. Author manuscript; available in PMC: 2022 Mar 1.

Published in final edited form as: Clin Pediatr (Phila). 2014 Dec 5;54(9):871–877. doi: 10.1177/0009922814561742

Table 2.

Regression Results Predicting Children’s Eighth-Grade Test Scores From Fifth-Grade Food Consumption.^a

Frequency of Fast Food Consumption at Fifth Grade	Eighth Grade Achievement Scores
	Reading			Math			Science
	b (SE)	95% Confidence Interval	B	b (SE)	95% Confidence Interval	B	b (SE)	95% Confidence Interval	B

1–3 times per week	−0.01 (0.72)	[−1.418, 1.399]	.00	−1.43^** (0.44)	[−2.29, −0.57]	−.03	−0.53 (0.42)	[−1.34, 0.30]	−.02
4–6 times per week	−1.88 (1.31)	[−4.439, 0.686]	−.02	−3.76^*** (0.74)	[−5.21, −2.30]	−.05	−2.65^*** (0.54)	[−3.71, −1.60]	−.05
Daily	−4.10^** (1.33)	[−6.708, −1.500]	−.05	−3.22^*** (0.77)	[−4.72, −1.72]	−.05	−3.46^*** (0.64)	[−4.71, −2.21]	−.07

Reference category is no fast food consumption in past week. Models control for all covariates listed in Table 1, including children’s fifth-grade test scores. Regression models were conducting using full information maximum likelihood estimation with longitudinal sampling and jackknife weights.

p < .05.

^**

p < .01.

^***

p < .001.