| Table 4. | ||
| OLS Regression of family involvement | ||
| Model 1 | Model 2 | |
| Enjoy | 0.56*** | 0.57*** |
| Race (White) | ||
| Black | 0.31 | |
| All Else | 1.49 | |
| R2 | 0.2112 | 0.2493 |
| Note. N=66; p<.05*, p<.01**, p<.001*** | ||
In the first regression model, the independent variable is the family enjoyment. The item that measured this is “How much did your family enjoy this activity?” with answer choices on a scale of 1-10 with “0=not at all” and “10=a great amount” The dependent variable in the first regression model is the level of family involvement. The item that measured this is “How engaged was your family in this activity (working together)?” with answer choices on a scale of 1-10 with “0=not at all” and “10=a great amount”. The model produced a regression coefficient of 0.56, meaning that with every one unit increase in enjoyment, family involvement increases 0.56 units. This coefficient was significant at the p<.001 level. The first regression model produced an R2 statistic of 0.2112, demonstrating 21.12% of the variation in family involvement is from the enjoyment variable.
In the second regression model, the independent variables are family enjoyment and race. Family enjoyment was measure the same as in the first model and race was measured using the item “Choose one or more races that you consider yourself to be” with answer choices “1=White or Caucasian (includes Latino/Hispanic)”, “2=Black or African American”, “3=American Indian/Native American or Alaska Native”, “4=Asian”, “5=Native Hawaiian or other Pacific Islander”, “6=Other”, and “7=Prefer not to say”. For the Race variable, we dummy coded it for the regression to run properly. We created categories of “White=1” and everything else=0; “Black=1” and everything else=0; and “All Else=1” and Black/White=0 to run the analysis.
The dependent variable is the level of family involvement and measured in the same way as in the first model. The second model produced a regression coefficient of 0.57 for enjoyment. This means that for every one unit increase in enjoyment, involvement increased 0.57 units. This coefficient was significant at the p<.001 level. For race, the regression coefficient for the “Black” group is 0.31, meaning for every one unit increase in the “Black” Race group, the family involvement increases 0.31 units. For the “All Else” group, the regression coefficient is 1.49, meaning that for every one unit increase in the “All Else” Race group, the family involvement increases 1.49 units. The second regression model produced an R2 statistic of 0.2493, demonstrating 24.93% of the variation in family involvement is from the enjoyment and race variables.