Item Response Theory Models and Spurious Interaction Effects in Factorial ANOVA Designs

摘要: In many psychological experiments, interaction effects in factorial analysis of variance (ANOVA) designs are often estimated using total scores derived from classical test theory. However, can be reduced or eliminated by nonlinear monotonic transformations a dependent variable. Although cross-over interactions cannot trans formations, the meaningfulness other hinges on achieving measurement scale level for which inappropriate (i.e., at least interval level). Classical do not provide according to contemporary item response theory (IRT). Nevertheless, rarely IRT models applied achieve more optimal properties and hence meaningful effects. This paper provides several condi tions under that scores, rather than trait misleading. Using asymptotic expecta...