Table 4.

Test Statistics from Analyses Comparing Each Numerical Format with the Control Condition on Accuracy Measures

Accuracy Measure	Absolute Frequency vs. Control	Percent vs. Control	Absolute Frequency + Percent vs. Control	Relative Frequency vs. Control	Relative Frequency + Percent vs. Control
	Print Ad
Absolute frequency	8.91, P<.001	7.10, P<.001	9.64, P<.001	−1.03, P=.30	5.86, P<.001
Percent	8.65, P<.001	7.87, P<.001	9.36, P<.001	−1.30, P=.19	6.46, P<.001
Relative frequency	0.47, P=.64	1.41, P=.16	0.53, P=.60	3.21, P=.001	3.98, P<.001
Placebo	9.64, P<.001	8.06, P<.001	11.08, P<.001	2.04, P=.04	8.39, P<.001
	Television Ad
Absolute frequency	6.63, P<.001	7.41, P<.001	9.83, P<.001	0.30, P=.76	5.28, P<.001
Percent	6.12, P<.001	7.10, P<.001	8.89, P<.001	0.10, P=.92	4.90, P<.001
Relative frequency	2.78, P=.005	1.74, P=.08	1.98, P=.05	4.10, P<.001	5.61, P<.001
Placebo	8.93, P<.001	9.02, P<.001	10.02, P<.001	−0.19, P=.85	7.71, P<.001

Note. Accuracy (absolute frequency, percentage, relative frequency, and placebo) was measured as the percentage who reported the correct quantitative benefit information. Bonferroni-adjusted significance was defined as P<.003. Print condition t-tests had 1456 degrees of freedom. Television condition t-tests had 1339 degrees of freedom.