Table 4. Linear regression analyses adjusted for socio-demographic variables^a.

Step 1. The association between impulsivity traits and children's BMI z-scores
	Total sample (n = 1377)	Subsample of non-overweight children (n = 1199)	Subsample of overweight children (n = 178)
	β	β	β
Reward sensitivity	0.01	0.02	−0.01
Disinhibition	0.04	0.03	−0.04
Step 2. Association between impulsivity traits and children's snack consumption
	Total sample	Subsample of non-overweight children	Subsample of overweight children
	β	β	β
Reward sensitivity	0.02	0.02	0.03
Disinhibition	0.05*	0.05	0.13
Step 3. Association between snack consumption and children's BMI z-scores
	Total sample	Subsample of non-overweight children	Subsample of overweight children
	β	β	β
Snack consumption	−0.10***	−0.06*	−0.02

β =  standardized regression coefficient.

^* p<0.05;

^** p<0.01;

^*** p<0.001.

^aAll regression analyses adjusted for children's age, gender, ethnicity (native Dutch vs. non-native Dutch) and socioeconomic status.