The rationale given in the Culverhouse et al. protocol for exclusion of small studies is that more small studies have claimed positive findings. They note small-N studies run a risk of publication bias. Such bias emerges when a small-N study with a negative finding is more often “file-drawered” because it is not deemed rigorous enough to constitute decisive rejection of the null, whereas a small-N study with a positive finding would be more often published because it was able to reject the null despite being under-powered. However, the simple fact that more small studies have obtained positive findings does not by itself constitute evidence of such publication bias, particularly when there are systematic differences in quality between small studies and large studies. Moreover, it has been commented before that in relation to this particular GxE finding, both researchers and editors have been quite keen to publish negative findings (6 nagative reports have appeared in the last 3 years, all of which are taking part in the meta-analysis, although curiously most positive reports appearing over this time period have not been
