(Quasi-)Experimental Research in OSM

Byron South
Friday, 3:15pm – 20 min
Abhishek Nagaraj

How should we maximize user contributions in OpenStreetMap? What is the role of imports in affecting data quality and user contributions in the long run? Is OpenStreetMap community weaker or stronger in a country where commercial alternatives are popular? What is the best strategy to retain and encourage newcomers to the platform?

All of these questions are of deep interest to the OSM community and yet we do not have credible, data-driven and quantitative answers to these problems. In this talk, I will leverage my training in applied economics and computational social science at MIT and my 3-years of research using OSM data to explain how one might conduct quantitative research on OpenStreetMap using both field experiments as well as naturally occurring experiments. I will illustrate these methodologies using my own research and brainstorm some additional research questions and propose designs on how one might answer them.

I will end with a call to build strong capacity of computational social science research within OpenStreetMap (similar to Wikimedia Research) so that we can better design the platform for long-run sustainability.