027, only very slightly higher than the first model. We note that for the second model, which includes an interaction term, the R 2 is. An applied researcher might want to develop a model with more explanatory variables to gain a better understanding of levels of satisfaction with the current state of the economy in the country. 0246 means that only about 2.5% of the variance in satisfaction with the economy is accounted for by the two variables. While detailed examination of these tables is beyond the scope of this example, we note that the R Square value in the second table measures the proportion of the variance in the dependent variable that is explained by the model. The first three tables in Figure 9 report the independent variables entered into Model 1 (the main effects model) and Model 2 (the model including an interaction term), some summary fit statistics for the regression models, and an analysis of variance for both models as a whole. We therefore have little concern about multicollinearity influencing this regression analysis.įigures 9 and 10 present a number of tables of results for both models that are produced by the multiple regression procedure in SPSS. In this case, the Pearson correlation coefficient between gndr and voter is. It is also useful to explore the possible correlation between your independent variables. The frequency distribution of the voter variable in Figure 8 shows that 71.5% of respondents voted in the last election compared with 28.5% of respondents who did not. The frequency distribution of the gender variable in Figure 7 shows that approximately 55% of respondents are female and 45% are male. Overall, there is little reason for concern as to the appropriateness of the variable for inclusion. It is possible to have three-way interactions or more, but we focus on the two-way case for ease of explanation.įigure 8: Frequency Distribution of Whether Respondents Voted or Not in the Last National Election, 2016 European Social Survey.įigure 6 shows a roughly normal distribution, with a peak at the lowest values. This example will focus on interactions between one pair of variables that are categorical in nature. An interaction can occur between independent variables that are categorical or continuous and across multiple independent variables. In a linear regression model, the dependent variables should be continuous. Focus is given instead to the difference in slopes which is described by the interaction coefficient.
![multiple regression spss multiple regression spss](https://www.statology.org/wp-content/uploads/2020/06/multRegSPSS1.png)
In a model including an interaction term, the slope estimates cannot be interpreted in the same way, as they are now conditional on other values. More attention is focused on the slope estimates because they capture the relationship between the dependent and independent variables. This requires estimating an intercept (often called a constant) and a slope for each independent variable that describes the change in the dependent variable for a one-unit increase in the independent variable. In a “main effects” multiple regression model, a dependent (or response) variable is expressed as a linear function of two or more independent (or explanatory) variables.
![multiple regression spss multiple regression spss](https://image.slidesharecdn.com/multipleregressioninspss-150404045801-conversion-gate01/95/multiple-regression-in-spss-1-638.jpg)
Interaction describes a particular type of non-linear relationship, where the “effect” of an independent variable on the dependent variable differs at different values of another independent variable in the model.