Second, we examined the types of foods consumed during
television viewing, the differences between the types of foods
consumed with the television on or off, and the associations
between the types of foods consumed during television viewing
and children’s weight status. The types of foods consumed with
the television on or off were described by the percentage of
energy from fat, the energy density (kilojoules per gram of food
or beverage containing1.20 kJ/serving), and the percentage of
energy that each of the 6 food categories contributed to the total
amount of energy consumed with the television either on or off.
To assess the relation between the types of foods consumed
during television viewing and children’s weight status, Spearman
correlations were calculated between children’s BMI and
both the percentage of energy from fat consumed during television
viewing and energy density of foods consumed during television
viewing for the subsample of children who consumed any
food while watching television. Finally, two-factor repeatedmeasures
analysis of variance was used to examine differences in
the types of foods consumed with the television on or off and in
the types of foods consumed on weekdays and weekend days and
the interaction between the types of foods consumed with the
television on or off on weekdays and weekend days. All analyses
were conducted by using SAS version 8.02 (SAS Institute Inc,
Cary, NC).