Skip to main content
. Author manuscript; available in PMC: 2014 Apr 1.
Published in final edited form as: Am J Prev Med. 2013 Apr;44(4):345–350. doi: 10.1016/j.amepre.2012.12.008

Table 2.

Logistic Regression Predicting Smoking Initiation

Characteristic Odds Ratio (95% CI)
Black Teen 0.09 (0.01,0.60)
Not-Black Teen 0.13 (0.02,0.84)
Gender a
 Male Referent
 Female 1.34 (1.07,1.69)
SES (standardized)b 0.82 (0.53,1.26)
Has TV in Bedrooma
 No Referent
 Yes 1.04 (0.82,1.32)
TV Hours per Dayc 0.91 (0.55,1.49)
Ageb 0.73 (0.51,1.05)
Peer Smokingc 2.21 (1.45,3.35)
Sibling Smokinga
 No Referent
 Yes 1.12 (0.81,1.56)
Adult Responsivenessb 0.96 (0.69,1.34)
Adult Monitoringb 0.92 (0.60,1.42)
Conduct Disorder Symptomsb 1.89 (1.23,2.90)
Sensation Seekingb 3.40 (2.14,5.41)
Black Teens
 Black-oriented MSE Linear 4.61 (1.22,17.4)
 Black-oriented MSE Quadratic 0.01 (0.00,0.67)
 Mainstream MSE Linear 0.68 (0.30,1.50)
 Mainstream MSE Quadratic 1.58 (1.04,2.38)
All other Teens
 Black-oriented MSE Linear 1.99 (1.15,3.45)
 Mainstream MSE Linear 2.50 (1.69,3.71)
 Mainstream MSE Quadratic 0.73 (0.59,0.89)

MSE = Movie Smoking Exposure. Coefficients for Black Teen and Not-Black Teen represent intercepts for each group.

a

Dichotomous variable; reference group as indicated.

b

Modeled as a continuous variable; trimmed at approximately the 5th and 95th percentile values to avoid possible distortion due to outliers and estimates are scaled to reflect the difference in initiation associated with moving from the 5th to the 95th percentile.

c

Ordered categorical variable; estimates are scaled to contrast the lowest level with the highest.

Significant p values are bolded. Analysis excludes 519 participants lost to follow up and 35 participants with missing data on covariates.