research concentrates on the effects of identification with pro-anorexic
sources on female students. Research showed that women are more susceptible to developing eating disorders than men and engage in dieting behavior more often (Wittchen & Hoyer, 2011). In addition, pro-anorexics present content mostly in a feminine way (Haas et al., 2010). Furthermore, anorexia nervosa and dieting in general occurs distinctively more for highly educated people (Wittchen & Hoyer, 2011). Within the context of this research, the
probability to reach primarily students was high. Students share a high level of education, a young age (18-30) and a similar lifestyle. Hence, they act as a rather homogeneous group,
which would serve the purpose of this study. In order to test identification with a large sample, it needed to be insofar homogeneous that high and low identification could be manipulated successfully.
Characteristics. To test the assumptions of this study, an online experiment was
conducted. Therefore, an online questionnaire including the material was distributed to 343 participants. Thirty-one participants were dropped from the study as they did not finish the questionnaire. In the final sample of 312 participants, the mean age was 23.9 (SD = 2.5) and all participants were female students. 87.5% of the participants stated to have Instagram and 77.9% use it frequently. Concerning the pro-anorexia movement 76.9% of the participants indicated that they had heard at least once about it.
Measures
Identification. In order to check whether the manipulation of identification was
successful a question block and two single questions were included at the end of the questionnaire. Four items were adapted from an identification scale by Sani et. al (2014) to measure whether the respondents identified with the pro-anorexic source of the profile in accordance to the condition. An exemplary item is “I feel somewhat similar to the person in the profile”. Participants were asked to indicate how much they agreed with these statements on a seven-point Likert scale from 1 (strongly disagree) to 7 (strongly agree). Reliability of the scale is good (Cronbach’s alpha = .80). Thus, it appears that the scale measures
identification with the pro-anorexic source (M = 2.79, SD = 0.92): The higher the score on the scale, the more a participant identifies with the pro-anorexic source. Furthermore, the
questions “How old was the person in profile 3?” and “What was the employment of the person?” measured whether participants noticed the manipulated material attentively. Answers options included both conditions and a wrong answer.
Self-esteem. In order to measure self-esteem, Rosenberg’s Self-Esteem scale (1965)
were asked to indicate how much they agreed with these statements on a seven-point Likert scale from 1 (strongly disagree) to 7 (strongly agree). Further, four items of the Body Esteem scale by Franzoi and Shields (1984) were added for the context of this study. An exemplary item was “I am satisfied with my weight”. Participants were asked to indicate how much they agreed with these statements on a seven-point Likert scale from 1 (strongly disagree) to 7 (strongly agree). A principal component analysis (PCA) showed that the 14 items form two uni-dimensional scales. Orthogonal rotation was used and illustrated that two components had an eigenvalue above one (component 1: 6.27; component 2: 1.97). The items of the second component seemed measure body satisfaction rather than self-esteem. Since this study focuses on effects on self-esteem in general, Rosenberg’s self-esteem scale (1965) was a better fit. The reliability of the scale is good (Cronbach’s alpha = .88). Therefore, the scale appears to measure self-esteem (M = 4.18, SD = 1.01): the higher the scale score, the higher a
respondent’s general self-esteem.
Dieting intentions. To test dieting intentions this study employed the dieting intention
scale (DIS) by Cruwsky, Platow, Rieger and Byrne (2013). An exemplary item is “In the next three weeks I intend to reduce my calorie intake”. Participants were asked to indicate how much they agreed with these statements on a seven-point-Likert scale from 1 (strongly disagree) to 7 (strongly agree). Another set of question items was introduced with “If I diet in the next three weeks, this would be…” and followed by answer options as for example from 1 (useless) to 7 (useful). Reliability of the scale is good (Cronbach’s alpha = .94). Thus, it appears that the scale measures dieting intentions and shows whether a participant intends to diet in the near future (M = 3.41, SD = 1.40): The higher a participants scores on the scale, the higher the dieting intentions of the person.
Message persuasiveness. This variable was operationalized with nine items. These
items were newly developed following Pounders and colleagues (2015) as well as Chandran and Menon (2004). They had to be adapted to match the pro-anorexic message of weight loss.
An exemplary item is “The profile was convincing in terms of weight loss tips”. All items are presented in Appendix B. A principal component analysis (PCA) showed that the nine items form a single uni-dimensional scale. Only one component had an eigenvalue above one (4.38) and there was a clear point of inflexion after this component in the scree plot. Reliability of the scale is good (Cronbach’s alpha = .84). Thus, it appears that the scale measures message persuasiveness (M = 2.97, SD = 1.17): The higher a respondent scores on the scale, the more persuasive the person perceives the message of weight loss to be.
