Supplementary Figure 18: The distribution of the reported p-values throughout the manuscript
From: Altered responses to social chemosignals in autism spectrum disorder
We take a meta-analysis approach to the experiments we conducted. In order to verify that our reported p values in the manuscript are not selective and to test whether our findings across experiments reflect true effects, we applied p-curve analysis (1). We found that the p-curve of all our experiments combined is significantly right skewed (Full p-curve, Z = −7.24, p < 10−5, Half p-curve, Z = −5.88, p < 10−5), indicating overall evidential value of the reported results. 1. Simonsohn, U., Nelson, L. D. & Simmons, J. P. P-curve and effect size: Correcting for publication bias using only significant results. Perspectives on Psychological Science 9, 666–681 (2014).
