Individual Differences Associated with Exposure to “Ana-Mia” Websites: An Examination of Adolescents from 25 European Countries

Abstract

This study explores the individual differences associated with adolescents’ exposure to “ana-mia” websites (i.e., websites where people discuss ways to be very thin, such as being anorexic). Participants were adolescents from a large cross-national survey in 25 European countries (N = 18,709, aged 11–16, 50% girls). Sociodemographic and individual factors (i.e., variables related to Internet use and personality traits) were included in a logistic regression performed separately for girls and boys. The results showed that sensation seeking and online disinhibition were both associated with an increased risk of exposure to “ana-mia” websites in girls as well as in boys, although some gender differences were apparent. In girls, but not in boys, the older the child and higher the socioeconomic status, higher the chance of being exposed to “ana-mia” websites. Further research is recommended to understand the real impact of “ana-mia” website exposure on adolescent health.

Introduction

The use of the Internet, and particularly social media, is ubiquitous among adolescents. They spend several hours per week engaged in online activities.¹ Consequently, adolescents are exposed to a myriad of appearance-related content/interactions. This exposure on the Internet contrasts with conventional media, in which adolescents are spectators of prefabricated content, such as the appearance-related content of films and magazines. For instance, new media offers youth a forum for discussion and feedback on their own appearance-related content, such as receiving comments on their looks after posting a new profile picture on Facebook.² In general, new media provides adolescents with the opportunity to create their own content online, to have their own spectators, and to drive online discussion and get feedback on their own content.

More importantly, the Internet offers the opportunity to bring together like-minded individuals,³ such as Internet users who already have body image concerns. That is the case for pro-eating disorder (pro-ED) websites that promote eating disorders as authentic lifestyle choices, although they may also provide help and support for recovery from an ED.^4,5 From personal websites or blogs to online social communities and social networks, pro-ED websites encourage disordered eating behaviors and can disseminate potentially harmful information.^6,7 They do so with “tips and tricks” for weight loss or for staying emaciated, by showing images of ultrathin bodies, by containing inspirational words/images known as “thinspiration,” and by providing areas for discussion, such as chat rooms, discussion threads in forums, and other outlets.⁵ Furthermore, this potentially harmful online content flows to different websites,⁸ and the same content from pro-ED websites can be found on websites not overtly pro-ED, which are actually the majority of websites.⁷

The potential risks and effects of pro-ED have recently been outlined and include negative effect, body dissatisfaction, and dieting, in addition to learning unhealthy strategies to lose weight.^4,5 Nevertheless, it is important to keep in mind that not all online content/interactions centered on eating disorders can be classified as pro-ED. As a result, the more neutral and representative term “ana-mia,” which labels websites with content potentially related to anorexia (ana) or bulimia (mia), but not necessarily pro-ED, has emerged as an alternative.⁹ So far, research attention is mostly focused on overtly pro-ED websites, but little is known about the wider range of “ana-mia” websites that are accessible on the Internet. The aim of this study is, therefore, to identify the individual factors associated with the exposure to “ana-mia” websites, with special focus on the personality traits associated with Internet use and risky online behavior (i.e., sensation seeking and online disinhibition).

Individual differences that increase the risk of exposure to “ana-mia” websites

Prior research has uncovered individual differences in online behavior patterns,³ such as sociodemographic factors (e.g., gender), knowledge and skills (e.g., digital skills), and individual characteristics (e.g., online disinhibition). Generally, it has been shown that higher Internet use, higher digital skills, and personality traits such as sensation seeking, are connected with increased risky behavior online,¹⁰ which includes exposure to potentially harmful online content. Nevertheless, few studies have examined the individual-level factors specifically associated with an increased probability of exposure to “ana-mia” websites. Typically, past studies were limited to the examination of associations between body image and eating concerns with the use or exposure to “ana-mia” websites,^11–13 social networks,¹⁴ and problematic Internet use.^15–17

