In his blog Data Colada, URI SIMONSON offers an hypothesis why a recent paper in the AER, Deming, Yuchtman, Abulafi, Goldin, & Katz (2016), failed to replicate Bertrand & Mullainathan’s (2004) famous paper on racial discrimination in labor markets.  The latter study focused on callback rates when CVs were mailed out using “black” versus “white” names.  B&M found that callback rates were lower for blacks.  DYAG&K found no difference.  Simonson argues that socioeconomic status (SES) may be a confounder.  In particular, B&M’s “black names” were also perceived to be low SES, while DYAG&K’s names arguably held SES constant.  So the final answer is…Yet To Be Determined.  To read more, click here.