features of a replication analysis as closely as possible to features of the original published study. The original published study used face-to-face clinical interviews. Thus, the protocol plans to include studies that fail to match the design of the original publication in the key area of measurement, and most such studies have very large Ns. Moreover, as noted above, large-N studies are even more unsuitable if their designs do not allow establishing clear temporal order between hypothetical cause and outcome. The protocol’s over-emphasis on sample size of individual studies, coupled with exclusion of many well-designed studies for testing the hypothesis, is misguided.
