file-drawer analysis
a statistical procedure for addressing the file-drawer problem by computing the number of unretrieved studies, averaging an effect size of .00, that would have to exist in file drawers before the overall results of a meta-analysis would become nonsignificant at p > .05, that is, would exceed an acceptable probability level (.05) of occurring by chance. A small computed value indicates a weak finding, that may just be due to chance. Conversely, a large value suggests that the finding is rather robust, as it would take a large number of nonsignificant findings to provide enough evidence to refute the results.