Results Manipulation Check
All participants identified the age of the source and the employment correctly. Thus, participants noticed the manipulation. An independent t-test indicated that the difference in mean identification displayed in the high identification condition (M = 4.19, SD = .45) was significantly different from participants in the low identification condition (M = 1.68, SD = .74), t (310) = 26.14, p < .001. In relation to the scale for identification, participants in both conditions identified according to the respective condition. Therefore, the manipulation for identification was successful.
Main Effects
In order to test the main effects of identification on self-esteem, dieting intentions and persuasiveness, three separate univariate analyses of variances were performed.
Self-esteem. Hypothesis 1a postulated that identification with the pro-anorexic source
would lower self-esteem. An ANOVA was carried out to assess the influence of identification on self-esteem. We found a significant, small main effect of identification on the respondent’s self-esteem, F (1,308) = 19.98, p < .001, η² = .06. Participants that highly identified with the pro-anorexic source had lower self-esteem (M = 4.09, SD = 1.09) than participants that identified low (M = 5.05, SD = .84). Thus, we accept H1a.
Dieting intentions. Identification with the source was expected to increase dieting
intentions (H1b). Another ANOVA was carried out to measure the influence of identification on dieting intentions. We found a significant, small main effect of identification with the pro-anorexic source on dieting intentions, F (1,308) = 12.00, p = .001, η² = .04. Participants that highly identified with the pro-anorexic source had higher dieting intentions (M = 4.44, SD = 1.64) than participants that identified low (M = 3.31, SD = 1.37). Thus, we also accept H1b.
Persuasiveness. The effect of identification on message persuasiveness was
hypothesized to be moderated by the resistance-increasing treatment. A third ANOVA was carried out to assess the influence of identification on message persuasiveness, in conjunction with the resistance-increasing strategy. We found a significant, small main effect of
identification on message persuasiveness, F (1,308) = 9.56, p = .002, η² = .03. Participants that highly identified with the pro-anorexic source perceived the message as more persuasive (M = 3.08, SD = 1.35) than participants that identified low (M = 2.25, SD = 1.11). Hence, H2a is accepted.
Furthermore, we found a significant, small main effect of the resistance-increasing treatment on persuasiveness, F (1,308) = 5.195, p = .023, η² = .02. Participants that received the treatment perceived the content less persuasive (M = 2.36, SD = 2.88) than participants that did not receive the treatment (M = 2.97, SD = 1.65). Therefore, H3a is accepted.
Moreover, there was a significant, small interaction effect between the resistance-increasing treatment and identification with the pro-anorexic source, F (1,308) = 4.119, p = .043, η² = .01. In the no treatment condition, people that identified highly with the source perceived the content distinctively more persuasive than those that identified low, M difference =
1.37, p < .001. In the treatment condition, no significant difference was found on whether or not participants identified with the pro-anorexic source, M difference = .28, p = .543. People that
received the treatment perceived the content less persuasive both when they identified highly (M = 2.5, SD = 2.12) and when they identified low (M = 2.22, SD = 1.14). Hence, for people
that identified highly with the pro-anorexic source, it makes a great difference whether they receive the resistance-increasing treatment (M = 2.5, SD = 2.12) or not (M = 3.65, SD = 2.64) in terms of message persuasiveness. Thus, hypothesis 3b is supported.
Mediation Effects
Both the effects of identification on self-esteem and dieting intentions were expected to be mediated by persuasiveness. Additionally, the mediation was expected to be moderated by the resistance-increasing treatment. Hence, two separate analyses of covariance were performed.
Self-esteem. It was assumed that the relationship of identification and self-esteem
would be mediated by message persuasiveness (hypothesis 2b). As shown before,
identification affects self-esteem directly. An ANCOVA was carried out using persuasiveness as covariate. For self-esteem, we found both a significant, small effect for identification, F (1,307) = 11.710, p = .001, η² = .03, and a significant, moderate effect for persuasiveness, F (1,307) = 69.229, p < .001, η² = .17. Since identification was still significant with
persuasiveness as a covariate, it shows that there is no indirect effect of identification on self-esteem that could be explained by message persuasiveness. Thus, H2b is rejected.
