Modeling Relationships of Multiple Variables with Linear Regression
Download 0.6 Mb. View original pdf
|
Stat Cheat Sheet
| Figure 7.2 Regression Output Interpreting the ANOVA F-test Although this table comes second in the output, the first statistics to look at are the F statistic and its significance value in the ANOVA table. This should look familiar to you—it’s the same type of ANOVA table we had in the one-way ANOVA test in Chapter 6. Ina regression model, the ANOVA F statistic tests whether the model as a whole is significant. In other words, do the independent variables, taken together, predict the dependent variable better than just predicting the mean for everything In the simple linear regression of this example, there is only one independent variable, so the F-test is testing if this one variable, the poverty rate, predicts the percent of births to teen mothers better than if we used the average teenage birth percentage to predict all states values. We would just use the average for all states if the relationship in Figure 7.1 were flat. The model clearly does better than a flat line—the p-value (in the Sig. column) is very low, less than .001. So there is less than a 1 in 1,000 chance that the relationship we found in this sample is actually best described by a flat line. Chapter 7 • Modeling Relationships of Multiple Variables with Linear Regression 166 Download 0.6 Mb. Share with your friends:
|