Reliability of GSHS questionnaire among school children and adolescents in Kampala, Uganda

Abstract

The Uganda Global School-based Health Survey (GSHS) measures behaviors and protective factors related to causes of mortality and morbidity among youth and adults in Uganda. This study aimed at assessing test-retest reliability of, and factors associated with reliability of the fruit and vegetable (FV) consumption and physical activity (PA) items of the adapted Uganda GSHS questionnaire among 8-19-year-old school children and adolescents in Kampala district, Uganda. The test-retest reliability of the FV consumption and PA items was determined on a sample of 621 children and adolescents aged 8-19 years, using Cronbach’s alpha. Association between reliability and strata defined by gender, age category, education level and type of school was also examined by chi square statistic. The reliability of the FV consumption and PA items was supported by Cronbach’s alpha coefficients of 0.80 and 0.72, respectively, indicating “good reliability.” Full consistency in FV consumption items was higher among secondary school participants (45%) compared to primary school participants (15%), 15–19-year-olds (43%) compared to 8–9-year-olds (19%) and 10–14-year-olds (21%), and private school participants (34%) compared to public school participants (24%). Similarly, full consistency for PA items was greater in secondary schools (34%) than in primary schools (0.7%), among 15–19-year-olds (30%) than 8–9-year-olds (0.0%) and 10–14-year-olds (12%), and in private schools (21%) than public schools (15%). The items therefore have the potential to measure the effects of nutrition interventions on children and adolescents because of their stability in making comparisons over time.