Development and validation of a scale to measure social isolation in adolescents

Abstract

There is a lack of questionnaires specifically designed to assess social isolation in adolescents. We developed and validated a self-report measure of social isolation in adolescents, the Social Isolation Questionnaire (QIS). A literature review on social isolation and mental health in adolescents indicated 20 questions to form QIS. Two cross-sectional surveys with 48 and 1135 adolescents, respectively, evaluated the reproducibility and validity of QIS. The Bland–Altman plot did not indicate a systematic difference between measurements 15 days apart. Bartlett's sphericity test verified a correlation between the questions and the Kaiser-Meyer-Olkin test showed good adequacy (.896) to the factor analysis. Exploratory factor analysis suggested the exclusion of three questions (loading factors <0.3), and eigenvalues (4.9–0.8) indicated that the questionnaire should be composed of three factors (dimensions). The Cronbach's alpha indicated high internal consistency of the 17 questions (0.850 overall; Dimensions: 0.815 ‘feeling of loneliness’; 0.760 ‘friendship’; and, 0.680 ‘Family support’). The QIS score ranged from 0 to 131 (maximum social isolation score). The correlation between QIS and depressive symptoms (r = .543) indicated the construct validity of QIS. We evaluated QIS in the Portuguese version, thus translation and cultural adaptation are necessary to evaluate the questionnaire in other settings. We constructed and validated the QIS questionnaire, a self-administered questionnaire to assess social isolation in adolescents, composed of three dimensions; feelings of loneliness, friendship and family support. The QIS questionnaire seems a very promising tool to support practitioners and researchers in assessing social isolation among adolescents.