Levene's test of error variances shows a significant difference in the variance of life span between wild-type and mutant lines in females (F9,377 = 4.086, p < 0.001), but not in males (F9,370 = 1.839, p = 0.060).
Bicondylar length, epicondylar breadth, anterior-posterior (AP) diameter, medial-lateral (ML) diameter and cortical area at the midshaft were measured by four observers to test the measurement error variance and observer agreement of the protocol (n = 6).
Levene test revealed the error variance of the dependent variables was equal across both groups, P ranged from 0.407 to 0.948.
Concerning model quality, one useful test is whether the forecast error variance in the testing data is approximately the same as that in the training data.
Using RT as dependent variable, we computed a 3 × 3 × 2 analysis of variance (ANOVA) with the factors group (LID, No‐LID, control), run, and task laterality (left, right) after testing equality of distribution of error variance (p = 0.902, Levene test).
†† Two outliers removed to not violate the Levene's test of equality of error variance.
To check for homogeneity, Levene's test of equality of Error Variance was performed.
For features that violated homogeneity of variance on Levene's test of equality of error variance, a more stringent α level was set at 0.01 (cf. [ 29]).
Accuracy rates were compared using a nonparametric Kruskal Wallis test, because Levene test showed that distribution of error variance was not equally distributed (p < 0.001).
Assumptions of linear regression were tested with Shapiro Wilk test (normality), partial residuals plots (linearity), Durbin Watson test (independence of errors), non-constant error variance (homoscedasticity), and variance inflation factor test (multicollinearity).