The resistance-increasing treatment was expected to lower the effects of identification on persuasiveness and thus increase self-esteem (H4a). There was no significant effect of the resistance-increasing strategy on self-esteem, F (1,307) = 3.50, p = .062. Neither was there a significant interaction effect of identification and resistance-increasing strategy on self-esteem, F (1,307) = 3.364, p = .068. Participants that received the treatment and highly identified with the pro-anorexic source did not score distinctively higher on self-esteem (M = 3.92, SD = 1.89) than participants that did not receive the treatment and also identified highly with the source (M = 4.26, SD = 1.01) when controlling for persuasiveness. Thus, we cannot conclude that there is a moderated mediation of identification on self-esteem. Therefore, hypothesis 4a is rejected.
Dieting intentions. Hypothesis 2c supposed that the effect of identification on dieting
intentions would be mediated by message persuasiveness. Previously, a direct effect of identification on dieting intentions was found. A second ANCOVA was carried out with message persuasiveness as covariate. For dieting intentions, we found both a significant, small effect for identification, F (1,307) = 6.466, p = .011, η² = .02, and a significant, moderate effect for persuasiveness, F (1,307) = 37.855, p < .001, η² = .10. Since identification was still significant when using persuasiveness as a covariate, there is no indirect effect of
identification on dieting intentions that could be explained by message persuasiveness. Thus, H2c is also rejected.
Further, the resistance-increasing treatment was supposed to lower the effects of identification on persuasiveness and thus decrease dieting intentions (H4b). There was no significant effect of the resistance-increasing strategy on dieting intentions, F (1,307) = .508, p = .477. Neither was there a significant interaction effect of identification and resistance-increasing strategy on dieting intentions, F (1,307) = .893, p = .893. Participants that received the treatment and highly identified with the pro-anorexic source did not score distinctively lower on dieting intentions (M = 4.27, SD = 1.53) than participants that did not receive the treatment but also identified highly with the source (M = 4.01, SD = 2.03). Therefore, we can conclude that there is no moderated mediation of identification on dieting intentions either. Hence, H4b is rejected.
Conclusion and Discussion
This research aimed to investigate the effects of anti-health promotion at the example of the pro-anorexia movement. Specifically, this research examined how identification with pro-anorexic sources affects self-esteem and dieting intentions. It was assumed that
persuasiveness mediates these effects. In addition, the resistance-increasing strategy of inoculation was employed to counter persuasiveness and consequently reduce effects of identification on self-esteem and dieting intentions.
The results illustrate that there is a direct effect of identification on self-esteem and dieting intentions and thus confirms the first hypothesis. In the high identification condition, people indeed had lower self-esteem than in the low identification condition. Likewise, the results confirm that people that identify highly with the pro-anorexic source have higher dieting intentions than people that identify low.
Furthermore, the results portray that message persuasiveness does not influence the effects of identification on self-esteem nor on dieting intentions indirectly. This refutes the second hypothesis that persuasiveness mediates the effects of identification on self-esteem and dieting intentions.
In terms of the resistance-increasing strategy, the results show that the effect of identification on persuasiveness can be moderated by the resistance-increasing treatment of inoculation. The third hypothesis is thus confirmed. Nonetheless, the moderation stops with persuasiveness. The treatment affected neither self-esteem nor dieting intentions. As there was no mediation effect of persuasiveness, there was no further effect of the resistance-increasing treatment on self-esteem or dieting intentions. This refutes the fourth hypothesis. The
resistance-increasing treatment was not successful.
To answer the research question, it can be concluded that identification with a pro-anorexic source affects self-esteem negatively and dieting intentions positively. This is not mediated by message persuasiveness. The resistance-increasing strategy of inoculation did not moderate these effects; it only affected persuasiveness. Thus, inoculation is not effective in countering effects of pro-anorexia promotion in relation to identification processes. Figure A3 summarizes the effects.
Implications
To date, this is the first study that investigated the effects of identification in the context of anti-health promotion. Health promotion consists of a large research base
This study adds insight into the processes of anti-health promotion. Previous studies on pro-anorexia indicated that exposure affects people’s self-esteem and dieting intentions. However, this study examined identification as a specific factor of exposure. It was shown that
identification processes contribute to the strength of exposure effects. As people identify with a source, they adapt behaviors and mindsets they are presented with, as proposed by the social cognitive theory (Bandura, 1988). Although people were informed and warned about the dangerous content, identification influenced self-esteem and dieting intentions stronger than the resistance-increasing treatment. This leads to the question of why identification affects outcomes stronger and what measures could counter the process of identification effectively.
