METHODS
Respondents. A random digit dial technique was employed by the Luntz Research Companies to conduct this survey. Randomly generated telephone numbers were pre-screened to determine if a adolescent in the 12-17 year-old age group resided in a household. Twelve hundred adolescents were inter-viewed as part of a survey designed to study the substance-related beliefs of adolescents and their parents. The adolescent sample was 51% male and 49% female. Sixty-six percent of the sample were Anglo, 12% were Hispanic, and 11% were African-American. The remaining 11% was comprised of Asian-Americans, Native Americans, and adolescents who reported their ethnicity as ‘‘other.’’ Table 1 presents sample characteristics.
The order of parent-adolescent interviews was based on opportunity and subsequent examination of these interviews revealed no order effects. In addition, because the adolescent samples’ characteristics with respect to gen-der, ethnicity, socioeconomic status, and family composition were quite simi-lar to those of the population of 12-17 year-olds in this country, no data weighting was required.
Procedure. Adolescents were told that the interviewer was a college stu-dent who was interviewing young people around the country ‘‘about current events and about how it is to be a teenager in America today.’’ The adoles-cents were also informed that the interview would take approximately 15 minutes and that there were no right or wrong answers to the questions.
Instrument. The adolescent survey contained 72 questions ranging from demographic questions including age, gender, and household composition to questions regarding the respondent’s substance use beliefs and behaviors.
To assess current smoking behavior, adolescents were asked, ‘‘Do you
Downloaded by [Tulane University] at 20:49 09 October 2015
JOURNAL OF ADDICTIVE DISEASES8