The cost of collaboration for code and art

In this paper, we use evidence from a remixing community to evaluate two pieces of common wisdom about collaboration. First, we test the theory that jointly produced works tend to be of higher quality than individually authored products. Second, we test the theory that collaboration improves the quality of functional works like code, but that it works less well for artistic works like images and sounds. We use data from Scratch, a large online community where hundreds of thousands of young users share and remix millions of animations and interactive games. Using peer-ratings as a measure of quality, we estimate a series of fitted regression models and find that collaborative Scratch projects tend to receive ratings that are lower than individually authored works. We also find that code-intensive collaborations are rated higher than media-intensive efforts. We conclude by discussing the limitations and implications of these findings.