By construction, this procedure creates two (possibly highly) imbalanced genetic groups that subsequently need to be compared in terms of mean or median trait differences.
Statistical differences between mean trait values of iLID mice and control mice were assessed using one-way ANOVA and Bonferroni post-hoc tests (p<0.05, Statistica 6.0, Statsoft, Tulsa, OK, USA).
False positives increased when items were very easy to endorse and when there with mode differences in mean trait level.
Though differences in mean trait levels were examined, deviations from prior assumptions concerning parameter variances or distribution types were not examined.
By calculating the fitness for each trait value at demographic equilibrium and weighing the distribution of trait values by their fitness, selection differentials are computed as the difference in mean trait value before and after selection (Fig. 1).
Though not significant at the univariate level, mean difference in mean trait level by mode of 1 logit was significantly and inversely related to true mode DIF detection in the multivariate model.
For the CrI procedure, size of DIF and difference in mean trait level by mode significantly and positively predicted false-positive DIF results, whereas item difficulty was significantly and inversely associated with false positive mode DIF.
