Examination for Professional Practice of Psychology (EPPP) Practice Test

When would a researcher choose to use stepwise multiple regression?

1. To predict performance based on criteria

2. To account for criterion variability with minimal predictors

3. To control for moderator variables in research

4. To categorize distinct criteria groups

The correct answer is: To predict performance based on criteria

A researcher would choose to use stepwise multiple regression primarily when the goal is to predict an outcome variable (performance, in this case) based on a set of predictor variables while automatically selecting the most significant predictors from a larger group. This technique helps in identifying which variables contribute meaningfully to predicting the dependent variable, allowing the researcher to simplify the model by including only those predictors that have a statistically significant relationship with the outcome. Stepwise multiple regression uses an algorithm to add or remove predictors from the model based on their statistical significance, which can be particularly useful when trying to optimize prediction without overfitting the model with unnecessary variables. This approach is particularly valuable in exploratory research where the relationships among variables are not well understood, enabling the researcher to build a model that efficiently captures the most relevant predictors. The other choices, while relevant to various statistical methods in regression analysis, do not align with the primary purpose of stepwise multiple regression. For example, accounting for criterion variability with minimal predictors focuses on reducing variance rather than solely on prediction. Controlling for moderator variables typically involves different analytical techniques, such as moderation analysis, rather than stepwise regression. Lastly, categorizing distinct criteria groups suggests a focus on classification rather than prediction, which is not the aim of step