Moreover, these studies are often focused on females, although the few existing studies suggest that males also participate on “ana-mia” websites.¹⁸ For instance, in the qualitative study by Wooldridge et al.,¹⁸ the authors included 12 forums in their content analysis and identified 689 posts by male participants, most of which were related to providing/seeking support and inspiration for weight loss.¹⁸ Yet, we still lack knowledge about males’ exposure to “ana-mia” websites. To the best of our knowledge, only one study on “ana-mia” websites included both adolescent girls and boys and examined a personality trait (in this case, perfectionism).¹³ Nevertheless, diverse personality traits, as well as other individual-level factors, are associated with different patterns of Internet use, as well as engagement in risky behavior online. For instance, a recent review revealed that sensation seeking, as well as gender, play an important role in shaping a person's online behavior.¹⁹ Furthermore, the authors hypothesize that the strength and saliency of certain factors on the Internet (e.g., anonymous communication with strangers in chat rooms) may attract certain personality types (e.g., online disinhibition among introverts).¹⁹ For this study, we take into account these general differences in online behavior patterns and focus specifically on their role in the exposure to “ana-mia” websites, an issue that has not yet been thoroughly studied. To fill this gap, the present study is aimed at examining two salient personality traits associated with Internet use and risky online behavior: sensation seeking and online disinhibition among adolescents of both genders.

Sensation seeking is mainly characterized by the openness to and seeking of new and intense sensations and experiences.²⁰ Previous studies suggest that sensation seekers offline tend to be also sensation seekers online, such as being more likely to communicate with unknown people on the Internet.²¹ Thus, considering that the Internet provides opportunities to gratify sensation-seeking needs and the exposure to “ana-mia” websites can be one of these opportunities, it could be plausible that, among certain individuals, sensation seeking is also associated with looking for “ana-mia” websites. For instance, a study of college women found that online appearance comparisons (i.e., comparing one's appearance to others’ on social media) and online “fat talk” (i.e., talking about others’ bodies and negatively about one's own) were both positively correlated with a measure that indicated sensation seeking by a lack of premeditation/perseverance.²² Similarly, other studies have found a positive association between sensation seeking and problematic Internet use in the general population,²³ as well as in female patients with eating disorders, and particularly in those of the binge/purge subtype.¹⁶ In fact, sensation seeking is considered an important trait associated with disordered eating, particularly with binge/purge behaviors.²⁴ Nevertheless, to the best of our knowledge, no study has examined the association between sensation seeking and the exposure to “ana-mia” websites in adolescents.

Although, “online disinhibition” has been considered a difficult term to define,²⁵ it denotes being less inhibited to exhibit/express behaviors, feelings, or thoughts online compared with face-to-face interactions.²⁶ Thus, online disinhibited individuals feel more comfortable online and are more likely to talk online than in face-to-face interactions, which can, in turn, become problematic, particularly for more vulnerable individuals. For instance, it has been suggested that vulnerable individuals, such as those who are socially anxious or those with body image concerns, can get caught up in a vicious circle by using online communication as a coping mechanism for their psychological difficulties.^17,27 Thus, online disinhibition can promote computer-mediated communication, interactions, and online exposure to diverse content. Therefore, special attention should be given to the effects of online disinhibition during developmental periods, such as childhood and adolescence, and particularly to vulnerable individuals, such as those with body image concerns.

The present study

The aim of this study is to examine the association between the exposure to “ana-mia” websites (i.e., websites where people discuss ways to be very thin, such as being anorexic) and two salient personality traits associated with Internet use (sensation seeking and online disinhibition) in a large cross-national sample of adolescents of both genders in 25 European countries. Additionally, we considered potential confounding variables, specifically: the daily use of the Internet; digital skills; socioeconomic status; and age. These have been found to be related to different online behavior patterns in prior studies.^13,17,28

