File Drawer Problem

The file drawer problem is the threat that the empirical literature is biased because nonsignificant research results are not disseminated. The consequence of this problem is that the results available provide a biased portrayal of what is actually found, so literature reviews (including meta-analyses) will conclude stronger effects than actually exist. The term arose from the image that these nonsignificant results are placed in researchers’ file drawers, never to be seen by others. This file drawer problem also has several similar names, including publication or dissemination bias. Although all literature reviews are vulnerable to this problem, meta-analysis provides methods of detecting and correcting for this bias. This entry first discusses the sources of publication bias and then the detection and correction of such bias.