- Analysis of covariance (ANCOVA) compares several means adjusted for the effect of one or more other variables (called covariates); for example, if you have several experimental conditions and want to adjust for the age of the participants.
- Before the analysis check that the independent variable(s) and covariate(s) are independent. You can do this using ANOVA or a t-test to check that levels of the covariate do not differ significantly across groups.
- In the table labelled Tests of Between-Subjects Effects, look at the column labelled Sig. for both the covariate and the independent variable. If the value is less than .05 then for the covariate it means that this variable has a significant relationship to the outcome variable; for the independent variable it means that the means are significantly different across the experimental conditions after adjusting them for the covariate.
- As with ANOVA, if you have generated specific hypotheses before the experiment use planned comparisons, but if you don’t have specific hypotheses use post hoc tests. Although SPSS will let you specify certain standard contrasts, other planned comparisons will have to be done by analysing the data using the regression procedure in SPSS.
- For parameters and post hoc tests, look to the columns labelled Sig. to discover if your comparisons are significant (they will be if the significance value is less than .05). Use bootstrapping to get robust versions of these tests.
- In addition to the assumptions in Chapter 5, test for homogeneity of regression slopes. This has to be done by customizing the ANCOVA model in SPSS to look at the independent variable × covariate interaction.
© SAGE Publications Ltd.
