Physical fitness of students in Indonesian during the COVID-19 period: Physical activity, body mass index, and socioeconomic status

_


INTRODUCTION
The COVID-19 pandemic has become an important problem in the education sector [1]. Learning activities that are usually carried out face-to-face have shifted to online-based learning which is increasing rapidly world wide [2], including in Indonesia. Previous studies have analyzed the factors that support online learning [3][4][5], and student satisfaction in online-based learning [6,7]. Learning like this can cause a decrease in students' motivation to learn in schools, because students only focus in front of the monitor and depend on smartphones. In addition, the limited PA carried out by students will have an impact on their PF [8,9],especially in physical education.
PA is often associated with PF [10]. This is due to the benefits of PA, such as calorie expenditure, weight loss, reducing stress, social interaction, reducing the risk of poor health, and improving selfimage [11,12]. Despite these benefits, a large proportion of children and youthworld widefail to actively engage in PA [13][14][15]. Globally, only 20% of adolescents meet this recommendation [14]. The total amount of PA carried out by adolescents is insufficient [16], leading to a sedentary lifestyle [17]. The World Health Organization recommends that children and adolescents aged 5-17 years engage in moderate to vigorous PA every day for 60 minutes [18]. Thus, PF can be obtained from regular PA, especially for the purpose of physical education, which promotes PA in the learning process.
Then, PA is also often recommended to prevent obesity during childhood and adolescence.It is frequently determined that nutrition-related health issues, such as diabetes, obesity, and cardiovascular disease, significantly affect one's health [19]. Obesity can be caused by the excessive intake of food from fat sources. If food intake from fat and carbohydrate sources is not accompanied by PA, there is an imbalance in the energy used [20]. The national basic health research survey shows that data on malnutrition in Indonesia is quite large, with more than 1/4 of the population of adolescents aged 13-15 years experiencing stunting (about 26%), and 9% being underweight (thin) [21]. Among adolescents aged 16-18 years, 27% were stunted and 8% were underweight, while 2013 data showed that the prevalence of anemia in adolescents aged 13-18 years was 12.4% for boys and 22.7% for girls [22]. The main factor related to being overweight among teenagers was lack of PA. Teenagers who regularly engage in physical activities make it possible for them to avoid being overweight compared to those who are less active, and spend a lot of time watching television and playing gadgets. A qualitative-quantitative study on nutritional intake and PA conducted by UNICEF in 2017, reported that PA at school is minimal, and no more than 90 minutes a week. The food diversity of Indonesian youth is poor, or only 25% consume sources of iron and micronutrients, such asanimal and vegetable sources [22].
The majority of research has also examined how SES affects PF, concentrating on the accessibility of recreational opportunities and PA in communities with low SES [23,24]. According to the study, communities with high SES are 4.5 times more likely to have recreational amenities than those with low SES [23]. According to estimates, people with low SES are less likely to engage in PA if there are fewer facilities available, which could have an impact on their physical health. However, this study focused on environmental demographics and facility availability. Very limited research has linkedSES to individual PF. Other studies have also addressed the differences in PA levels among different SES. The findings demonstrate that low SES groups of all ages tend to be more sedentary than high SES groups [25][26][27]. However, several of these studies have compared groups according to race. Moreover, no association wa found between SES and PF. In one study, the FITNESSGRAM battery of fitness tests was used to evaluate the PF of 1.314 female high school students [28]. The findings indicate that there are disparities in body composition, running, and PA across groups of high and low socioeconomic classes, and that overall lower SES groups had lower levels of PF [28]. However, no previous studies have been conducted using male samples.
This study aimed to evaluate the effects of PA, BMI, and SES on students' PF during the COVID-19 period. This research is very important, because PF is the basis for quality child survival. Therefore,some of these factors need to be analyzed and proven. This research is expected to be useful for school principals, teachers (especially physical education teachers), parents, and students to evaluate the importance of these factors for the survival of quality children.

Study design
This research is associative quantitative, and aims to prove the relationship between two or more variables. The relationship used was causal, consisting of PA (X1), BMI (X2), and SES (X3) as independent variables, and PF (Y) as the dependent variable.

