Predicting smoking behavior among male Saudi Arabian college students

This study describes the relationships of smoking behavior among a sample of male college students in Saudi Arabia to their religious practice, parents\u27 smoking behaviors and attitudes, peers\u27 smoking behaviors and attitudes, and knowledge about the dangers of smoking. A 49-item questionnaire was developed and pilot tested in Saudi Arabia. This questionnaire was completed by 715 undergraduate male students at the King Saud University in Riyadh, Saudi Arabia. Results show that 29.8% of the students were smokers (13.8% cigarette smokers, 7.3% sheesha smokers, and 27% cigarette and sheesha smokers). Logistic regression analysis suggested that students who were more faithful in their practice of Islam were 15% less likely to smoke than those who were less faithful. Students who reported having a few friends who smoked where almost three times more likely to smoke than students who reported that none of their friends smoked. Students who reported that more than half of their friends smoked were almost seven times more likely to smoke, and students who reported that all of their friends smoked were eight times more likely to smoke than students who reported that none of their friends smoked. Students who reported that their friends had a neutral attitude toward their smoking were almost twice as likely to smoke as students who reported their friends disagreed with their smoking. Students who reported that their friends agreed with their smoking were three times as likely to smoke as students who reported their friends disagreed with their smoking. Students who were more knowledgeable about the dangers of smoking were 8% less likely to smoke than students who were more knowledgeable about the dangers of smoking. It was not possible to assess the effect of parents\u27 behaviors or attitudes on college student smoking, since 90% of the parents disagreed with their children smoking and only 12% of the parents smoking. There was insufficient variance in the data to conduct an analysis. The logistic analysis identified peers (friends) as the most powerful factor in predicting smoking. The four-factor model had an overall classification accuracy of 78%. Results suggest a clear need to understand more fully the dynamics of peer relations among Saudi Arabian males as a basis for developing tobacco education/prevention programs