Table 2

Results from the regression models predicting adolescents' physical activity

Parameters from each regression model

β (SE)

R² for model

Main effect of risk

-.37 (.17)**

.23**

Main effect of maternal support

.12 (.10) **

Interaction between risk and maternal support

.23 (.18) **

Main effect of risk

-.31 (.17) **

.26**

Main effect of paternal support

.25 (.12) **

Interaction between risk and paternal support

.14 (.18) †

Main effect of risk

-.33 (.17) **

.24**

Main effect of sibling support

.16 (.12) **

Interaction between risk and sibling support

.21 (.18) **

Main effect of risk

-.25 (.17) **

.33**

Main effect of peer support

.34 (.11) **

Interaction between risk and peer support

.22 (.17) *

Note: † p < .10 * p < .05 ** p < .01

High risk coded as 1, low risk coded as 0.

The beta-weight for risk is slightly different across models because a different source of support was included in each model. The resulting beta weight is the independent effect of risk.

Davison and Schmalz International Journal of Behavioral Nutrition and Physical Activity 2006 3:5   doi:10.1186/1479-5868-3-5