TABLE 2

Results of Linear Regression Models Predicting BMI by Type of Screen Media Use

TelevisionVideo GamesComputer
Model 1Model 2Model 1Model 2Model 1Model 2
BSEβBSEβBSEβBSEβBSEβBSEβ
Age0.180.820.020.530.790.070.050.810.010.050.820.010.140.810.020.170.810.02
Female gender1.681.170.151.771.120.161.461.180.131.341.250.121.691.160.151.421.210.13
Physical activity (h)−0.590.28−0.22b−0.670.28−0.25b−0.600.28−0.22b−0.600.29−0.23b−0.660.29−0.25b−0.690.29−0.26b
Duration (h)a−0.100.13−0.08−0.260.14−0.20−0.190.23−0.09−0.140.32−0.06−0.240.18−0.14−0.160.21−0.09
Percent primary attentiona0.130.040.32c−0.020.08−0.05−0.060.07−0.10
Percent secondary/tertiary attentiona−0.030.05−0.050.060.250.03−0.020.11−0.02
Adjusted R2.05.13.05.03.06.05
  • Model 1 tests the association between the duration of the media use listed in the column and BMI adjusted for the other listed variables. Model 2 adds primary and secondary/tertiary attention to model 1.

  • a Duration and attention variables correspond to use of the single medium type identified with the column header.

  • b P < .05.

  • c P < .01.