2.1 Study Design
This cross-sectional study was conducted from November 2018 to November 2019, with the aim of assessing the impact of policy, school, family, and community environmental factors on PA and PF among children and adolescents in China. Employing a stratified cluster random sampling method, the research selected 17 schools across ten districts within five Chinese cities: Shanghai, Guangzhou, Harbin, Yinchuan, and Guiyang. The sampling framework encompassed a variety of educational levels, including primary schools (grade 4), junior middle schools (grade 7), and high schools (grade 10). Participants were involved in a comprehensive questionnaire survey designed to evaluate environmental factors and collect data on PA levels and basic demographic information. Additionally, the PF levels of the children and adolescents participating in the study were assessed through on-site testing. The study protocol received approval from the Ethics Committee of the Shanghai University of Sport. Consent was obtained from the teachers and principals of the participating schools. Participation in the study was entirely voluntary, with informed oral consent secured from all children and their parents or guardians prior to data collection. Data were collected and analysed anonymously, ensuring the privacy and confidentiality of all participants.
2.2 Participants
In this cross-sectional study, the participants consisted of 3,060 children and adolescents aged 9-18 years, drawn from 17 schools across five cities, including 10 primary schools, 4 junior middle schools, and 3 high schools. Due to incomplete or aberrant questionnaire responses, 268 participants were excluded from the study. Additionally, 45 students were unable to partake in the PF tests due to health reasons, resulting in a final sample size of 2,747 participants, yielding an effective response rate of 89.8%. This study included 1,665 children aged 9-12 years (60.6%) and 1,082 adolescents aged 13-18 years (39.4%), with a gender distribution of 48.2% males and 51.8% females. Furthermore, the sample comprised 1,500 students (54.6%) from urban areas and 1,247 students (45.4%) from rural areas. This diverse participant pool ensures a comprehensive analysis of the impact of environmental factors on PA and PF across different age groups and genders.
2.3 Procedure
Prior to the commencement of this study, comprehensive training was conducted for all individuals involved in administering the questionnaire survey and conducting the PF tests, with all participating researchers being graduate students specializing in sports science. Before initiating the survey and testing, researchers introduced the study objectives and methods to participants in classroom settings. Under guidance, each student completed a 4-page questionnaire within 20 minutes, either online or on paper, in a classroom or computer lab setting. This questionnaire primarily assessed policy, school, family, and community environmental factors, PA levels, and demographic information. Additionally, parents or guardians of each student were invited to complete a corresponding 4-page home questionnaire, which included questions about parental support for physical exercise and other family-related characteristics. Following the completion of the questionnaire, students underwent PF testing conducted by trained research assistants during scheduled class times.
To ensure the integrity and accuracy of the data collected, a rigorous double data entry system was employed. Two experienced assistants independently input the collected data into a secure computerized database, which was then meticulously cross-checked to ensure consistency and accuracy. Both questionnaire responses and physical test results were encoded with unique identifiers to maintain participant confidentiality. Access to this database was strictly limited to authorized researchers, ensuring data privacy and security.
2.4 Measures
In this study, policy, school, family, and community environmental factors and PA levels were based on self-reports from survey questionnaires, and PF test was assessed through on-site testing. The specific methodologies for each measurement area are detailed below:
2.4.1 Policy, school, family, and community environmental factors
In this study, policy (Qp), school (Qs), family (Qf), and community (Qc) environmental factors were assessed using a questionnaire survey, which is the "Child and Adolescent Sports Fitness Survey Questionnaire" developed by Shanghai University of Sport. This questionnaire has been extensively utilized nationwide and has contributed data to numerous published articles, demonstrating its high reliability and validity(16, 19, 24).
Policy Environment (Qp): Parents responded to two questions about their knowledge of the National Physical Fitness Standards for Students and the Interim Measures for the Prevention and Control of Risks in School Physical Activity policies in order to assess their awareness of sport and fitness policies.
School Environment (Qs): Students answered six questions on various aspects, including the adequacy of their school’s sports facilities and equipment, the impact of PE classes, satisfaction with PE teaching, support from other teachers for engaging in physical exercises, the sufficiency of time allocated for extracurricular physical activities and overall exercise atmosphere in the school.
Family Environment (Qf): Parents completed a series of ten questions evaluating their encouragement of their children’s participation in sports, attendance at their sports activities, communication regarding the health benefits of sports, active interest in their children’s PE learning at school, the importance of parental involvement in sports competitions or performances, joint participation in sports activities, accompaniment to sports events, family participation in sports as leisure activities, financial support for sports activities, and leading by example in sports participation.