In addition, this study examined effects in the social media context. Until now, effects of pro-anorexic content were investigated solely on blogs. However, as Sidani et al. (2016) proposed, the investigation of pro-anorexia on social media is urgent due to its popularity. This study found that effects on social media occur in line with previous research on pro-anorexia (Bardone-Cone & Cass, 2008; Custers & Van den Bulck, 2009).
Limitations
This study focused on the effects of identification with pro-anorexic sources on female students. This group is very specific, which is why the findings of this study are not
generalizable to all populations. Based on eating disorder research, exposure to pro-anorexic content mostly affects subgroups of the general public (Custers & Van den Bulck, 2009; Wittchen & Hoyer, 2011). Teenage girls or girls with different educational backgrounds might be susceptible to the effects as well, as Custers and Van den Bulck (2009) established effects of pro-anorexia blogs on schoolchildren. However, the chosen sample fits the context since it is a highly susceptible population in terms of dieting and the development of eating disorders (Wittchen & Hoyer, 2011). Future research should investigate whether identification processes with pro-anorexic sources affect other populations as well.
Additionally, this study used dieting intentions as measures of future behavior. Research has shown that intentions predict behavior (Ajzen, 1991). However, behavior does not always succeed intentions. Extensive research on the intention-behavior gap demonstrates that intentions do not equal behavior (Conner & Norman, 2005; Sheeran, 2002). Consequently, this study cannot provide information about actual dieting behavior as a cause of identification. When people report dieting behavior, several factors can influence the correctness and the preciseness of these self-reports. Social desirability has been proven to influence the truthfulness of self-reported dieting behavior (Herbert et al., 1995). Further, people tend not to remember precisely how they behaved in different situations (Neumark-Sztainer et al., 1997). Due to these methodological challenges, this study focused on dieting intentions.
Since this research had to comply to ethical standards effects might differ when participants view the actual variety of pro-anorexic profiles. Graphic images might evoke factors as disgust or social norms that could influence people’s perceptions. However, this study shows that subtle forms of pro-anorexia profiles influence users as well. In fact, subtle forms of anti-health promotion raised concern in recent literature on fitspiration and pro-anorexia (Custers, 2015; Sidani et al., 2016). In line with the findings of pro-pro-anorexia research, studies indicated that the fitspiration movement poses a threat and affects body satisfaction and dieting behavior negatively (Holland & Hons, 2016; Tiggemann & Zaccardo, 2016). These two communities are difficult to distinguish and demand further investigation.
Future Research
The findings of this study could be useful for other anti-health groups. The chosen dependent variables are specifically important for the context of the pro-anorexia community. Nonetheless, identification could play a role in other anti-health promotion contexts as well. It might be useful to investigate whether identification affects promoting smoking, self-harm or other anti-health concepts.
Identification influenced behavioral intentions and feelings despite reduced
persuasiveness and resistance-increasing treatment. Hence, future research should concentrate on minimizing the identification process with pro-anorexic sources. However, the
implementation might pose a challenge. Entertainment-Education suggests that identification can lower counter arguing and make a message more acceptable (Moyer-Gusé, 2008).
Bandura (1995) proposes that behavioral effects will be stronger the more an observer identifies with a model. Both these theories might aid in creating a successful counter
measure. In an entertainment environment people might be more inclined to listen to a person they identify with that presents a warning instead of a pro-anorexic source. Consequently, the effects of self-esteem and dieting intentions might decrease if the identification with the person presenting the warning is stronger than with the pro-anorexic source. Future research should inspect whether identification could be transferred to an educational source and whether this might decrease the effects of pro-anorexia promotion.
Moreover, the concept of identification needs further examination. This research revealed a lack of knowledge as to how identification works on a cognitive and affective level. Even though people were informed about a danger, the fact of identifying with the source had a greater impact on the outcomes. Since the identification manipulation was very subtle, it poses the question of how consciously the identification affected the participants. Although the participants in the treatment condition were warned about the content, it could not hinder the effects of identification. This suggests that unconscious processes accompany the effects of identification on behavioral intention and feelings. The underlying processes of identification demand further investigation. Future research should examine which aspects of identification consciously and unconsciously influence people. The concept of demonstrating people their own vulnerability to be manipulated might serve to gain insight into these processes (Sagarin et al., 2002). A follow-up study could investigate how participants would
react when presented with the results of this study and whether this could decrease effects on self-esteem and dieting intentions.
