Fig. 1.

Data pooling sample characteristics. a Total number of participants with ASD by sex for each contributing site ordered as a function of sample size (labelled alphabetically, see Table 1 for label key). The same site labels are used for (b–f). b–f are ordered by median sample statistic per site. b Violin plot of chronological age in years for all individuals per site. c Distribution of nonverbal IQ scores per site. Short-dashed line NVIQ for ADI-R datasets, long-dashed line NVIQ for ADOS datasets. Solid black lines indicate median NVIQ per site. d–f Tukey’s box-whiskers plots overlaid with scatterplots of individual data points per site for (d) ADOS Calibrated Severity Scores (CSS) Total, e ADI-R Social scores (ever/diagnostic) and f ADI-R Social scores (current)