. Author manuscript; available in PMC: 2021 Sep 1.

Published in final edited form as: Am J Prev Med. 2020 Jun 21;59(3):e135–e139. doi: 10.1016/j.amepre.2020.02.021

Table 1:

Associations between demographic groups and predicted probability of being likely affected by a portion size regulation

Demographic	Excluding convenience stores^a		Including convenience stores^a
Demographic	Average effect^b	95% CI	Average effect	95% CI
Age less than 35 years (ref: age 35–65)	0.034	(0.015, 0.053)	0.067	(0.043, 0.091)
Male (ref: female)	0.027	(0.011, 0.042)	0.069	(0.049, 0.089)
Race/ethnicity (ref: non-Hispanic white)
Non-Hispanic Black	0.001	(−0.020, 0.022)	0.018	(−0.012, 0.048)
Hispanic	0.000	(−0.024, 0.025)	−0.024	(−0.057, 0.009)
Low education (ref: college degree)	0.045	(0.025, 0.066)	0.090	(0.067, 0.113)
Low income (ref: income >185% of FPL)	−0.003	(−0.028, 0.022)	0.008	(−0.018, 0.033)

Note: Boldface indicates statistical significance (p<0.05). Analyses include 6,594 individuals from NHANES 2013–14 and 2015–16.

Dependent variables differ based on whether they exclude or include convenience stores. Both version include restaurants, fast food or pizza restaurants, bars/taverns, sports/entertainment facilities, street vendors.

Average effects and 95% confidence intervals calculated using margins command in Stata 16.1. Average effect is the difference in average predicted probability of being affected by a portion size regulation in the group of interest minus the reference group calculated using the method of recycled predictions. Average effects can be multiplied by 100% to yield approximate effects in percentage points (e.g., an average effect of 0.034 indicates a difference of approximately 3.4 percentage points).

CI, confidence interval. FPL, Federal Poverty Line. NHANES, National Health and Nutrition Examination Survey.