Conclusion
Overall, this study demonstrated that not only exposure affects self-esteem and dieting intentions in the context of pro-anorexia but specifically identification with a pro-anorexic source. In fact, identification affects self-esteem negatively and dieting intentions positively. It does not matter whether people perceive the content as persuasive or not when
identification processes occur. The resistance-increasing treatment could not counter effects of identification, which points to future research. The field of anti-health should be further investigated in relation to identification processes and explore further counter measures.
References
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179 –211. doi:10.1016/0749- 5978(91)90020-T
Autenrieth, Ulla (2014). Das Phänomen ‚Selfie‘: Handlungsorientierungen und
Herausforderungen der fotografischen Selbstinszenierung von Jugendlichen im Social Web. In J. Lauffer, R. Röllecke (Eds.), Lieben, Liken, Spielen: Digitale Kommunikation und Selbstdarstellung Jugendlicher heute: Medienpädagogische Konzepte und
Perspektiven. (pp. 52-59). München: Kopaed.
Banas, J. A., & Rains, S. A. (2010). A meta-analysis of research on inoculation theory. Communication Monographs, 77(3), 281–311. doi.org/10.1080/0 3637751003758193 Bandura, A. (1988). Organizational application of social cognitive theory. Australian
Journal of Management, 13(2), 275–302, doi:10.1177/031289628801300210 Bandura, A. (1995). Self-efficacy in changing societies. Cambridge; New York: Cambridge
University Press.
Bardone-Cone, A. M., & Cass, K. M. (2006). Investigating the impact of pro-anorexia websites: A pilot study, European Eating Disorders Review, 14(June), 256–262.
Belch, G. E., & Belch, M. A. (2015). Advertising and promotion. An integrated marketing cmmunications perspective (10th ed.). New York: McGraw-Hill Education.
Boyd, D., Ryan, J., & Leavitt, A. (2010). Pro-self-harm and the visibility of youth-generated problematic content. Journal of Law and Policy for the Information Society. 7(1), 1-32. Bryman, A. (2012). Social research methods (4th ed.). New York: Oxford University Press. Burgoon, M., Alvara, E., Grandpre, J., & Voulodakis, M. (2002). Revisiting the theory of
psychological reactance: Communicating threats to attitudinal freedom. In J.P. Dillard & M. Pfau (Eds.), The Persuasion Handbook: Developments in Theory and Practice (pp. 213-232). Thousand Oaks: Sage Publications.
Chandran, S., & Menon, G. (2004). When a day means more than a year: Effects of temporal framing on judgments of health risk, Journal of Consumer Research, 31(2), 375–389. Compton, J., Jackson, B., & Dimmock, J. A. (2016). Persuading others to avoid persuasion:
Inoculation theory and resistant health attitudes. Frontiers in Psychology, 7(February), 1–9. doi.org/10.3389/fpsyg.2016.00122
Conner, M., & Norman, P. (2005). Predicting health behaviour: Research and practice with social cognition models (2nd ed.). Maidenhead: McGraw-Hill Education.
Csipke, E., & Horne, O. (2007). Pro-eating disorder websites: User’s opinions. European Eating Disorders Review, 15(3), 196–206. doi.org/10.1002/erv.789
Cruwys, T., Platow, M. J., Rieger, E., & Byrne, D. G. (2013). The development and validation of the Dieting Intentions scale (DIS). Psychological assessment, 25(1), 264-278. Custers, K. (2015). The urgent matter of online pro-eating disorder content and children:
clinical practice. European Journal of Pediatrics, 174(4), 429-433. doi.org/10.1007 /s00431-015-2487-7
Custers, K., & Van den Bulck, J. (2009). Viewership of pro-anorexia websites in seventh, ninth and eleventh graders. European Eating Disorders Review, 17(3), 214–219. https://doi.org/10.1002/erv.910
Epstein, J. A., & Botvin, G. J. (2008). Media resistance skills and drug skill refusal
techniques: What is their relationship with alcohol use among inner-city adolescents? Addictive Behaviors, 33(4), 528–537. https://doi.org/10.1016/j.addbeh.2007.11.001 Fardouly, J., Diedrichs, P. C., Vartanian, L. R., & Halliwell, E. (2015). Social comparisons on
social media: The impact of Facebook on young women’s body image concerns and mood. Body Image, 13, 38–45. doi.org/10.1016/j.bodyim.2014.12.002
Fardouly, J., & Vartanian, L. R. (2015). Negative comparisons about one’s appearance mediate the relationship between Facebook usage and body image concerns. Body Image, 12(1), 82–88. doi.org/10.1016/j.bodyim.2014.10.004
Festinger, L. A. (1954). A theory of social comparison processes. Human Relations, 7, 117-140.
