The file drawer would need 558 null studies to overturn the priming effect.

Evidence

Using the procedure suggested by Rosenthal and Rosnow (2008), we found there must be 558 studies put away in file drawers that, if added to the sample of studies we analyzed, will produce an overall nonsignificant priming effect. — Alexandra Snell, Miron Zuckerman, and Bonnie M. Le, Does Religious Priming Induce Greater…, 79

Categorization

• SHARE: A concrete, striking robustness number (558 hidden null studies needed) for a colleague doing meta-analytic work on religious priming specifically.
• XREF: connects to: File-drawer tests assume uncovered unpublished studies form a random sample. — Thomas F. Pettigrew, Linda R. Tropp, A Meta-Analytic Test of Intergroup…, 50

Broader Connections

• File-drawer tests assume uncovered unpublished studies form a random sample. #qualifies (0.82) — Thomas F. Pettigrew, Linda R. Tropp, A Meta-Analytic Test of Intergroup…, 50