Statistical packages for social sciences

All statistical analyses were performed using statistical packages for social sciences (SPSS, Chicago, IL, USA, version 23.0). Data are presented as mean ± standard deviation (SD) or standard error (SE). One−sample Kolmogorov–Smirnov test was used to test the normal distribution of the data, and in case of a skewed distribution, logarithmic transformation was applied. Within−group differences were compared using paired t-test in each group. To compare the change values among the intervention groups, one-way analysis of variance (ANOVA) and least significant difference (LSD) post-hoc tests were utilized. Age, gender, and baseline values were adjusted as covariates in the analysis of covariance (ANCOVA). P−values less than 0.05 were considered statistically significant in all analyses.

Basic descriptive statistics are presented as mean (x) and standard deviation (SD). Sex and age differences were calculated by using analysis of variance (ANOVA) with post hoc comparison test between the groups. To calculate correlations between age and all the study variables, we used Pearson coefficient of correlation (r). For each variable, we determined sex- and age- specific percentile values (5^th, 10^th, 25^th, 50^th, 75^th, 90^th and 95^th) and used Cole’s Lambda, Mu and Sigma (LMS) method, in which the optimal power to obtain normality is summarized by a smooth (L) curve and trends in the mean (M) and coefficient of variation (S) are similarly smoothed^{18 (link)}. Next, all three curves (L, M and S) are summarized based on the power of age-specific Box–Cox power transformations for normalizing the data^{18 (link)}. All analyses were performed in Statistical Packages for Social Sciences (SPSS Inc., Chicago, Illinois, USA) and in LMS Chartmaker Pro version (The Institute of Child Health, London, UK). A p value of <0.05 (two sided) was considered statistically significant.

Age (without decimal places) and sex were self-reported. Basic descriptive statistics are presented as mean and standard deviation (SD). Sex differences were calculated by using Student t-test for independent samples. Spearman’s coefficient (r) was used to calculate the correlation between relative schoolbag weight and age. Next, we created smoothed sex- and age-specific 5^th, 10^th, 25^th, 50^th, 75^th and 90^th percentiles for relative schoolbag weight with Lambda (L), Mu (M) and Sigma (S) method. In this analysis, the optimal power to obtain normality is summarized by a smooth (L) curve and trends in the mean (M) and coefficient of variation (S) are similarly smoothed [13 ]. Next, all three curves (L, M and S) are summarized based on the power of age-specific Box–Cox power transformations for normalizing the data. Children carrying ‘optimal’ relative (defined as ≤10% of body weight) vs. ‘overload’ (>10% of body weight) schoolbag weight are presented in percentages for each sex and age group. All analyses were performed in Statistical Packages for Social Sciences (SPSS Inc., Chicago, Illinois, USA) with statistical significance of p<0.05.