TABLE 3.

Mean within-meta-analysis percentiles of point estimates ($\overline{Q}$) overall, by journal tier, and by a study's status as one of the first three (“early”) published versus as one of the later studies published

Group	n	${\overline{\mathit{Q}}}_{\text{higher}-\text{tier}}$	${\overline{\mathit{Q}}}_{\text{lower}-\text{tier}}$	${\overline{\mathit{Q}}}_{\text{early}}$	${\overline{\mathit{Q}}}_{\text{later}}$
PLoS One	3636			0.46 [0.35, 0.58]	0.51 [0.49, 0.54]
Top medical	558	0.40 [0.26, 0.55]	0.52 [0.46, 0.58]	0.40 [0.08, 0.72]	0.52 [0.47, 0.57]
Top psychology	7501	0.50 [0.47, 0.54]	0.51 [0.49, 0.52]	0.56 [0.44, 0.69]	0.51 [0.49, 0.52]
Metalab	799	0.61 [0.52, 0.71]	0.49 [0.46, 0.52]	0.62 [0.43, 0.80]	0.49 [0.46, 0.52]

Note: PLoS One was omitted from journal tier analyses. n: number of point estimates in group (which may exceeded number analyzed due to missing data).