Skip to main content
. Author manuscript; available in PMC: 2017 Sep 1.
Published in final edited form as: Addict Behav. 2016 Apr 21;60:154–159. doi: 10.1016/j.addbeh.2016.04.008

Table 3.

Negative binomial models predicting frequency of e-cigarette and hookah use.

Predictor Coefficient Standard Error z-score p-value
E-cigarettes: model χ2 (6) = 46.73, p < .001
Age −0.02 0.08 −0.31 .758
Sex −0.52 0.22 −2.32 .020
Race/ethnicity 0.04 0.12 0.35 .729
Student status −0.14 0.27 −0.50 .615
Cigarette days −0.01 0.03 −0.25 .800
E-cigarette expectancies 1.38 0.23 6.04 <.001
Hookah: model χ2 (6) = 34.89, p < .001
Age −0.02 0.07 −0.24 .809
Sex −0.32 0.24 −1.36 .175
Race/ethnicity 0.10 0.11 0.89 .375
Student status 0.24 0.25 0.96 .336
Cigarette days 0.11 0.03 3.94 <.001
Hookah expectancies 0.78 0.17 4.66 <.001