Methods

Procedure

Data from the EU Kids Online II project (April/October 2010) was used. Approximately 1,000 Internet-using youth in each of 25 participating European countries were sampled using a stratified random probability sampling approach, yielding an overall sample of 25,142 children (50% girls) aged 9–16. Trained interviewers collected data at the child's home, where the child filled out both administered and self-completed questionnaires focused on their online experiences. The research study was conducted in accordance with ESOMAR ethical guidelines and approved by the LSE Research Ethics Committee. Confidentiality and anonymity was guaranteed and all the information and questions were explained. Informed consent was obtained from the parents and children.²⁹ The sample for this study consists of children who were asked about exposure to websites where people discuss eating disorders (aged 11–16). Due to the low prevalence of the studied phenomenon, pooled data from all countries were used for the analysis; N = 18,709, M_age = 13.50, SD = 1.70, 50% girls.

Measures

Exposure to “ana-mia” websites

We asked participants, “In the past 12 months, have you seen websites where people discuss ways to be very thin (such as being anorexic or bulimic)?,” with answers Yes ( = 1; 11.5%) and No ( = 0; 82.5%). Answers “Don't know” (4.2%), “Prefer not to say” (1.4%), and missing values (0.4%) were excluded from the analysis.

Daily use of the Internet

Children were asked, “How often do you use the Internet?” and were divided into those who use it “every day or almost every day,” that is, daily users ( = 1; 73.6%), and those who use it less often, that is, nondaily users ( = 0).

Digital skills

Children were asked, “Which of these things do you know how to do on the Internet?” followed by eight Yes/No items (e.g., “Delete the record of which sites you have visited” or “Block unwanted adverts or junk mail/spam”); positive answers were counted to create a scale; M = 4.43, SD = 2.58, α = 0.76.

Online disinhibition

This was computed as the mean of three statements (e.g., “I find it easier to be myself on the Internet than when I am with people face to face”) answered by “Not true” ( = 0), “A bit true” ( = 1), and “Very true” ( = 2), M = 0.53, SD = 0.52, α = 0.65.

Sensation seeking

This was computed as the mean of two items,³⁰ “I do dangerous things for fun” and “I do exciting things, even if they are dangerous,” answered by “Not true” ( = 0), “A bit true” ( = 1), and “Very true” ( = 2), M = 0.39, SD = 0.53, α = 0.78.

Socioeconomic status of the household

This was assessed through the household's main wage earner's level of education and occupation. Three levels of socioeconomic status (SES) were calculated: Low ( = 1; 20%), Medium ( = 2; 45%), and High ( = 3; 35%).

Gender

The child's gender was coded by the interviewer.

Age

Parents reported the age of the child.

Table 1 includes the descriptive statistics separately by gender.

Table 1.

Descriptive Statistics by Gender (M, SD)

	Age	Digital skills	Sensation seeking	Online disinhibition	Daily Internet use (%)	SES (low, %)	SES (middle, %)	SES (high, %)	Exposure (%)
Boys	13.49 (1.69)	4.56 (2.63)	0.48 (0.57)	0.55 (0.53)	75	20	46	34	7.8
Girls	13.51 (1.66)	4.29 (2.52)	0.29 (0.47)	0.51 (0.51)	73	20	36	43	16.7

The percentages were computed from valid values only.

SES, socioeconomic status.

Data analysis

We conducted three-step hierarchical logistic regression analyses to predict the odds of children's exposure to “ana-mia” websites. Demographics (SES, age), factors related to Internet use (non/daily use and digital skills), and individual characteristics (sensation seeking and online disinhibition), were added consecutively to the models. The analyses were conducted separately for girls and boys.

In the model for girls (Table 2), all variables predicted the outcome positively. Specifically, the older the age, the higher the SES, the more the digital skills, the more the daily Internet use, and higher the levels of sensation seeking and online disinhibition, then higher the chance of being exposed to “ana-mia” websites.

Table 2.

