Missing data are a common challenge facing empirical researchers. This paper presents a general GMM framework and estimator for dealing with missing values of an explanatory variable in linear regression analysis. The GMM estimator is efficient under assumptions needed for consistency of linear-imputation methods. The estimator, which also allows for a specification test of the missingness assumptions, is compared to existing linear imputation, complete data, and dummy variable methods commonly used in empirical research. The dummy variable method is generally inconsistent even when data are missing completely at random, and the dummy variable method, when consistent, can be less efficient than the complete data method.

Using a very large sample of matched author-referee pairs, we examine how referees' and authors' genders affect the referees' recommendations. Relying on changing author-referee matches, we find no evidence of gender differences among referees in charitableness, nor is there any effect of the interaction between the referees' and authors' genders. With substantial laboratory research showing gender differences in fairness, the results suggest that outside the laboratory, an ethos of objectivity can overcome possible tendencies toward same-group favoritism or opposite-group discrimination.