Table 1. Meta-analytic results for all experiments and for independent replications of Bem’s experiments.
Number of
experiments |
Number of
participants |
Effect size
(Hedges’ g) |
95%CI or
Crl |
Combined
z or
Bayes factor |
p
(One-tailed) |
I 2 | τ 2 | |
---|---|---|---|---|---|---|---|---|
All experiments
a
Bayesian analysis |
90 | 12,406 | 0.09
0.08 |
[0.06, 0.11]
[0.02, 0.15] |
z = 6.33
BF = 5.1×10 9 |
1.2 × 10 -10 | 41.4 | .005
.028 |
Independent replications
b
Bayesian analysis |
69 | 10,082 | 0.06
0.07 |
[0.03, 0.09]
[0.01, 0.14] |
z = 4.16
BF = 3,853 |
1.2 × 10 -5 | 36.1 | .004
.035 |
Exact replications
Modified replications |
31
38 |
2,106
7,976 |
0.08
0.05 |
[0.02, 0.13]
[0.02, 0.09] |
z = 2.90
z = 3.00 |
.0018
.0013 |
31.7
38.9 |
.007
.004 |
Pre-2011 replications
Post-2011 replications |
30
39 |
2,193
7,889 |
0.09
0.05 |
[0.04, 0.15]
[0.02, 0.08] |
z = 3.20
z = 2.88 |
.0007
.004 |
39.5
32.3 |
.009
.003 |
Peer reviewed
Not peer reviewed |
35
34 |
7,477
2,605 |
0.06
0.06 |
[0.02, 0.10]
[0.02, 0.10] |
z = 2.93
z = 3.21 |
.0017
.0007 |
51.4
8.7 |
.001
.006 |
Note. In a Bayesian analysis, the analogue to the 95%CI is Crl, “credible intervals of the posterior distributions.” I 2 is an estimate of the percent of variance across studies due to differences among the true effect sizes. τ 2 is the between-studies variance.
a Assuming a null ES of .01 and a variance of .005 (the observed variance, τ 2, in the random-effects model), the statistical power of this meta-analysis is 0.95 ( Hedges & Pigott, 2001).
b These analyses exclude Bem’s own experiments and the eleven experiments that had not been designed as replications of those experiments.