Logistic Regression Predicting Exposure to “Ana-Mia” Websites Among Girls

	Step 1			Step 2			Step 3
	B	SE	OR	B	SE	OR	B	SE	OR
Constant	−6.556	0.292	0.001^***	−6.069	0.299	0.002^***	−6.133	0.309	0.002^***
SES	0.173	0.042	1.189^***	0.103	0.043	1.109^*	0.108	0.044	1.114^*
Age	0.332	0.019	1.394^***	0.233	0.021	1.263^***	0.215	0.021	1.24^***
Daily Internet use				0.364	0.087	1.439^***	0.276	0.088	1.318^**
Digital skills				0.155	0.015	1.168^***	0.139	0.015	1.149^***
Sensation seeking							0.794	0.059	2.213^***
Online disinhibition							0.291	0.06	1.338^***
Cox & Snell, R²	0.04			0.06			0.09
Nagelkerke, R²	0.07			0.10			0.15

***

p < 0.001; ^**p < 0.01; ^*p < 0.05.

OR, odds ratio; SE, standard error.

The results for boys (Table 3) were similar, with two exceptions. First, SES was not significantly linked to the outcome, nor was the age of the boys after the inclusion of other predictors. Second, the final model for boys explained less than the final model for girls (Cox & Snell, R² = 0.03; Nagelkerke, R² = 0.07 for boys versus Cox & Snell, R² = 0.09; Nagelkerke, R² = 0.15 for girls).

Table 3.

Logistic Regression Predicting Exposure to “Ana-Mia” Websites Among Boys

	Step 1			Step 2			Step 3
	B	SE	OR	B	SE	OR	B	SE	OR
Constant	−4.592	0.374	0.01^***	−4.077	0.386	0.017^***	−4.441	0.395	0.012^***
SES	0.028	0.059	1.028	−0.041	0.06	0.959	−0.037	0.06	0.963
Age	0.15	0.025	1.162^***	0.048	0.028	1.049	0.046	0.028	1.047
Daily Internet use				0.412	0.125	1.51^***	0.333	0.126	1.395^**
Digital skills				0.138	0.02	1.148^***	0.105	0.021	1.11^***
Sensation seeking							0.219	0.034	1.245^***
Online disinhibition							0.562	0.076	1.755^***
Cox & Snell, R²	0.01			0.02			0.03
Nagelkerke, R²	0.01			0.04			0.07

***

p < 0.001; ^**p < 0.01.

Discussion

The aim of this study was to examine if certain individual differences (sensation seeking and online disinhibition) are associated with adolescents’ exposure to “ana-mia” websites. We employed data from a cross-national sample of adolescents in 25 European countries.

The results showed that sensation seeking and online disinhibition were both associated with an increased risk of exposure to “ana-mia” websites in girls as well as in boys. Although the past literature examining these topics is scarce, prior studies have found that sensation seeking and online disinhibition are associated with online appearance-related interactions, particularly with appearance-related teasing and cyberbullying.^31,32 Thus, it could be possible that sensation seeking and online disinhibition not only increase the probability of exposure to “ana-mia” websites, but also shape the online interactions within these websites. However, these ideas are speculative and further research would be needed to help clarify which individual differences are the most salient regarding the exposure to and interactions within “ana-mia” websites.

In our investigation, about one out of 10 adolescents reported being exposed to “ana-mia” websites, and this experience was more than double in girls than in boys. This gender difference in the exposure to “ana-mia” websites is very similar to previous reports¹³ and can be explained by the content and target population of these websites. As usually happens with online weight loss advertising,³³ thinspirational online content is represented by and directed to women, typically portraying thin or underweight women to inspire weight loss.³⁴ Likewise, adolescent females place notable importance on body image and weight concerns³⁵ and are more likely to develop body dissatisfaction and eating disorders over time^36,37; this can render them more prone than boys seeking information or support through “ana-mia” websites. Indeed, our results showed that the explained variance was much higher for girls compared with boys, which could suggest that the exposure to “ana-mia” websites would be more intentional for girls with the studied individual characteristics, whereas for boys it would be more random or even unintentional. Moreover, while the odds of exposure increased with age among girls, this did not apply for boys. We controlled both daily Internet use and digital literacy, which could help explain the link to (in some cases probably also unintentional) exposure and the overall amount of Internet use. Yet, we see that the older girls were the more likely to visit an “ana-mia” website. This probably reflects a growing focus on thin appearance in girls, and we suggest that this also indicates that the visits were more often intentional and in line with girls’ overall behavioral focus on weight loss and appearance management. Nevertheless, the fact that boys were also exposed to “ana-mia” websites, as found in studies with community and clinical populations,^13,38 confirms that this phenomenon is not confined to girls and that we should pay more attention to the male population.¹⁸

