Table 4:

Results from OLS Regression of Outcome Variables on Response Mode with Demographic Control Variables

	1	2	3
	Weekly Sweetened Beverage Consumption	Agrees with Tax	Impact Score
Phone Response	−0.78***	0.03	0.54**
	[0.21]	[0.03]	[0.24]
Weekly Sweetened Beverage Consumption		−0.02***	−0.17***
		[0.00]	[0.03]
Lives in Seattle	−0.70***	0.01	0.07
	[0.21]	[0.03]	[0.23]
High Income (≥260% FPL)	−0.32	−0.02	−0.01
	[0.21]	[0.03]	[0.24]
Non-Hispanic White	0.21	0.05	0.04
	[0.34]	[0.04]	[0.40]
Non-Hispanic Black	0.97*	−0.04	−0.42
	[0.50]	[0.06]	[0.49]
Non-Hispanic Asian	−0.54	−0.05	−0.37
	[0.44]	[0.06]	[0.53]
Hispanic	0.79**	0.02	−0.09
	[0.36]	[0.04]	[0.40]
50 or Younger	0.99***	0.10***	0.26
	[0.21]	[0.03]	[0.24]
Completed College or Above	−0.80***	0.13***	1.29***
	[0.23]	[0.03]	[0.25]
Married or Partnered	0.23	0.04	0.53**
	[0.21]	[0.02]	[0.23]
Democrat	−0.45*	0.14***	1.60***
	[0.27]	[0.03]	[0.28]
Independent	−0.28	0.01	−0.13
	[0.30]	[0.04]	[0.32]
Observations	1,702	1,609	1,702
R-squared	0.07	0.07	0.10

Note: High income is family income ≥ 260% of the Federal Poverty Line. Linear regression with robust standard errors in brackets.

^***p<0.01

^**p<0.05

^*p<0.1