Fig. 1

Monthly Volume Purchasing Trends by Beverage Tax Status, 2016–2017 (Liters per Capita). Results are from separate weighted OLS regressions of total volume purchases, untaxed beverage volume purchases, lower-sugar taxed volume purchases, and high-sugar taxed volume purchases (L) per capita per month on a linear time trend with cluster-robust standard errors at the household level. The model also included month dummies (1-12) to account for seasonality (omitted from output). Time is a continuous variable (1-24). β is the beta coefficient, 95% CI is the 95% confidence interval, and p is the p-value for the statistical significance of the beta coefficient