Moreover, because of the future ubiquity of online information and the difficulty to regulate its content, policy stakeholders should ensure that the population receives education and training in core critical media literacy skills.³⁹ Indeed, media literacy is also considered a key contributor to the prevention of eating disorders.⁴⁰ Furthermore, this knowledge and these activities can be integrated within the promotion of social responsibility (i.e., investment in the well-being of others and the community).⁴¹ For instance, girls can be educated to recognize the importance of taking an active role in online communities to promote a climate of empathy, respect, and ethical participation among members.⁴²

Finally, future research can examine the paths and ways that promote selective exposure to specific online content, such as “ana-mia” websites, as well as how adolescents perceive this online content and why they do so. For instance, it is known that the online content that users see on the Internet is not entirely random since it is usually filtered by computer algorithms that in turn influence online behavior patterns.^43–45 Therefore, future research can examine how this algorithmic regulation affects exposure to “ana-mia” websites. Similarly, an experimental study with adolescent girls found that peer comments about extremely thin media models exert influence on how girls perceive these media models.⁴⁶ Therefore, future research can examine the external factors that influence adolescents’ perceptions of “ana-mia” websites. Future research should also differentiate between intentional and unintentional exposure to “ana-mia” web sites. As already mentioned, it is crucial to further understand the real impact of “ana-mia” website exposure on adolescent health.

Limitations

Although our study has strengths, such as a large sample size, it also has limitations, and our results should be interpreted with caution. First, our findings cannot be interpreted in terms of causality because of the cross-sectional nature of our study. Second, although we included several potentially confounding variables to account for individual differences that can affect exposure to “ana-mia” websites, there are other factors which can be linked to the studied outcome. For instance, it could be possible that some parents use parental control software to block their child's access to specific websites and perhaps this software could block the access to “ana-mia” websites. Finally, we relied on self-reports, which might, in some cases, deviate from the actual experiences and which did not provide specific or detailed information about the duration of exposure, the kind of content seen, and the number and nature of interactions maintained within the “ana-mia” websites. Nevertheless, the characteristics of our study, including a large sample of adolescents of both genders, as well as the variables included in the analyses, provide an important step toward a better understanding of the individual variables that increase the risk of exposure to “ana-mia” websites.

Footnotes

Acknowledgments

The authors acknowledge the support of the Czech Science Foundation (THINLINE—GA15-05696S) and the Faculty of Social Studies, Masaryk University, the Czech Republic.

Author Disclosure Statement

No competing financial interests exist.

References

Perloff

. Social media effects on young women's body image concerns: theoretical perspectives and an agenda for research. Sex Roles, 2014; 71:363–377.

de Vries

, Peter

, de Graaf

, et al. Adolescents’ social network site use, peer appearance-related feedback, and body dissatisfaction: testing a mediation model. Journal of Youth and Adolescence, 2016; 45:211–224.

Amichai-Hamburger

. (2013) The social net: understanding our online behavior. Oxford, UK: Oxford University Press.

Rodgers

, Lowy

, Halperin

, et al. A meta-analysis examining the influence of pro-eating disorder websites on body image and eating pathology. European Eating Disorders Review, 2016; 24:3–8.

Rouleau

, von Ranson

. Potential risks of pro-eating disorder websites. Clinical Psychology Review, 2011; 31:525–531.

