Replication Data for: Countering Underproduction of Information Public Goods

Information public goods are susceptible to a risk called “underproduction”: a mis- alignment of the quality of the good and the demand for it. The practical con- sequences vary by the type of good. When the information public good is digital infrastructure, underproduced goods may be plagued by widescale security vulner- abilities, and when the information public good is detailed factual information, un- derproduced goods spread misinformation and ignorance. Yet contributors to in- formation public goods are often volunteers who choose their own tasks, and are not assigned to track consumer demand or to serve public needs. Although under- production has been shown to be a common feature of peer-produced information public goods, very little is known about how it can be counteracted. To understand whether contributor experience and limits on their personal identifiability might serve to counteract underproduction, we use a detailed longitudinal dataset from English Wikipedia. First, we show that more experienced editors tend to contribute to underproduced articles. Second, we show that this tendency to contribute in ways that counter underproduction is still present although generally weaker among those who contribute without an account. Third, we use a within-person analysis to show that contributors tend to shift toward underproduced articles over time. These find- ings illustrate the value of retaining peer production contributors, including those contributing without accounts, as a means to counter underproduction.