Stats Lecture 14

Why not multiple t-tests?

increases the risk of Type I error, as each test carries a chance of incorrectly rejecting the null hypothesis

Post-hoc test

uses the omnibus error term from the ANOVA to determine which means are significantly different

When do we use post-hoc tests?

after an ANOVA when it indicates significant differences among group means, allowing researchers to identify specifically which groups differ from one another

ANCOVA

analysis of covariance; compares a factor and continuous response factor on a single continuous variable output

ANCOVA example

DV: health score; IV: education level and hours working per week

ANCOVA null hypothesis

the group means are equal AFTER controlling for the covariate

ANCOVA alternative hypothesis

the group means are not equal AFTER controlling for the covariate

MANOVA

multivariate analysis of variance; used to compare means of multiple dependent variables across groups

One-way MANOVA example

IV: education level; DV: insurance risk score and debt

Two-way MANOVA example

IV: education level and zodiac sign; DV: insurance risk score and debt

Sphericity

the condition where the variances of the differences between all possible parts of within-subject conditions are equal (the variability in the differences between any two groups is the same)