Boepple

, Thompson

. A content analytic comparison of fitspiration and thinspiration websites. International Journal of Eating Disorders, 2016; 49:98–101.

Borzekowski

DLG

, Schenk

, Wilson

, et al. e-Ana and e-Mia: a content analysis of pro-eating disorder web sites. American Journal of Public Health, 2010; 100:1526–1534.

Ghaznavi

, Taylor

. Bones, body parts, and sex appeal: an analysis of #thinspiration images on popular social media. Body Image, 2015; 14:54–61.

Casilli

, Mounier

, Pailler

, et al. Les jeunes et le web des troubles alimentaires: dépasser la notion de “pro-ana.” Rapport du projet de recherche ANR ANAMIA “La sociabilité Ana-mia: une approche des troubles alimentaires par les réseaux sociaux en ligne et hors-ligne.” www.webcitation.org/6ce6HJibd (accessed Oct. 29, 2015).

10.

Livingstone

, Smith

. Annual research review: harms experienced by child users of online and mobile technologies: the nature, prevalence and management of sexual and aggressive risks in the digital age. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 2014; 55:635–654.

11.

Harper

, Sperry

, Thompson

. Viewership of pro-eating disorder websites: association with body image and eating disturbances. International Journal of Eating Disorders, 2008; 41:92–95.

12.

Rodgers

, Skowron

, Chabrol

. Disordered eating and group membership among members of a pro-anorexic online community. European Eating Disorders Review, 2012; 20:9–12.

13.

Custers

, Van den Bulck

. Viewership of pro-anorexia websites in seventh, ninth and eleventh graders. European Eating Disorders Review, 2009; 17:214–219.

14.

Tiggemann

, Miller

. The Internet and adolescent girls’ weight satisfaction and drive for thinness. Sex Roles, 2010; 63:79–90.

15.

Koronczai

, Kökönyei

, Urbán

, et al. The mediating effect of self-esteem, depression and anxiety between satisfaction with body appearance and problematic internet use. The American Journal of Drug and Alcohol Abuse, 2013; 39:259–265.

16.

Claes

, Müller

, Norré

, et al. The relationship among compulsive buying, compulsive Internet use and temperament in a sample of female patients with eating disorders. European Eating Disorders Review, 2012; 20:126–131.

17.

Rodgers

, Melioli

, Laconi

, et al. Internet addiction symptoms, disordered eating, and body image avoidance. Cyberpsychology, Behavior, and Social Networking, 2013; 16:56–60.

18.

Wooldridge

, Mok

, Chiu

. Content analysis of male participation in pro-eating disorder web sites. Eating Disorders: The Journal of Treatment & Prevention, 2014; 22:97–110.

19.

Orchard

, Fullwood

. Current perspectives on personality and Internet use. Social Science Computer Review, 2010; 28:155–169.

20.

Zuckerman

. (2007) Sensation seeking and risky behavior. Washington, DC: American Psychological Association.

21.

Livingstone

, Helsper

. Taking risks when communicating on the Internet: the role of offline social-psychological factors in young people's vulnerability to online risks. Information, Communication & Society, 2007; 10:619–644.

22.

Walker

, Thornton

, De Choudhury

, et al. Facebook use and disordered eating in college-aged women. Journal of Adolescent Health, 2015; 57:157–163.

23.

Amichai-Hamburger

. (2009) Personality, individual differences and Internet use. In Joinson

, McKenna

KYA

, Postmes

, Reips

, eds. The Oxford handbook of Internet psychology. New York: Oxford University Press, pp. 187–204.

24.

Cassin

, von Ranson

. Personality and eating disorders: a decade in review. Clinical Psychology Review, 2005; 25:895–916.

25.

Joinson

. (2007) Disinhibition and the Internet. In Gackenbach

, ed. Psychology and the Internet: intrapersonal, interpersonal and transpersonal implications. San Diego: Academic Press, pp. 75–92.

26.

