Search
Program Calendar
Browse By Day
Browse By Time
Browse By Person
Browse By Room
Browse By Unit
Browse By Session Type
Search Tips
What to do in Chicago
Personal Schedule
Sign In
X (Twitter)
School psychology outcome variables do not often comply with typical assumptions of parametric analysis, such as homogeneity and normality. A central problem is that the outcomes in educational research are often counts or proportions, which follow alternative distributions best handled by different variants of Generalized Linear Models (GLiMs). We reviewed all articles in the top three school psychology journals in the last ten years, finding that in almost all cases researchers did not consider alternative GliM for count variables, and subsequently applied patchwork fixes or ignored the problem inappropriately given advances in analysis. We report on the findings of this literature review, introduce several GLiMs appropriate to school psychology, and conclude with several application examples.