Fleming-May, R. A., & Miller, L. E. (2010). “I’m scared to look. but I’m dying to know”: Information seeking and sharing on pro-ana weblogs. Proceedings of the American Society for Information Science and Technology, 47(1), 1-9. doi.org/10.1002/meet. 14504701212
Franzoi, S. L., & Shields, S. A. (1984). The Body Esteem scale: Multidimensional structure and sex differences in a college population. Journal of Personality Assessment, 48(2), 173-178. doi.org/10.1207/s15327752jpa4802_12
Gielen, A. C., & Mcdonald, E. M. (2002). Using the PRECEDE-PROCEED planning model to apply health behavior theories. In K. Glanz, B. K. Rimer, & F. M. Lewis, Health behavior and health education: Theory, research and practice (pp. 409-436). San Francisco: Jossey-Bass.
Groothof, H. A. K., Siero, F. W., & Buunk, A. P. (2007). Doing worse, but feeling happy: Social comparison and identification in response to upward and downward targets. Revue Internationale de Psychologie Sociale, 20(1), 125–144.
grounded model of online negative enabling support groups in the pro-anorexia movement. New Media & Society, 13(1), 40–57. doi.org/10.1177/1461444810363910 Harshbarger, J. L., Ahlers-Schmidt, C. R., Mayans, L., Mayans, D., & Hawkins, J. H. (2009).
Pro-anorexia websites: What a clinician should know. International Journal of Eating Disorders, 42(4), 367–370. doi.org/10.1002/eat.20608
Hebert, J. R., Clemow, L., Pbert, L., Ockene, I. S., & Ockene, J. K. (1995). Social desirability bias in dietary self-report may compromise the validity of dietary intake measures. International Journal of Epidemiology, 24, 389 –398. doi:10.1093/ije/24.2.389 Holland, G., & Tiggemann, M. (2016). “Strong beats skinny every time”: Disordered eating
and compulsive exercise in women who post fitspiration on Instagram. International Journal of Eating Disorders, 1–4. doi.org/10.1002/eat.22559
Instagram. (2016a). Guidelines. Retrieved from https://help.instagram.com/477434105621119
/?helpref=hc_fnav
Instagram. (2016b). Using Instagram. Retrieved from https://help.instagram.com/936495066
470190/?helpref=hc_fnav
Javaras, K. N., Laird, N. M., Reichborn-Kjennerud, T., Bulik, C. M., Pope, H. G., & Hudson, J. I. (2008). Familiarity and heritability of binge eating disorder: Results of a case-control family study and a twin study. The International Journal of Eating Disorders, 41(2), 174–179. https://doi.org/10.1002/eat
Jordan, C.H., Ziegler-Hill, V. & Cameron, J.J. (2015). Self-esteem. In International Encyclopedia of the Social & Behavioral Sciences (2nd ed., pp. 522-528).
Kok, G., Schaalma, H., Ruiter, R. A. C., van Empelen, P., & Brug, J. (2004). Intervention mapping: A protocol for applying health psychology theory to prevention programmes. Journal of Health Psychology, 9(1), 85–98. doi.org/10.1177/1359105304038379 Lee, S. Y. (2010). Ad-induced affect: The effects of forewarning, affect intensity, and prior
10.1080/13527260902869038
Linn, A. J., van Weert, J. C. M., Smit, E. G., Perry, K., & van Dijk, L. (2013). 1+1=3? The systematic development of a theoretical and evidence-based tailored multimedia
intervention to improve medication adherence. Patient Education and Counseling, 93(3), 381–388. doi.org/10.1016/j.pec.2013.03.009
Manago, A. M., Monique Ward, L., Lemm, K. M., Reed, L., & Seabrook, R. (2015). Facebook involvement, objectified body consciousness, body shame, and sexual assertiveness in college women and men. Sex Roles, 72(1), 1–14. doi.org/10.1007/ s11199-014-0441-1
Martijn, C., Smeets, E., Jansen, A., Hoeymans, N., & Schoemaker, C. (2009). Don’t get the message: The effect of a warning text before visiting a proanorexia website.
