Skip to main content
. 2011 Sep 8;13(1):55–63. doi: 10.1007/s11121-011-0244-3

Table 1.

Risk/protective factors and tried smoking incidence by race/ethnicity

Race/ethnicity Significant Contrast
White Black Hispanic
Media Exposures
 Movie smoking [mean(SD)] for 0 to 1 scale .30 (.27) .39 (.29) .31 (.28) H, W < B
 Weekday television viewing W < H < B
None 5.9% 6.6% 4.3%
< 1 h 22.0% 9.2% 17.2%
1–2 h 49.5% 35.4% 48.6%
3–4 h 16.8% 27.3% 23.1%
>4 h 5.8% 21.5% 6.7%
 Has TV in the bedroom 52.2% 78.3% 65.8% W < H < B
Socio-demographics
 Parent education H < B < W
HS grad or less 27.9% 45.0% 67.6%
Associate’s 31.3% 33.9% 19.0%
> = Bachelor’s 40.8% 21.1% 13.5%
Social influences and school
 Friend(s) smoke 15.1% 17.3% 16.9% None
 Sibling(s) smoke 12.2% 12.9% 9.2% H < B, W
 Parent(s) smoke 30.2% 32.3% 25.0% H < B, W
 Parenting style .60 (.27) .61 (.28) .60 (.29) None
 School performance average or below average 63.3% 73.7% 74.4% W < B, H
 Extracurricular activities .50 (.27) .51 (.29) .45 (.29) H < B, W
Characteristics of the adolescent
 Sensation seeking .34 (.28) .36 (.27) .33 (.28) None
 Rebelliousness   .23 (.28) .30 (.30) .25 (.30) H, W < B
Smoking onset
 Baseline → 8M 7.1% overall 6.2% 8.5% 7.6% None
 8 m → 16M 6.8% overall 6.7% 7.8% 8.0% None
 16 m → 24M 6.4% overall 5.9% 6.9% 7.7% None
 Baseline → 24M 14.3% overall 14.0% 14.1% 15.8% None

Significance is presented based upon differences in means by race. In the first row, for example, H, W < B would be interpreted as there being no statistically significant difference in the prevalence of movie smoking between White and Hispanic youth, but both groups having less exposure than Black youth. All continuous variables were scaled so that 0 represents the 5th percentile and 1 represents the 95th percentile for the distribution, values below 0 were set to 0, and values above 1 were set to 1; thus, range was always 0–1 for these variables. Means and SDs presented for continuous variables and percentages for ordered categorical or dichotomous variables. Significance tests for race/ethnicity contrasts based on t tests for continuous variables, ordered logistic regression for ordered categorical variables and logistic regression for dichotomous variables