Physical Phenomena

The original item pool was constructed from a comprehensive literature review on the concept of physical literacy between 1994 and 2015, followed by three focus group interviews which lasted an average of 90 minutes per interview. Eleven Hong Kong Chinese physical education teachers with teaching experience ranging from 3 to 18 years were interviewed on three separate occasions. The focus group participants were all physical education subject specialists and came from primary (N = 5) and secondary (N = 6) schools in Hong Kong. This number was appropriate, as focus group sizes generally range from 4 to 12. [33 ] In the focus group discussions, participants were given opportunities to share their perception on physical literacy and to provide diverse experiences on how physical literacy impacts their students’ lifestyles. The researchers used reviewed literature to guide the focus group interviews that helped to identify some key attributes of physical literacy and relationships between all attributes such as “sense of self and self-confidence”, “self-expression and communication with others” and “knowledge and understanding” [26 ] for the development of the instrument.
The interviews were audio-taped and transcribed by the researchers. Content analysis was used to identify codes in an iterative process. [34 ] The researchers discussed the codes and their relationships in order to facilitate analysis for drafting different items of the instrument. The role of codes was to distinguish overall themes in which more specific patterns can be interpreted. After drafting and revising items, participants from the focus group were asked to complete an 18-item version of the instrument. They responded to the items, evaluated their clarity, and provided feedback on the response scale. The feedback suggested slight changes to 2 items and that the instrument could be achieved by using a 5-point Likert scale anchored by 1) “strongly disagree” and 5) “strongly agree” with a midpoint 3) “no comment”.
A second (revised) version of the instrument was then sent to a panel of four experts who were teaching and doing research in the area of sport science, physical education, health education and instrument development in local and regional universities. They were invited to evaluate the initial instrument for item wording, instrument length, clarity of the statement and response format. Revisions were made based on their feedback with a few items rewritten. There were no suggested items being added by the experts relating to the construct. The initial instrument was well received and commented as useful. A full outline of the items of the instrument which contained 18 questions on the understanding of physical literacy is shown in Table 1.

We used SAS version 9.1 (SAS Institute Inc, Cary, NC) for data analysis. Descriptive statistics (frequencies, percentages, means, standard deviations, and ranges) were generated to characterize the study sample in terms of socio-demographic parameters. We used several criteria to assess the subscale validity and reliability of the MFI-20.
Internal consistency of each of the five MFI-20 subscales was determined using three reliability tests: 1) inter-item correlation; 2) corrected-to-total (or item-total) subscale correlation; 3) and Standardized Cronbach's α coefficients (and item discrimination). The cutoff criteria for acceptance on reliability tests are as follows. First, item-total subscale correlations of not less than 0.30 and inter-item correlations of 0.30 to 0.70 were retained. Second, a fairly high reliability coefficient (Cronbach's α > 0.70) was required to assess the internal consistency reliability [30 ,31 (link)]. Floor/ceiling effects were considered significant if more than 15% of the subjects had either the lowest possible or highest possible score on the subscales [32 ]. A significant floor effect was expected in the well group.
As an indication of discriminant (known-group) validity, group differences in the five MFI-20 subscales were calculated using analyses of variance to examine the ability of the MFI-20 instrument to distinguish three groups: CFS-like, chronically unwell, and well. Using a Tukey correction, the alpha per test for each subscale was 0.01, for an overall alpha of 0.05. Two-way analyses of variance were performed to test the age and sex effects on the five MFI-20 subscales. Post-hoc analysis with Tukey p-value adjustment was performed for multiple subgroup comparisons.
To further assess construct validity of the subscales, an exploratory factor analysis was performed. A principle component analysis was used to extract factors. The obtained factors were rotated oblique using the Varimax procedure. A minimum eigenvalue of 1 was specified as the extraction criterion [33 ]. The desired criterion of factor loadings was set at 0.50 or above, slightly higher than the typical cutoff value of 0.40 [34 ].
Finally, the convergent validity of the MFI-20 was evaluated through comparisons of the MFI-20 with other instruments administered in the protocol. Pearson correlation coefficients were used to assess linear associations between the multi-item scales of SF-36, SDS, and STAI. We chose these instruments based on the association between fatigue and other measures on psychosocial functioning, such as health-related quality of life (measured by SF-36), depression (measured by SDS), and anxiety (measured by STAI) as well as the existing data from the source study.
The most valid SF-36 subscales for measuring physical health include the physical functioning, role physical, and bodily pain subscales and the physical component summary score [26 (link)]. The most valid SF-36 subscales for measuring mental health include the mental health, role emotional, and social functioning subscales and the mental component summary score [26 (link)]. For the concept of physical and mental health, we investigated correlations between MFI-20 subscales and physical and mental health as measured by the SF-36.