International Journal of Eating Disorders, 42(2), 139–145. doi.org/10.1002/eat.20598 McGuire, W. J., & Papageorgis, D. (1961). The relative efficacy of various types of prior
belief-defense in producing immunity against persuasion. The Journal of Abnormal and Social Psychology, 62(2), 327–337. doi.org/10.1037/h0042026
Moreno, M. A., Ton, A., Selkie, E., & Evans, Y. (2016). Secret society 123: Understanding the language of self-harm on Instagram. Journal of Adolescent Health, 58(1), 78–84. doi.org/10.1016/j.jadohealth.2015.09.015
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.org/10.1111/j.1468-2885.2008.00328.x
Neumark-Sztainer, D., Jeffery, R. W., & French, S. A. (1997). Self-reported dieting: How should we ask? What does it mean? Associations between dieting and reported energy intake. International Journal of Eating Disorders, 22, 437–449.
van Noort, G. & Voorveld, H.A.M. (2014). Social media in advertising campaigns.
Examining the effects on perceived persuasive intent, campaign and brand response. Journal of Creative Communication, 9 (3), 253-268.
Ojvan, N. (2015). Hungern 3.0? Eine Untersuchung von Pro Ana-Profilen auf Instagram. (Unpublished bachelor’s thesis). Universität Hamburg, Germany.
Perry, C. L., Slinkard, L. A., Killen, J., & McAlister, A. (1980). Peer teaching and smoking prevention among junior high students. Adolscence, 15(58), 277-281.
Pfau, M., & Van Bockern, S. (1994). The Persistence of inoculation in conferring resistance to smoking initiation among adolescents. The second year. Human Communication Research, 20(3), 413–430. doi.org/10.2975/31.3.2008.183.184
Pounders, K. R., Lee, S., & Mackert, M. (2015). Matching temporal frame, self-view, and message frame valence : Improving persuasiveness in health communications, Journal of Advertising, 44(4), 388-402. doi.org/10.1080/00913367.2015.1071210
van Reijmersdal, E. A., Fransen, M. L., van Noort, G., Opree, S. J., Vandeberg, L., Reusch, S., & van Lieshout, F. (2016). Effects of disclosing sponsored content in blogs : How the use of resistance strategies mediates effects on persuasion. American Behavioral
Scientist, 60(12), 1458-1474. doi.org/10.1177/0002764216660141
Rodgers, R. F., Skowron, S., & Chabrol, H. (2012). Disordered eating and group membership among members of a pro-anorexic online community. European Eating Disorders Review, 20(1), 9–12. doi.org/10.1002/erv.1096
Rosenberg, M. (1965). Society and the adolescent self-image.Princeton, NJ: Princeton University Press.
Rothman, A. J., Bartels, R. D., Wlaschin, J., & Salovey, P. (2006). The strategic use of gain- and loss-framed messages to promote healthy behavior: How theory can inform practice. Journal of Communication, 56(Supplement). doi.org/10.1111/j.1460-2466.2006.00290.x Sagarin, B. J., Cialdini, R. B., Rice, W. E., & Serna, S. B. (2002). Dispelling the illusion of
invulnerability: The motivations and mechanisms of resistance to persuasion. Journal of Personality and Social Psychology, 83(3), 526–541. doi.org/10.1037/h0087893
Sani, F., Madhok, V., Norbury, M., Dugard, P., & Wakefield, J. R. H. (2015). Greater number of group identifications is associated with lower odds of being depressed: Evidence from a Scottish community sample. Social Psychiatry and Psychiatric Epidemiology, 50(9), 1389–1397. doi.org/10.1007/s00127-015-1076-4
Santarossa, S., Coyne, P., Lisinski, C., & Woodruff, S. J. (2016). #fitspo on Instagram: A mixed-methods approach using Netlytic and photo analysis, uncovering the online discussion and author/image characteristics. British Journal of Health Psychology, 11(2), 235-248. doi.org/10.1177/1359105316676334
Schmidt, J.-H. (2009). Das neue Netz: Merkmale, Praktiken und Folgen des Web 2.0. Konstanz: UVK Verlagsgesellschaft mbH.
