> "It is unclear whether their statistical model will 'find' social contagion in every outcome they examine because of the limitations," Jason M. Fletcher of Ya

> "It is unclear whether their statistical model will 'find' social contagion in every outcome they examine because of the limitations," Jason M. Fletcher of Yale University wrote in an e-mail. He and a colleague conducted a similar analysis using data from a large federal survey to show that acne, headaches and even height could appear to be spread through social networks if not analyzed properly. The inclusion of this caveat makes this an above-average piece of science reporting.