Table 6:

Estimated Effect of Phone Survey Mode on Survey Responses using Alternative Specifications

	Propensity Score Match	Nearest Neighbor Match	Inverse Probability Weighted and Regression Adjusted
Weekly Sweetened Beverage Consumption	−0.22	−1.08***	−0.60**
	[0.40]	[0.27]	[0.26]
Agrees with Tax	0.03	0.04	0.04
	[0.03]	[0.03]	[0.03]
Impact Score	0.65*	0.83***	0.54**
	[0.35]	[0.29]	[0.27]
	Panel B – Redefine Consumption
Weekly Sweetened Beverage Consumption	−0.12***	−0.13***	−0.12***
	[0.04]	[0.03]	[0.03]
Agrees with Tax	0.04	0.02	0.03
	[0.04]	[0.03]	[0.03]
Impact Score	0.51	0.64**	0.43
	[0.39]	[0.28]	[0.27]
	Panel C – Seattle Only
Weekly Sweetened Beverage Consumption	−0.67*	−0.47	−0.37
	[0.35]	[0.30]	[0.31]
Agrees with Tax	0.04	0.02	0.08**
	[0.04]	[0.04]	[0.03]
Impact Score	0.55	0.91**	0.93***
	[0.41]	[0.41]	[0.35]
	Panel D – Remove Inconsistent Incomes
Weekly SSB Consumption	−1.01**	−0.91***	−0.68***
	[0.47]	[0.26]	[0.26]
Agrees with Tax	0.07**	0.02	0.05
	[0.03]	[0.03]	[0.03]
Impact Score	0.66**	0.52*	0.58**
	[0.28]	[0.29]	[0.28]

Note: Robust standard errors in brackets.

^***p<0.01

^**p<0.05

^*p<0.1