Predictive factors for loneliness in female high school students: An unvariate and multivariate logistic regression analysis

Background and aims: Loneliness typically includes anxious feelings. It is particularly relevant to adolescence period. It has effect on physical and mental health. The present study aimed to identify the predictive factors of loneliness among high schools female students. Methods: A cross– sectional survey was carried out among high schools female students in Ilam during the academic year 2014-15. Sampling was done by multistage method. The student's consent to participation in the study obtained by full filled the questionnaires. Data were collected by demographic and University of California, Los Angeles questionnaires. Questionnaires with incomplete information were excluded. The Cronbach’s alpha coefficient was measured as an index of internal identicalness of the questionnaire to verify its reliability. Results: A total of 400 female high school students were studied. Overall, 62.8% of students put into non- loneliness group and 37.3% of all have loneliness. The univariate logistic regression analysis demonstrates that education field, father’s education and father’s occupation were different between the groups (P<0.05). The risk of loneliness was higher in students with a mathematical sciences education field in comparison to general education field (OR=1.75). In multivariate logistic regression analysis the education field, father’s education and father’s occupation were considered as independent predictive variables for female students’ loneliness. The AUROC criterion was applied to compute both the sensibility and the specificity of the manikin. The overall percent of correct classification of the model is 64%. Conclusion: Identify the causes of students loneliness can prevent complications and provide appropriate solutions.