Fig. 4: p-curve mixture models show robustness to effect size heterogeneity under the alternative hypothesis. | Communications Psychology

Fig. 4: p-curve mixture models show robustness to effect size heterogeneity under the alternative hypothesis.

From: Bayesian p-curve mixture models as a tool to dissociate effect size and effect prevalence

Fig. 4

a Distributions of participant-level accuracies under the alternative hypothesis, at various levels of between-participant heterogeneity. The mean of each distribution is 0.65. b False coverage rates of a 95% posterior highest-density interval (HDI) over 1000 simulations with a ground-truth population prevalence of 0.4.

Back to article page