Variables, Dependent

A dependent variable is dependent upon the presence or absence of an independent variable. The dependent variable is what the researcher is trying to measure or explain, and is the object of the research; thus, it is sometimes called the outcome variable. Dependent variables are most often represented in quantitative research where the focus is on using defined variables to measure outcomes. This entry describes the nature of the relationship between dependent and independent variables, offers examples of such relationships, and indicates types of statistical tests used to examine dependent variables.

A researcher’s goal is to determine if and to what extent changes in one variable influence certain types of changes in another, the dependent variable. A researcher will manipulate independent variables to see how dependent variables respond; the dependent variable represents the measureable outcome of this manipulation. A researcher will measure the dependent variable to determine whether and how much it changes. The researcher’s goal is to accurately predict how the dependent variable will change in the presence of the independent variable. To do so, the researcher must find a significant relationship between independent and dependent variables, represented by correlation or causal claims. In graphical representations of these relationships, the dependent variable is generally put on the y-axis.

For example, the amount of sleep an individual gets before a test could influence the grade a student receives on that test. In this case, the researcher is looking to see how much the test grade depends on the amount of sleep. It is important to remember that a variable isolated as dependent in one research study is not necessarily a dependent variable in other research designs. In the example, the test grade was dependent on the amount of sleep. However, if a researcher wanted to know if test grades influenced self-esteem, the test grade would become an independent variable and level of self-esteem dependent on the grade. Research design determines which variables are independent and which variables are dependent.

To measure dependent variables, researchers must determine the type of variable (e.g., ordinal, interval) and determine appropriate statistical tests. Researchers looking for relationships between one independent and one dependent variable, such as the sleep and test score example, have a number of options for testing the significance of the relationship. Some research designs [Page 1845]are more complex and feature one or more independent or dependent variables. For example, some research designs call for a measure of significant differences between the effect of two categorical independent variables and one categorical dependent variable, so a researcher might use a t-test. Sometimes research designs call for researchers to analyze differences in a dependent variable between two or more groups. To make such comparisons, the researcher would use analysis of variance (ANOVA) procedures, such as one-way ANOVA to analyze one categorical independent variable among three or more groups with one continuous dependent variable. Or, a researcher could use multivariate ANOVA (MANOVA) to determine how one or more independent variables influences one or more dependent variables. Finally, a researcher looking to estimate how much of the variance in a dependent variable is accounted for by variance in a predictor variable would use multiple regression. There are numerous statistical tests a researcher can use to determine the relationship between independent and dependent variables; the appropriate test depends on the specifics of the research design.

...