Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
This paper analyzes the implications of attrition for the internal and external validity of the results of four randomized experiments and proposes a new method to correct for attrition bias. We find that not including those found during the intensive tracking can lead to a substantial overestimation or underestimation of the intention-to-treat effects, even when attrition without such tracking is balanced. We propose to correct for attrition using inverse probability weighting with estimates of weights that exploit the similarities between missing individuals and those found during an intensive tracking phase.