Exchange: Confusing Categories, Shifting Targets

Findings in the social sciences too often turn out to be unstable because of the difficulty of and lack of incentives for replication. Bogaards’ initiative to reexamine existing findings on the role of elections in democratization is laudable in principle. Yet, replication is a delicate art and Bogaards attempt is not entirely cogent. The cited works are not accurately presented; the original indicators and dependent variables are shifted making comparison with the original findings difficult at best; and his own analysis is distorted by the use of muddled, partially overlapping categories.