Table 4

Logistic regression model results†

Final model
−2logL3441.43423.03416.6
Difference244.718.46.4
p Value‡<0.00010.00040.09
β Coefficientsβ Coefficientsβ Coefficients
† −2*log likelihood, β coefficients and their standard errors. Assessing the effects of experiments and respondent characteristics on current smoking status (n=3996).
‡p Values are for forward selection for comparing the logistic regression models.
*The p value for the β coefficient is <0.05.
**The p value for the β coefficient is <0.01.
Intercept−0.12 (0.16)−0.32 (0.18)−0.31 (0.18)
Introduction 0.11 (0.12) 0.54 (0.16)** 0.49 (0.18)**
Filter−0.08 (0.12) 0.02 (0.17)−0.04 (0.19)
Intro*Filter−0.26 (0.17)−0.58 (0.24)*−0.38 (0.25)
High school−0.37 (0.14)**−0.37 (0.14)**−0.38 (0.14)**
Some college−0.61 (0.14)**−0.63 (0.14)**−0.63 (0.14)**
College graduate−1.56 (0.16)**−1.56 (0.16)**−1.56 (0.16)**
Female−0.44 (0.09)** 0.03 (0.17)−0.03 (0.17)
Hispanic−0.54 (0.13)**−0.52 (0.13)**−0.55 (0.23)**
Spanish−0.78 (0.21)**−0.84 (0.21)**−0.85 (0.21)**
$20000 to $50000−0.32 (0.11)**−0.32 (0.11)**−0.32 (0.11)**
$50000 and up−0.72 (0.14)**−0.73 (0.14)**−0.73 (0.14)**
Income not given−0.72 (0.18)**−0.71 (0.18)**−0.72 (0.18)**
Intro*female−0.98 (0.24)**−0.97 (0.24)**
Filter*female−0.19 (0.24)−0.18 (0.24)
Intro*filter*female 0.72 (0.35)* 0.70 (0.35)*
Intro*Hispanic 0.22 (0.29)
Filter*Hispanic 0.25 (0.30)
Intro*filter*Hispanic−0.95 (0.44)*