Abstract
A questionnaire relating to smoking behavior was developed and administered to 3,786 college students who were attending introductory personal health classes at selected colleges in the state of Oregon during the 1963–64 school year. Analysis of the data was not completed until 1987 (twenty-three years later) at which time the data cards were discovered in storage. During the 1986–87 school year a modified questionnaire was developed and administered to 863 college students attending introductory personal health classes at three of the same four universities that were utilized in the 1963–64 study. The study addressed the smoking behavior of an often neglected population of older adolescents and younger adults over a time period of twenty-three years. Stepwise logistic regression equations were used to determine the set of variables that best accounted for smoking status in 1963–64 and 1986–87. Results indicated that the variables that increased the probability of an individual smoking in 1963–64 did not increase the probability of an individual smoking in 1986–87. The characteristics and behaviors of family members have significantly less influence on the smoking behavior of adolescents and young adults today when compared to adolescents and young adults in 1963–64.
Get full access to this article
View all access options for this article.