Descriptive statistics at baseline for anthropometric and demographic data, media use, cardiovascular endurance, self-concept, and self-worth are reported as mean ± SD for continuous variables and as frequencies and percentages for categorical variables. Mean changes from baseline to program end are represented as $\mathrm{\Delta }t1t2$ and changes from program end to one year later are represented as $\mathrm{\Delta }t2t3$ . Mean differences in boys’ and girls’ baseline characteristics [39 (link)] were analyzed using independent two-tailed t-tests in a between-subjects design. Based on a within-subject design, paired two-tailed t-tests were conducted to detect significant differences from $t1$ to $t2$ and from $t2$ to $t3$ .
Backward stepwise multiple linear regression analysis with $p\ge 0.1$ for variable removal was performed to examine the predictors of mean changes in $\mathrm{\Delta }t1t2$ BMI-SDS and $\mathrm{\Delta }t2t3$ BMI-SDS. The predictors included in the regression models are illustrated in Table S2 in the Online Resources, including a dichotomous variable to express the participants’ adolescence stage [40 (link)] so as to account for the nonlinearity of the relationship between physical self-concept and age [41 (link)].
To identify outliers, leverage values (< 0.2) [42 ], studentized excluded residuals (< 3 and >  −3) and Cook distances (> 1) were analyzed. No extreme values were found. Homoscedasticity, linearity, and normal distribution are assumed, based on visual inspection of quantile–quantile and scatter plots of the unstandardized predicted values and studentized residuals [43 ]. It is furthermore presumed that no autocorrelation existed between the residuals, since the Durbin–Watson statistics for all models had values close to 2. No multicollinearity existed between the predictors, as the variance inflation factor values were less than 10 in all regression models [44 ].
For all statistical analyses, IBM SPSS version 28.0 was used, and significance was set at p < 0.05.

For the assessment of physical self-concept and global self-worth, this study used two subscales that resulted from a German version of Harter’s Self-Perception Profile for Children [36 ] by Wünsche and Schneewind, named FSK-K (Fragebogen zur Erfassung von Selbst-und Kompetenzeinschätzungen bei Kindern) [37 ]. In the 30-item questionnaire, each item was rated on a scale of 1 to 4 in an alternative statement format, with a positive statement on one side (e.g., “I like my body the way it is”) and a negative statement on the other side (e.g., “I want my body to be different”). The child/adolescent decided which side of the description was kind of true/almost true/really true for him/her, sometimes with parental assistance. The test was conducted at all three measurement time points ( $t1,t2,t3$ ). Results were adjusted to fall within a range of 0–100 and recoded so that high scores indicated high self-concept/self-worth. Cronbach’s $\alpha$ was calculated for reliability analysis [38 ]. The internal consistencies of the subscales at baseline ( $t1$ ) and follow-up ( $t2,t3$ ) were α_t1 = 0.79; α_t2 = 0.81; α_t3 = 0.82 for physical self-concept (n_t1 = 214; n_t2 = 102; n_t3 = 72) and α_t1 = 0.71; α_t2 = 0.80; α_t3 = 0.81 for self-worth (n_t1 = 210; n_t2 = 209; n_t3 = 74).