*2.3. Statistical Analysis*

A descriptive analysis of various sociodemographic variables was conducted for the total study population by country of residence and BMI status. Psychosocial, behavioral, and socio-environmental factors were also examined by country of residence and BMI status. Di fferences between proportions

were assessed using chi-square tests of homogeneity, and t-tests were used to calculate differences between means. Test for trend *p*-values were calculated to determine whether there was a linear association between the study variables and BMI status. Odds ratios and 95% confidence intervals for the association between psychosocial, behavioral, and socio-environmental factors and being overweight or obese were calculated using multinomial logistic regression. These results were adjusted for sex, age, race/ethnicity, and country of residence. Standardized odds ratios were determined to facilitate comparisons of the study variables since their score range varied considerably. Standardized odds ratios improve comparison and interpretability of the logistic regression results. Multinomial logistic regression models for males and females were also used to examine any differences by sex, after adjusting for age, race/ethnicity, and country. All *p*-values presented are 2-tailed and a *p*-value of <0.05 was considered statistically significant. All statistical analyses were performed using STATA software, version 12.0 (StataCorp LP, College Station, TX, USA).
