Drawing Conclusions from Non-Random Samples: A Comment on “Race and Art: Prices for African American Painters and their Contemporaries” by Richard Agnello

In a recent paper in the Journal of Black Studies, “Race and Art: Prices for African American Painters and Their Contemporaries,” economist Richard Agnello examines price differentials between paintings sold at auction from 1972 to 2004 for a set of 16 African American artists, and a group of white artists identified as similar contemporaries. In this short paper we examine Agnello’s control group of white artists and confirm that this is a non-randomly selected sample of artists that are, on average, quantifiably more famous than the black artists in Agnello’s treatment group. In light of this selection bias, we discuss the difficulty of disentangling the effects of race and fame on auction prices.