ABSTRACT
Analysis of Covariance (ANCOVA) can be used when people have a two-group pre-test/post-test design (e.g. comparing the impact of two different interventions, taking before and after measures for each group). ANCOVA is also handy when readers have been unable to randomly assign their participants to the different groups and instead have had to use existing groups (e.g. classes of students). ANCOVA assumes that covariates are measured without error, which is a rather unrealistic assumption in the majority of social science research. Some variables that reader may wish to control, such as age, can be measured reasonably reliably; others that rely on a scale may not meet this assumption. The impressive-sounding assumption requires that the relationship between the covariate and dependent variable for each of readers' groups is the same. The final assumption concerns the relationship between the covariate and the dependent variable for each of their groups.
