The data were analyzed as follows: First, a basic statistical analysis was performed to identify the mean, median, standard
deviation and relative frequency of each factor to be identified. Second, as the sample predominantly included white
females, the data were examined both with and without males and non-white subjects, to determine whether the inclusion
of these subjects influenced the pattern of findings. To check for differences in average scores for the three ASTAAA
factors, between the groups formed by men, women, whites and non-whites and interactions between gender and race, a
general linear model was used. To investigate the possibility that the nursing students’ socio-demographic characteristics influenced participants'
responses, binary logistic regression analysis was performed. This resource was used to estimate the Odds Ratio (OR),
with 95% confidence intervals in each of the three factors analyzed. The first step was to classify the 144 nursing students
into two groups, according to their scores on each of the three factors. Group 1 included participants with scores below the
median scores of each factor, whereas Group 2 included participants with scores that were higher or equal to the median
for each factor. Students scoring higher than the median were considered to have a positive attitude. Those scoring below
the median were considered to present negative attitudes.
Data were analyzed after the subjects were categorized into these two groups. Binary logistic regression was used to
identify the predictor variables that significantly influenced the participants’ scores. The variables investigated in this
study were gender, age, school, race and the number of hours of lectures on the subject of alcohol. All statistical tests were
conducted using the Statistical Package for Social Science (SPSS) version 18.0 for Windows.