Casale

, Fiovaranti

, Caplan

. Online disinhibition. Journal of Media Psychology, 2015; 27:170–177.

27.

Stanley

, Barnes

, Short

. Appearance-focused Internet use and the thin-beauty ideal. Studies in Media and Communication, 2015; 3:38–50.

28.

Ransom

, La Guardia

, Woody

, et al. Interpersonal interactions on online forums addressing eating concerns. The International Journal of Eating Disorders, 2010; 43:161–170.

29.

Livingstone

, Haddon

, Görzig

, et al. (2011). Technical report and user guide: The 2010 EU Kids Online Survey. London: EU Kids Online. http://eprints.lse.ac.uk/45270/ (accessed Jan. 29, 2015).

30.

Stephenson

, Hoyle

, Palmgreen

, et al. Brief measures of sensation seeking for screening and large-scale surveys. Drug and Alcohol Dependence, 2003; 72:279–286.

31.

Rodgers

, Melioli

. The relationship between body image concerns, eating disorders and Internet use, Part I: a review of empirical support. Adolescent Research Review, 2016; 1:95–119.

32.

Rodgers

. The relationship between body image concerns, eating disorders and Internet use, Part II: an integrated theoretical model. Adolescent Research Review, 2016; 1:121–137.

33.

Slater

, Tiggemann

, Hawkins

, et al. Just one click: a content analysis of advertisements on teen Web sites. Journal of Adolescent Health, 2012; 50:339–345.

34.

Lewis

, Arbuthnott

. Searching for thinspiration: the nature of Internet searches for pro-eating disorder websites. Cyberpsychology, Behavior, and Social Networking, 2012; 15:200–204.

35.

Markey

. Invited commentary: why body image is important to adolescent development. Journal of Youth and Adolescence, 2010; 39:1387–1391.

36.

Keel

. (2010) Epidemiology and course of eating disorders. In Agras

, ed. The Oxford handbook of eating disorders. New York: Oxford University Press, pp. 25–32.

37.

Ferreiro

, Seoane

, Senra

. Toward understanding the role of body dissatisfaction in the gender differences in depressive symptoms and disordered eating: a longitudinal study during adolescence. Journal of Adolescence, 2014; 37:73–84.

38.

Wilson

, Peebles

, Hardy

, et al. Surfing for thinness: a pilot study of pro-eating disorder Web site usage in adolescents with eating disorders. Pediatrics, 2006; 118:e1635–e1643.

39.

Jenkins

. (2009) Confronting the challenges of participatory culture: media education for the 21st century. Cambridge, MA: The MIT Press.

40.

Ciao

, Loth

, Neumark-Sztainer

. Preventing eating disorder pathology: common and unique features of successful eating disorders prevention programs. Current Psychiatry Reports, 2014; 16:453.

41.

Wolk

. (2013) Caring hearts and critical minds: literature, inquiry, and social responsibility. Portland, ME: Stenhouse.

42.

Johnson

. (2015) Digital literacy and digital citizenship: approaches to girls’ online experiences. In Bailey

, Steeves

, eds. eGirls, eCitizens. Ottawa: University of Ottawa Press, pp. 339–360.

43.

O'Reilly

. (2013). Open data and algorithmic regulation. In: Goldstein

, Dyson

, eds. Beyond transparency: open data and the future of civic innovation. San Francisco: Code for America Press, pp. 289–300.

44.

Wells

, Thorson

. (2015). How gatekeeping still matters: understanding media effects in an era of curated flows. In Vos

, Heinderyckx

, eds. Gatekeeping in transition. New York: Routledge, pp. 26–44.

45.

Graells-Garrido

. (2015) Biased behavior in web activities: from understanding to unbiased visual exploration. http://hdl.handle.net/10803/294280 (accessed Oct. 31, 2015).

46.

Veldhuis

, Konijn

, Seidell

. Negotiated media effects. Peer feedback modifies effects of media's thin-body ideal on adolescent girls. Appetite, 2014; 73:172–182.