Mind your Ps and Qs

Little is known about the impact of politeness in online communities. We use an inductive approach to automatically model linguistic politeness in online discussion groups and determine the impact of politeness on desired outcomes, such as increased reply rates. We describe differences in perceived politeness across a variety of groups and find that, controlling for group norms of responsiveness and message length, politeness increases reply rates in some technical groups, but rudeness is more effective in some political groups. The perceived politeness scores will be used to validate linguistic politeness strategies from theory and to inform the creation of a machine learning model of linguistic politeness that can be applied as a "politeness checker" to educate newcomers to write in ways likely to elicit response from specific communities or as a rudeness detection tool for moderators.