#	Assumption	Step	Result	Decision
1	Normality of the data associated with the dependent variable	We computed the Mahalanobis values through the HLR in which PA enjoyment was the dependent variable	The significance values associated with the Mahalanobis values met the condition p < 0.001; thus, normality was confirmed. Each variable also produced a skewness and kurtosis value not greater than 2 in absolute terms [26,28]	We confirmed normality of the data for HLR analysis
2	Linearity	We plotted standardised residuals against standardised predicted values of the dependent variable in the above HLR analysis. We observed the linearity of the lines of best fit	The graph shows a straight line as recommended [28]	Assumption or condition met for HLR analysis
3	Independence of errors	Durbin–Watson statistics were generated for all the HLR models fitted	Durbin–Watson statistic was approximately 2 as recommended [26]	The assumption was met for HLR analysis
4	Multi-collinearity	Tolerance values were computed through the above HLR analysis	The tolerance values are >0.2 as recommended [28]	The assumption was met for HLR analyses
5	Homogeneity of variances	We plotted standardised residuals against standardised predicted values of the dependent variable in all HLR models	The graphs produced a satisfactory pattern as recommended [28]	The assumption was met for HLR analyses
Note: HLR—hierarchical linear regression; PA—physical activity.