Results
First, in order to investigate the differences in content between these groups, we conducted series of independent samples t-tests. We determined the type of t-test by using Levene’s Test for Equality of Variances, which varied throughout our analyses. Assuming equal variances, we found that medical blog entries (M=.25) had a higher percentage of words regarding harm/vice compared to pro-Ana blog entries (M=.13 ) on average (t (119)=2.77, p<.01). We also found that medical blog entries (M=.25) had a higher percentage of words regarding authority/virtue than pro-Ana blogs (M=.15) on average (t (68.73)=2.02, p<.01) when not assuming equal variances.
Medical blog entries (M=.48) also had more words pertaining to ingroup/virtue than pro-Ana blog entries (M=.07) in general when not assuming equal variances (t (43.76)=4.62, p<.01). To ensure that this result was not due to medical blog entries that primarily had to do with the topic of family therapy, we took out those samples. The result was still significant in that medical blog entries (M=.35) still consistently had more language pertaining to ingroup/virtue than pro-Ana blogs (M=.07) without assuming equal variances (t (40.52)=4.11, p<.01). We found as well that medical blog entries (M=.08) had more words pertaining to ingroup/vice than pro-Ana blogs entries when not assuming equal variances (t(43.03)=3.69, p<.01).
In addition, we also observed that pro-Ana blog entries (M=.06) had more words pertaining to purity/virtue than medical blog entries (M=0) on average not assuming equal variances (t (78)=-3.55, p<.01). We also removed any outliers which may have been unrepresentative of the sample. We still found significant results in that pro-Ana blog entries (M=.04) had more words regarding purity/virtue than the medical blogs when not assuming equal variances (M=0) (t (76)=-4.29, p<.01). We also found that pro-Ana blog entries (M=.14) included more language in regards to purity/vice than medical blog entries (M=.05) in general when not assuming equal variances (t (118.77)=-3.03, p<.01). We found that pro-Ana poetry (M=.07) had more words pertaining to fairness/virtue than mental health poetry (M=.01) when not assuming equal variances (t(52.05)=-2.41, p=.02). We also found that pro-Ana poetry (M= .25) included more words regarding purity/vice than mental health poetry (M=.09) on average when not assuming equal variances (t (65.31)=-2.80, p<.01). Furthermore, we observed that authors of pro-Ana poetry (M=.34) included more words pertaining to harm/vice than pro-Ana blogs (M=.14) on average when assuming equal variances (t (114)=-2.52, p<.01).
My conscience hath a thousand several tongues, and every tongue brings in a several tale.
William Shakespeare (Richard III)
Discussion
Overall, we found several significant differences between the pro-Ana groups and their controls. In particular, we saw a trend that pro-Ana authors were more likely to use more language referring to purity/vice. We believe this may have to do with the fact that victims who suffer from anorexia nervosa are more likely to have feelings of disgust in regards to food and their own self-concept. Troop & Baker (2009) have noted that when significant correlations are found between disgust and eating disorder symptoms, there are correlations after taking out effects of depression, anxiety, fear, and neuroticism (p. 236). It is also interesting that Murray and colleagues (2008) found no difference between patients with and without eating disorders in regards to recognizing facial displays of emotions, but patients with eating disorders were more sensitive to recognition of disgust in facial expressions (as cited in Troop & Baker, 2009, p. 242). Perhaps, persons struggling with anorexia nervosa are more likely to focus upon the consequences of not being pure. However, that being said, we find it interesting that the pro-Ana blog entries were more likely to refer to purity with positive language than the medical blogs. This result could be due to the fact that pro-Ana bloggers care about purity more than the medical bloggers. Giles (2006) noted that many pro-Ana users in his study spoke of “ana” being a practice of purity, and saw anorexics as having a higher moral ground due to their self-discipline from dieting (p.468). Nevertheless, the pro-Ana bloggers’ mean percentage of words pertaining to purity/virtue was extremely small (M=.03), and it is possible that this significant result happened due to not having enough samples of medical blogs.
Some other of our significant results also had very small means, which also makes us wary of drawing conclusions from the different sample groups. For instance, medical blog entries had a mean percentage of .09 for language pertaining to ingroup/vice, while pro-Ana blog entries had a mean of .01. It is possible that some words pertaining to ingroup/vice such as individual may have come up a couple of times for the medical blogs, but only once or twice for the pro-Ana blog entries. While the medical blog entries may have used individual more than the pro-Ana groups, that might not necessarily mean that they valued ingroup loyalty more than the pro-Ana users. Hopefully, future studies will verify whether our results were legitimate.
Limitations
The main limitation of our study is that the pro-Ana blog and poetry samples could be from the same user and or users with a different alias. It is possible we are not getting an accurate representative sampling of the pro-Ana community. Moreover, the 500 or 300 words with which we have taken from each blog and poetry page may not be an accurate representation of blogs have thousands of words and the most recent entries or poems may simply be anomalies from the rest of the text. In the future, we hope we can take pro-Ana poetry and blog entries in their entirety and use them as samples which would be more likely to be representative of the pro-Ana population.
References
-
Bardone-Cone, A. M. and K. M. Cass, 2006. "Investigating the impact of pro-anorexia websites: a pilot study. “ European Eating Disorders Review 14:256-262.
-
Borzekowski, D. L. G., Schenk, S., Wilson, J.L., and R. Peebles, 2010. "e-Ana and e-Mia: A Content Analysis of Pro-Eating Disorder Web Sites." American Journal of Public Health100:1526-1534.
-
Brynolf, M.T. 2009. "Innovations in Clinical Practice: A 21st Century Sourcebook." in Pro-ana: Analysis and implications for treatment of anorexia nervosa vol. 1, edited by J. B. Allen, E. M. Wolf, and L. VandeCreek. Sarasota, FL: Professional Resource Press.
-
Custers, K., and Van den Bulck, J. 2009. "Viewership of pro-anorexia websites in seventh, ninth and eleventh graders." European Eating Disorders Review 17:214-219.
-
Giles, D. 2006. "Constructing identities in cyberspace: The case of eating disorders." British Journal of Social Psychology 45:463-477.
-
Graham, J., J. Haidt, and B. A. Nosek. 2009. "Liberals and conservatives rely on different sets of moral foundations.“ Journal of Personality and Social Psychology 96.
-
Griffin, J. and E. M. Berry. 2003. "A modern day holy anorexia? Religious language in advertising and anorexia nervosa in the West." European Journal of Clinical Nutrition 57:43.
-
Harper, K., Sperry, S., and J. K. Thompson. 2008. "Viewership of pro-eating disorder websites: Association with body image and eating disturbances." International Journal of Eating Disorders 41:92-95.
-
Harshbarger, J.L., Ahlers-Schmidt, C.R., Mayans, L., Mayans, D., and J.H. Hawkins. 2009. "Pro- anorexia websites: What a clinician should know." International Journal of Eating Disorders 42:367-370.
-
Schwartz, L. 2009. "Arresting Ana." Pp. 28 minutes. U.S.: Women Make Movies.
-
Troop, N. and A.H. Baker. 2009. "Food, body and soul: The role of disgust in eating disorders." In Disgust and Its Disorders, edited by B. O. Olatunji and D. McKay. Washington D.C.: American Psychological Association.