The Lemming Effect as a Mechanism of Individual Change in Group Treatments

College-aged women are at an increased risk for the development of subclinical levels of eating disordered symptoms, which have been correlated to lasting deleterious cognitive, physical, and academic effects. The Body Project (TBP) is a universal group-based eating disorder prevention program that targets undergraduate women and challenges thin ideal messages through cognitive dissonance. Burlingame, Strauss, and Joyce (2013) in a meta-analysis of group treatments proposed five factors that independently and congruently work to promote individual change in group treatments: formal change theory, patient characteristics, leader characteristics, structural factors, and small group processes. A host of literature within TBP exists surrounding four of these factors, however, little research has been conducted on the small group processes that moderate individual change within this eating disorder prevention program.
The current study was designed to replicate and extend previous findings on the prevalence of the lemming effect within TBP, as well as examine how the lemming effect is related to outcome of treatment at a 3-month follow-up. Thirty-two participants aged 18-24 were examined. Groups ranged from 3 to 21 participants, including peer leaders. Twenty-nine audio recordings of session one of TBP were coded for lemming effects by the main research, and ten were coded by blind raters for inter-rater reliability measures. Three scales, the Ideal Body Stereotype Scale-Revised (IBSS-R), the Body Parts Satisfaction Scale-Revised (BPSS-R), and the Eating Disorder Examination Questionnaire (EDE-Q), were used to measure levels of thin-ideal internalization, body satisfaction, and frequencies of eating disordered (ED) behaviors, respectively.
Partial correlations revealed nonsignificant relationships between the number of lemming effects and the change in thin-ideal internalization and body satisfaction from baseline to follow-up. Additionally, a reliable change index revealed that the majority of change from baseline to follow-up was reliable for the IBSS-R, and the majority of change for the BPSS-R was unreliable. Lastly, chi-square tests of independence revealed nonsignificant relations between the number of lemming effects and change in ED behaviors.
Due to the small sample and lack of findings, future research would benefit from including a larger sample. This would enable larger power to detect effects and allow for more thorough statistical analyses to be performed to compare the relation of lemming effects to changes in outcome. However, this was the first study to look at the lemming effect variable as a small group process within TBP and added to the growing literature on how small group processes result in efficacious outcomes of treatment within group treatments.