Participant
A total of 69 junior high school students number 2 from 50 cities, West Sumatra, Indonesianwere used as research samples, recruited by proportional stratified random sampling. The sample is aged 13-15 years, male (n=34), and female (n=35). This research was conducted after obtaining research ethics approval from the Health Research Ethics Committee, Padang State University approval number: No.08.01/KEPK-UNP/II/2021.

Procedure and instruments Physical activity
PA was measured using the GPAQ. The GPAQ was developed as a tool to evaluate and compare levels of PA on a local and international scale [29]. The GPAQ consists of 16 questions designed to estimate the level of PA, and consists of three important domains that students do during one week (work, travel, and leisure), and time spent sedentary behavior. The PA score was calculated as the total PA Metabolic Equivalent (MET) min/week.
The formula above is a calculation of the PA score based on the MET according to the analysis guide of the GPAQ questionnaire version 2. After obtaining the total PA value in MET min/week, it was categorized into three levels of PA: high, moderate, and low (Table 1). Body mass index BMI was measured using anthropometry based on the participants' weight and height. After the results are obtained using the formula above, the classification is determined by looking at the calculation table based on the index of weight according to height in children aged 5-18 years [30] ( Table 2).

Socioeconomic status
The SES of the parents was obtained using a questionnaire, consisting of the level of education and income of the parents for 1 month. The categories for the parents'education level were as follows: no school was coded 1, elementary school was coded 2, junior high school was coded 3, senior high school was coded 4, and undergraduate/above was coded 5. The categories for parental income for one months are low income (≤ IDR 3.000.000) (code 1) and high income (≥ IDR 3.000.000) (code 2).

Physical fitness
PF was obtained from the "Indonesian physical fitness test" for the age group13-15 years. This test consists of several components: 50 meters run, pull-ups/elbow-bend hanging test, sit-ups, vertical jump, and middle distance running 800/1000 meters [31] (Tables 3 and 4). Then, each of the test items in Tables 3 and 4 was accumulated as a whole into the level of PF (Table 5).   Very less

Statistical analysis
Data were analyzed using descriptive statistics to determine the classification level of each research variable. Correlation and regression analyses were then used to prove the relationship between variables. All stages were analyzed using IBM SPSS version 24.

RESULT
The data description aims to describe the characteristics of the test and measurement of each variable, so that the level/classification of the sample is known. Table 6 shows that the average value of PA for male is 608.97 or is in the moderate classification, and for female is 514.29 or is in the low classification. The BMI for male is 0.97 or is in the normal Physical Activity Review, vol. 11 (1), 2023 www.physactiv.eu _____________________________________________________________________________________ 81 classification, and for female is 1.79 or is in the fat classification. Education level for parents of male 3.32 and female 3.00 or junior high school graduates. Parental income for male is 1.26 and female is 1.23 or is in the low classification. Then, PF for male is 13.41 and female is 9.51 or is in the less classification. Figure 1 shows that the average score for males is superior to that of females, in terms of PA, BMI, SES, and PF. The results of the Kolmogorov-Smirnov normality test and inearity test also showed that each variable was normally and linearly distributed (P>0.05) ( Table 7). Correlation and regression analyses were used to determine the interrelationships between these variables. Table 8 shows that PA, BMI, and SES have a significant relationship withPF, either partially or simultaneously (P<0.05). The correlation between PA and PF was 0.644, regression significance was 47.53, and contribution was 41.50%. The correlation between BMI and PF was 0.571, regression significance was 32.40, and contribution was 32.60%. The correlation between SES and PF was 0.622, the regression significance was 45.27, and the contribution was 38.70%. Furthermore, the correlation of PA, BMI, and SES withPFwas 0.822, the regression significance was 45.17, and the simultaneous contribution was 67.60%.    *Regression and correlation were significant (p<0.05); "B" is the regression model; "R" is the magnitude of the relationship; "RSquare" is the amount of contribution; "F" is the regression significance; "t" is the significance of the relationship.