Setodji, C. M., Martino, S. C., Scharf, D. M., & Shadel, W. G. (2014). Quantifying the persistence of pro-smoking media effects on college students’ smoking risk. Journal of Adolescent Health, 54(4), 474–480. doi.org/10.1016/j.jadohealth.2013.09.011
Sheeran, P. (2002). Intention-behavior relations : A conceptual and empirical review. European Review of Social Psychology, 12(1), 1–36. doi.org/10.1080/14792772143 000003
Sidani, J. E., Shensa, A., Hoffman, B., Hanmer, J., & Primack, B. A. (2016). The association between social media use and eating concerns among US young adults. Journal of the Academy of Nutrition and Dietetics, 116(9), 1465-1472. doi.org/10.1016/j.jand. 2016.03.021
Tierney, S. (2008). Creating communities in cyberspace: Pro-anorexia web sites and social capital. Journal of Psychiatric and Mental Health Nursing, 15(Commentary), 340–343. Tiggemann, M., & Zaccardo, M. (2015). “Exercise to be fit, not skinny”: The effect of
1016/j.bodyim.2015.06.003
Tiggemann, M., & Zaccardo, M. (2016). ’Strong is the new skinny': A content analysis of #fitspiration images on Instagram. Journal of Health Psychology, 2(17), 132-135. doi.org/ 10.1177/1359105316639436
Wilcox, K., & Laird, J. D. (2000). The impact of media images of super-slender women on women’s self-esteem: Identification, social comparison, and self-perception, Journal of Research in Personality, 34, 278–286. doi.org/10.1006/jrpe.1999.2281
Wittchen, H.-U., & Hoyer, J. (2011). Klinische Psychologie und Psychotherapie (2nd ed.). Berlin; Heidelberg; New York: Springer-Verlag.
World Health Organisation. (2005). The Bangkok charter for health promotion in a globalised world, 1–6. Retrieved from http://www.who.int/healthpromotion/conferences
/6gchp/hpr_050829_ BCHP.pdf
Yom-Tov, E., & Boyd, D. M. (2014). On the link between media coverage of anorexia and pro-anorexic practices on the web. International Journal of Eating Disorders, 47(2), 196–202. doi.org/10.1002/eat.22195
Appendix A Conceptual Models
Figure A1. Conceptual model including hypothesized main effects of identification on self-esteem and dieting intentions and mediation effects of persuasiveness on identification and self-esteem and dieting intentions.
Figure A2. Conceptual model including hypothesized main effects of identification on self-esteem and dieting intentions and mediation effects of persuasiveness on identification and self-esteem and dieting intentions and the moderated mediation of resistance-increasing strategy.
Figure A3. Conceptual model including results of main effects of identification on self-esteem and dieting intentions and moderation effects of resistance-increasing strategy on
identification and persuasiveness.
Appendix B
Experimental Questionnaire
Welcome to the experiment "Instagram perceptions"!
Thank you very much for clicking the link and thus helping me to find insights into a very interesting topic: On Instagram people present themselves in various ways and often declare themselves as experts of certain topics. How do followers perceive profiles where people like you and me share tips and tricks or motivate one another in certain areas? How do we feel about these people? Are they similar to us?
In the following you will be presented with Instagram profiles. Please take a close look at them. Afterwards you will answer a couple of questions about your perceptions about these profiles as well as some questions about yourself.
Everything is absolutely anonymous, so you can feel free to answer openly and honestly. Let's get started!
I hereby declare that I have been informed in a clear manner about the nature and method of the research, as described in the invitation for this study. I agree, fully and voluntarily, to participate in this research study. With this, I retain the right to withdraw my consent, without having to give a reason for doing so. I am aware that I may halt my participation in the
experiment at any time. If my research results are used in scientific publications or are made public in another way, this will be done such a way that my anonymity is completely
safeguarded. My personal data will not be passed on to third parties without my express permission. If I wish to receive more information about the research, either now or in future, I can contact natalieojvan@gmx.de. Should I have any complaints about this research, I can contact the designated member of the Ethics Committee representing the ASCoR, at the following address: ASCoR secretariat, Ethics Committee, University of Amsterdam, Postbus 15793, 1001 NG Amsterdam; 020‐ 525 3680; ascor‐secr‐fmg@uva.nl.
I understand the text presented above, and I agree to participate in the research study.
WATCH OUT!
On Instagram, not everyone is the expert he or she claims to be. You have to be very cautious because some people will try to promote unhealthy thoughts and behaviors.
One group that does this frequently is the pro anorexia movement. Here, young girls promote and glorify the mental illness anorexia as a lifestyle and give tips and tricks on how to lose weight dangerously fast. The so called pro anas (short for anorexia) will try to subliminally convince you that their way is the only way to live and lose weight. Since IG knows about the existence of these pro ana profiles and tries to act against them, the pro anas try to fit into other categories on Instagram like fitness and other lifestyle profiles.
