Analysis of variance (ANOVA) is a statistical method that allows the researcher to compare several different groups, rather than the t test that only compares two groups. The scores of a dependent variable are compared within groups of the independent variable. In our discussion of the t test, for example, we compared word recognition scores (dependent variable) within two groups of the independent variable “noise” (high noise versus low noise). Using ANOVA allows us to compare three or more independent variable groups. For example, we might want to compare subjects’ word recognition scores depending on their group membership in low, medium, or high noise conditions.

There are several variations of this test, but we focus primarily on the “one-way” ANOVA in this section. One-way ANOVA refers to the number of independent variables. In research, independent variables are known as factors. Therefore, if we have a research problem that has several groups of one independent variable, we can use one-way ANOVA to detect any differences on the dependent variable measure among the groups.

We do not have the space in this book to talk about variations of ANOVA, but there are several. Each kind is particularly designed to handle multiple independent variables (e.g., factorial ANOVA), designs that introduce control variables (ANCOVA), those that include multiple measures of the same subjects (within-subjects ANOVA), and those that include more than one dependent variable ...

Get Understanding and Applying Research Design now with the O’Reilly learning platform.

O’Reilly members experience live online training, plus books, videos, and digital content from nearly 200 publishers.