DISCUSSION
This finding suggests, that PA, BMI, and SES on students' PF (simultaneous effect 67.60%). PA had an effect of 41.50%, BMIof 32.60%, and SES 38.70%. Of these three factors, PA is the dominant factor affectingPF, without ignoring other factors (BMI and SES). The results of this study are consistent with previous studies, reporting that PA is one of the most effective ways to improve PF [32]. Numerous physiological and psychological consequences, including cardiorespiratory fitness, have been positively correlated with moderatetostrenuous PA during adolescence. [33], metabolic diseases [34], and better mental health [35,36].The advantages of a physically active lifestyle in adolescence for both metabolism [37] and psycho-social behavior [38] have been well-documented in several research studies. Previous studies reported that, students who participated in carefully developed physical training programs experienced considerable changes in their physical development, fitness, cardiorespiratory system functionality, and body balance [39]. Students' PF indicators vary widely, necessitating routine customization of the training approach [40].Previous studies have also reported that games that involve PA can improve students' psychomotor, psychological, memory, and attention, and thus have an impact on their PF [41].
Schools through physical education are ideal for making students aware of fitness, which affects their physical and psychological well-being [42][43][44][45]. This opinion is consistent with earlier research showing that physical education benefits children's and teenagers' health [46,47] and physical and psychological well-being [42,48]. The results of previous studies show a significant difference between the percentage of adolescents who performPA on school days in physical education compared to nonclass physical education school days. The percentage of students meeting PA targets on non-physical education class days (i.e., less than 50%) [43] is in line with other studies [43,49,50]. Thus, it is not surprising that students who have a good level of PA have an impact on good PF; therefore, they are more productive than students who have a low level of PA.
Previous studies have reported that a lack of PA is associated with overweight and obesity during childhood and adolescence [51]. Intense PA can help develop healthy musculoskeletal tissues and maintainideal body weight [52]. The results of this study are consistent with those of previous studies, reporting that a healthy lifestyle and balanced nutritional intake are important for avoiding various adverse health effects [53][54][55][56]. Lack of knowledge is one of the main causes of nutritional problems; therefore,it has a negative impact on students' health and fitness [57]. Another study also reported that students who eat nutritious food have an influence on exams at school; they show better behaviors, such as school attendance, and complete assignments given by the teacher carefully, compared to students who do not consume nutritious foods [58].
SES has an impact on PF. The findings of this study are consistent with those of earlier research, which showed that children from low socioeconomic homes frequently have issues with nutrition and _____________________________________________________________________________________ 84 unhealthy behavior [59][60][61]. Families with sufficient financial conditions fulfill, their children's nutritionalneeds [62]. A poor diet is also linked to lower socioeconomic levels; minorities tend to skip breakfast and consume fewer fruits and vegetables [60,61]. The results of this study are in accordance with previous studies, which reported that SEShas a higher impact on student engagement in activities that are better compared to students who have a low SES [27]. According to another study, SES significantly affects people's physical health, but not their psychological health. The quality of one's lifestyle significantly improves one's mental and physical health. Additionally, the link between SES and health is mediated by lifestyle [63]. This study supports earlier studies that found that higher SES groups have better health outcomes than lower SES groups [64,65]. Numerous reports on SES and ways of life have focused on physical health [66]. In addition, higher income is associated with good nutritional status, housing conditions, medical services, etc [62]. Additionally, families with higher education levels typically have better health awareness and understanding of their health [67]. Based on dataobtained in the field, the level of PF is influenced by several factors, including PA, BMI, and SES. However, there are some limitations that need to be validated in future research. The sample used was junior high school students aged 13-15 years, so a wider sample size and diversity are needed. The factors used are still limited, namely PA, BMI, and SES; therefore, it is necessary to add other factors that affect PF. Then, the classification for the level of family income (SES) is divided into onlytwo classifications: low income and high income. This is because there is no definite reference for this classification, so we use 2 classifications (low and high income) which are guided by the "Regional Minimum Wage" (different for each region in Indonesian based on the research subject area). It is also important for future research on moderate income classifications, so that there are differences in each income classification.

CONCLUSION
Based on these findings, we conclude that physical activity, body mass index, and socioeconomic Based on these findings, we conclude that PA, BMI, and SES affect students' PF (simultaneous effect 67.60%). PA had an effect of 41.50%, BMI of 32.60%, and SES 38.70%. Of these three factors, PA is the dominant factor affecting PF, without ignoring other factors (BMI and SES). This means that regular PA, balanced nutritional intake, and good SES have an impact on students' PF. This research is expected to be useful for principals, teachers (especially physical education teachers), parents, and students to evaluate the importance of these factors for the survival of quality children. Future studies should include a larger sample size and other factors that influence PF.