Common voice: Analysis of behavior modification and content convergence in a popular online community

This study analyzes images and associated commentary from an online image-sharing community (www.imgur.com). We hypothesize that, in the presence of overt communication of social rules, site content will reflect a somewhat consistent, content convergence irrespective of who comments, given that an individual's social identity, rather than their individual identity, will dominate their online interactions. We began with a random sample of 5000 images, and we grouped those images into six categories. We then randomly selected 10 images from each category, and conducted a close qualitative analysis of the micro-comments submitted in response to those images. Our results suggest that there is an overt communication of behavioral standards through explicit behavior correction by other site members. Content convergence can also be observed both in responses to a single image, and more broadly in comment similarity across images and even categories. Cultural norms and resulting content convergence suggests that individuals may come to see themselves as representatives of the site when they post - and may modify their behavior accordingly. This may lead to the perception that a common voice appears throughout the site, which may have implications for the communication of social support over the Internet, and for more quantitative analyses.