BACKGROUND: The validity of self-administered questionnaires used to study habits during adolescence is often put into doubt. The aim of this study was to analyze the reliability and the aspects of validity of the presentation of a self-administered questionnaire concerning smoking habits, alcohol intake and degree of exercise. MATERIAL AND METHODS: A descriptive transversal study was undertaken in 1990 in 846 primary school children in Lleida (Spain) (444 of 6th course and 402 from 8th course). The children were randomly selected by conglomerates (unit = class) and stratified according to the type of school. The degree of exercise, smoking habits and alcohol intake of each student and the consumption perceived within the environment as well as the attitudes and beliefs related to the three habits were studied. Another test was performed in a subsample at two months to study the reliability and aspects related to validity were analyzed by concordance of interrelated questions. RESULTS: The reliability of the second test regarding the degree of exercise was found to have a correlation higher than 0.69. For the main variables related to alcohol consumption, the percentage of agreement was greater than 70% being 78% for the smoking habit. The results did not allow conclusions concerning the reliability of the variables for quantifying consumption and the age of initiation of the same to be obtained. The reliability concerning the attitudes regarding smoking (0.79-0.82) was higher that for alcohol (0.66-0.72) and exercise (0.36-0.54). CONCLUSIONS: The reliability of the variables of the questionnaires on habits of alcohol intake, smoking and degree of exercise within the environment and the attitudes is acceptable. The questionnaire studied was considered useful for studying these habits in school children.
BACKGROUND: The validity of self-administered questionnaires used to study habits during adolescence is often put into doubt. The aim of this study was to analyze the reliability and the aspects of validity of the presentation of a self-administered questionnaire concerning smoking habits, alcohol intake and degree of exercise. MATERIAL AND METHODS: A descriptive transversal study was undertaken in 1990 in 846 primary school children in Lleida (Spain) (444 of 6th course and 402 from 8th course). The children were randomly selected by conglomerates (unit = class) and stratified according to the type of school. The degree of exercise, smoking habits and alcohol intake of each student and the consumption perceived within the environment as well as the attitudes and beliefs related to the three habits were studied. Another test was performed in a subsample at two months to study the reliability and aspects related to validity were analyzed by concordance of interrelated questions. RESULTS: The reliability of the second test regarding the degree of exercise was found to have a correlation higher than 0.69. For the main variables related to alcohol consumption, the percentage of agreement was greater than 70% being 78% for the smoking habit. The results did not allow conclusions concerning the reliability of the variables for quantifying consumption and the age of initiation of the same to be obtained. The reliability concerning the attitudes regarding smoking (0.79-0.82) was higher that for alcohol (0.66-0.72) and exercise (0.36-0.54). CONCLUSIONS: The reliability of the variables of the questionnaires on habits of alcohol intake, smoking and degree of exercise within the environment and the attitudes is acceptable. The questionnaire studied was considered useful for studying these habits in school children.