Table 4. Caliper test for exactly recalculated p-values per journal for different binwidths.

Binwidth	0.00125				0.0025				0.005				0.01
	x	N	Pr	p	x	N	Pr	p	x	N	Pr	p	x	N	Pr	p
All	1,404	2,808	0.5	0.508	2,808	5,761	0.487	0.973	5,761	11,824	0.487	0.997	11,824	25,142	0.47	>.999
DP	184	382	0.482	0.779	382	829	0.461	0.989	829	1,710	0.485	0.9	1,710	3,579	0.478	0.996
FP	30	69	0.435	0.886	69	172	0.401	0.996	172	376	0.457	0.956	376	799	0.471	0.955
JAP	73	145	0.503	0.5	145	270	0.537	0.124	270	556	0.486	0.765	556	1,168	0.476	0.952
JCCP	160	308	0.519	0.265	308	633	0.487	0.763	633	1,267	0.5	0.522	1,267	2,706	0.468	>.999
JEPG	81	164	0.494	0.593	164	332	0.494	0.608	332	683	0.486	0.778	683	1,535	0.445	>.999
JPSP	640	1,268	0.505	0.379	1,268	2,557	0.496	0.668	2,557	5,174	0.494	0.802	5,174	10,976	0.471	>.999
PLOS	125	260	0.481	0.752	260	541	0.481	0.828	541	1,170	0.462	0.995	1,170	2,544	0.46	>.999
PS	111	212	0.524	0.268	212	427	0.496	0.577	427	888	0.481	0.88	888	1,835	0.484	0.919

Notes.

x

frequency of p-values in .05 minus binwidth through .05

N

total frequency of p-values across both intervals in the comparison, Pr, x∕N

p

p-value of the binomial test

Significant results (α = .05, one-tailed) indicating excess of p-values just below .05 and are reported in bold.