Many Labs 2: Investigating Variation in Replicability Across Samples and Settings

Abstract

We conducted preregistered replications of 28 classic and contemporary published findings, with protocols that were peer reviewed in advance, to examine variation in effect magnitudes across samples and settings. Each protocol was administered to approximately half of 125 samples that comprised 15,305 participants from 36 countries and territories. Using the conventional criterion of statistical significance ( p < .05), we found that 15 (54%) of the replications provided evidence of a statistically significant effect in the same direction as the original finding. With a strict significance criterion ( p < .0001), 14 (50%) of the replications still provided such evidence, a reflection of the extremely high-powered design. Seven (25%) of the replications yielded effect sizes larger than the original ones, and 21 (75%) yielded effect sizes smaller than the original ones. The median comparable Cohen’s ds were 0.60 for the original findings and 0.15 for the replications. The effect sizes were small (< 0.20) in 16 of the replications (57%), and 9 effects (32%) were in the direction opposite the direction of the original effect. Across settings, the Q statistic indicated significant heterogeneity in 11 (39%) of the replication effects, and most of those were among the findings with the largest overall effect sizes; only 1 effect that was near zero in the aggregate showed significant heterogeneity according to this measure. Only 1 effect had a tau value greater than .20, an indication of moderate heterogeneity. Eight others had tau values near or slightly above .10, an indication of slight heterogeneity. Moderation tests indicated that very little heterogeneity was attributable to the order in which the tasks were performed or whether the tasks were administered in lab versus online. Exploratory comparisons revealed little heterogeneity between Western, educated, industrialized, rich, and democratic (WEIRD) cultures and less WEIRD cultures (i.e., cultures with relatively high and low WEIRDness scores, respectively). Cumulatively, variability in the observed effect sizes was attributable more to the effect being studied than to the sample or setting in which it was studied.

Voice Patterns in Schizophrenia: A systematic Review and Bayesian Meta-Analysis

“…Schizophrenia is a heterogeneous disorder, and indeed several studies attempted to more specifically investigate the relation of acoustic features with the symptomatology of the disorder. However, given the limited meta-analytic effect sizes and the awareness that replications tend to show a marked shrinkage of effect sizes (Klein et al, 2018), we need to move beyond small heterogeneous studies. The majority of the studies analyzed include between 20 and 30 patients, plausibly due to the difficulty in accessing clinical populations.…”

Enhancing Statistical Inference in Psychological Research via Prospective and Retrospective Design Analysis

“…As mentioned at the beginning of this paragraph, the Study 2 of Kay et al (2014) was selected for illustrative purposes and in a constructive perspective. For a more comprehensive picture, we invite interested readers to consult the "Many Labs 2 project" (Klein et al, 2018), which showed that with a large sample size (n = 6506) the original conclusion of Study 2 cannot be supported (i.e., t(6498.63) = −0.94, p = .35, d = −0.02, 95%CI = [−0.07, 0.03], as well as the subsequent response of the original authors (Laurin et al, 2018).…”

Practical Methodological Reform Needs Good Theory

“…Beyond providing guidance in the selection of replication projects, theory is absolutely essential when designing meta-science projects that purport to address significant theoretical claims. One example here is Many Labs 2 (Klein et al, 2018), which sought to ask whether sample source was a moderator of experimental effects. It provided a nice test of sampling variability for an idiosyncratically selected subset of effects, and perhaps surprisingly revealed that -at least for the tasks chosen -results were fairly consistent across volunteer sites.…”