Motivated by research from the late 1990s and early 2000s, researchers often examine analyst performance as a function of individual analyst characteristics using variables representing analysts’ experience, busyness, and resources. Studies make different choices for variable construction and fixed effects, sometimes with varying results, suggesting there is no standard approach to evaluating the effect of analyst characteristics. We reproduce analyses from two early studies (Clement (1999) and Clement and Tse (2003)) and extend these analyses to a more recent sample period and across different methodological approaches and alternative analyst performance measures. We provide robust evidence that analyst experience is positively associated with earnings forecast accuracy but that associations for analyst resources and busyness are sensitive to variable measurement and fixed effect choices. Our results can inform accounting and finance scholars examining analyst performance as they consider which analyst characteristics to include and how to test the robustness of their findings.