Figure 1
Participants’ demographic characteristics as regards age and autistic symptomatology. ASRS, Autism Spectrum Rating Scales; SC, social/communication; UB, unusual behaviors; SR, self-regulation; TOT, total score, DSM-5 criteria; PS, peer Socialization; AS, adult socialization; SER, social/emotional reciprocity; AL, atypical language; ST, stereotypy; BR, behavioral rigidity; SS, sensory sensitivity; AT, attention. A repeated-measure two-way ANOVA was conducted to assess the effects of scales and groups (high- and low-ASRS) on T-scores (groups: F(1,30) = 94.50, p-value = 8.804e−11, η2G = 0.46). Mauchly’s test of sphericity was used to evaluate homogeneity of variances and co-variances. Because of violation of sphericity (p-value < 0.05), a Greenhouse–Geisser correction was performed (scales: FGGe(1,30) = 0.29, p-value = 2.374e−09, Groups:Scales: FGGe(1,30) = 0.36, p-value = 0.27). Also, normality of residuals was assessed and confirmed with Shapiro–Wilk test (W = 0.995, p-value = 0.170). Post-hoc analyses were performed by Tukey’s procedure. A sensitivity analysis was performed to assess the influence of outliers (i.e., values out of 1.5 times the interquartile range) on statistical significance. The ANOVA was conducted without outliers and was robust to their exclusion. For the “Age” variable, the statistical result was obtained using an independent sample t-test between high-ASRS and low-ASRS. Normality was assessed and confirmed with Shapiro–Wilk test (WLow-ASRS = 0.934, p-valueLow-ASRS = 0.07; WHigh-ASRS = 0.882, p-valueHigh-ASRS = 0.06). No skewness nor presence of outliers was outlined. Homogeneity of variances was confirmed (F = 0.657, p-value = 0.25). “***” p < 0.001, NS: Non significance.