Community Environment (Qc): Students were asked four questions about the prevalence and quality of youth sports activities in their community or neighbourhood, including the organization of sports events, availability of free sports skill training, establishment of youth sports organizations, and accessibility to sports facilities suitable for young people.
Each of these questions was scored on a five-point scale where 1 equated to 100 points, 2 to 75 points, 3 to 50 points, 4 to 25 points, and 5 to 0 points (Appendices). The total scores for each section of the questionnaire were calculated, and the average score for each environmental factor was determined based on these totals.
2.4.2 PA Level
To ascertain the levels of PA among students, this study utilized two adapted items from the International Physical Activity Questionnaire Short Form (IPAQ-SF), renowned for its robust psychometric properties. These items facilitated the assessment of activities across three distinct intensity levels: (1) low-intensity activities, such as walking; (2) moderate-intensity activities, such as carrying light loads and cycling at a normal pace; and (3) vigorous-intensity activities, such as running quickly and performing aerobics dance(25). For each grade level, respondents indicated the frequency (days of engagement in each activity) and duration (minimum of 10 minutes per session) of these activities over the past seven days. The average daily minutes of low-intensity physical activity (LPA) were calculated by dividing the total minutes of low-intensity activities by seven. Similarly, the average daily minutes of MVPA were obtained by summing the minutes of moderate and vigorous-intensity activities and dividing by seven.
2.4.3 PF Test
PF indicators in this research included BMI, waist-to-height ratio (WHtR), grip strength, vertical jump, and the 20-meter shuttle run test (20-mSRT).
BMI: Children's barefoot weight (in kilograms) and height (in centimeters) were measured using a portable device (GMCS-IV, Beijing Jianmin Company, China). BMI was calculated as the weight in kilograms divided by the square of height in meters (kg/m²).
WHtR: Reflecting abdominal fat accumulation and a key indicator of central obesity, WHtR was measured by wrapping a tape measure horizontally around the waist at a point 1 cm above the navel and midway between the lower margin of the last rib and the top of the iliac crest. The measurement was recorded to an accuracy of 0.1 mm.
Grip Strength: A convenient and effective method for assessing upper limb strength and muscle development, grip strength was measured using a portable dynamometer (T.K.K 5401, Japan). Participants stood with their arms hanging naturally, palms facing inwards, and squeezed the dynamometer with maximum force. The highest value from two attempts for each hand was recorded in kilograms to one decimal point.
Vertical Jump: To evaluate lower limb explosive power and muscle strength, participants performed vertical jumps using a portable device (T.K.K 5406, Japan). The height of the jump was determined by measuring the distance pulled by a connected thread at the peak of the jump. The best of two attempts was recorded in centimeters to one decimal point.
20-mSRT: The 20-mSRT, used to assess maximal oxygen uptake (VO2max), involved participants running back and forth between two markers, 20 meters apart, with increasing speed every minute starting at 8.5 km/h. The test ended when participants could no longer maintain the pace or elected to stop.
2.4.4 Demographic Measures
Throughout the survey, demographic information such as age, gender, grade, and residential area (rural or urban) of the participants was collected.
2.5 Statistical Analysis
Baseline characteristics and descriptive statistics were calculated to summarize the demographic and main study variables, employing means and standard deviations for continuous data, and frequencies and percentages for categorical data. SEM was employed to investigate the causal relationships among the variables. SEM allowed for the analysis of complex relationships, where PA and PF were modeled as dependent variables, influenced by policy, school, family, and community environmental factors as latent variables. Specific measurement indicators for each latent variable were incorporated into the model, ensuring an accurate representation of the theoretical constructs.
The fit of the model to the data was rigorously evaluated using several indices: the chi-square value (χ2), Root Mean Square Error of Approximation (RMSEA), Comparative Fit Index (CFI), and Tucker-Lewis Index (TLI). These indices were employed collectively to assess the adequacy of the model, with a focus on the χ2 value, RMSEA less than 0.08, and CFI and TLI values exceeding 0.90, indicating a well-fitting model. Statistical analyses of baseline features were carried out using SPSS, and SEM was analysed using Mplus version 7.4. The significance threshold for all hypothesis tests was set at α=0.05, ensuring the statistical validity of the findings.