This video illustrates how to perform and interpret a multiple regression statistical analysis in SPSS.
YouTube Channel (Quantitative Specialists): https://www.youtube.com/user/statisticsinstructor
Inferential course: https://www.udemy.com/inferential-statistics-spss
Descriptives course: https://www.udemy.com/descriptive-statistics-spss
Questionnaire/Survey & Likert Course: https://www.udemy.com/survey-data
ANOVA course: https://www.udemy.com/anova-spss
MANOVA course: https://www.udemy.com/manova-spss
Video Transcript: In this video, we'll take a look at how to run a multiple regression in SPSS. And on your screen as an example we have four variables SAT score, social support, gender, and college GPA. And in this example we're using the first three variables SAT score, social support, and gender, to predict first year college GPA. And here SAT score was taken in high school, social support is a measure of how much support a student felt that they received from others, where higher scores indicate greater support, and that was taken in the first year in college, and then gender, our dichotomous variable, where 1 is male and 2 is female, and the variable, college GPA, was the GPA after the first year in college. And in regression what we're trying to predict in this case, college GPA, is known as our criterion variable. It's also known as the dependent variable (DV). And then the variables that we're using to predict the criterion variable, SAT score, social support, and gender, those are known as are predictors or predictor variables, and we also refer to those as independent variables (IV). And those once again are SAT score, social support, and gender. Now in multiple regression you always have one criterion or dependent variable, and for it to be multiple regression you have to have two or more predictors or independent variables. if you just had one predictor or independent variable, such as SAT score, then that would be simple regression. But since we have two or more, in this case we have three once again, we're doing multiple regression. OK so to run multiple regression SPSS we want to go to Analyze, and then Regression and then go ahead and select Linear. And here we want to move college GPA to our Dependent box and then we want to select all the predictors and move those to our Independent(s) box. And then go ahead and click OK. And our output opens here and the first table, Variables Entered/Removed, this confirms that we had the variables gender, SAT score, and social support as our predictors, and then our dependent variable, or criterion variable, was college GPA, so that looks good. OK our next two tables, Model Summary and ANOVA, these two tables, they're looking at whether are predictors, once again, SAT score, social support, and gender, when those are taken together as a set or as a group, do they predict college GPA. And the Model Summary and ANOVA table are getting that slightly different things, but they're very closely related. So let's go ahead and start with Model Summary and take a look at that. So for Model Summary in this video we're going to focus on R square and then in another video we'll talk about these measures in more detail. But for this general overview the most commonly reported value in the Model Summary table is the R square value. And R squared, if I round this to two decimal places and then convert it to a percentage, so this would round two .50 or 50%, I could interpret R squared as follows. R squared once again is equal to .50 and then taken as a set the predictors SAT score, social support, and gender, account for 50% of the variance in college GPA. OK so R squared is a measure of the amount of variance in the dependent variable that the independent variables or predictors account for when taken as a group. And that's very important, it doesn't measure how much a given individual predictor accounts for, but only when we take them all as a group, this Model Summary table says overall, the regression model, which is what is referred to sometimes as a model, these three predictors predicting college GPA, that overall model accounts for 50% of the variance. Which is pretty good in practice. OK next we have our ANOVA table