Design and validation of scales to measure adolescent attitudes toward sexuality, addictive substances and road safety. Are they related to behaviour?

Abstract

Background. To validate scales to measure adolescent attitudes toward sexuality (EAS), toward addictive substances (EASA) and toward vial security (EASV), and to examine their correlations with behaviour on these areas.

Methods. We designed the scales by literature review; we validated their content using a Delphi-on-line panel with 18 experts, and a pre-test with a sample of 21 fourth-grade high-school students (similar to 10th in US) in Seville, Spain. The scales were administered to 188 fourth-grade high-school students in order to test their construct validity (exploratory factor analysis), their reliability (Alpha Cronbach test), and to determinate their correlations with age, gender, experience of sexual intercourse, substance use and road behaviour.

Results. EAS-10 obtained α=0.73, EASA-12 obtained α=0.80, EASV-12 obtained α=0.79. They showed evidence of one-dimensional interpretation after factor analysis, a) all items achieved weights r>0.30 in the first factor, which explained a significant proportion of variance before rotations, and b) total variance explained by the main factors extracted was greater than 50%. We found correlations between EASA-12 and tobacco use (r=0.43, p<0.001), alcohol use (r=0.30, p<0.001), age of alcohol initiation (r=0.32, p=0.003), and substance use (r=-0.55, p<0.001), and between EASV-12 and safe road behaviour (r=0.41, p<0.001), tobacco use (r=0.23, p=0.003), alcohol use (r=0,22, p=0,003), age of first experience of drunkenness (r=0.22, p=0.003), and substance use (r=-0.27, p=0.013).

Conclusions. The scales showed their reliability and validity. They could be employed for assessing adolescent attitudes to these priority intervention areas, for carrying out studies on their mediating role on adopting behaviour, and for designing and evaluating health programs.