Mechanism Between Physical Activity and Academic Anxiety: Evidence from Pakistan

A plethora of research has explored the benefits of physical activities on several variables such as academic performances, quality of life, fitness, levels of stress, and depression. However, less research considered the impact of physical activity on academic anxiety, self-enhancement, and self-criticism. This study examines the influence of physical activity on academic anxiety with self-enhancement and self-criticism playing a mediating role. Data for this model has been collected by using structured questionnaires administered to 418 Pakistani university students (undergraduate, graduate and post-graduate) aged between 18 and 36 years old. Structural equation modeling through analysis of moment structures (AMOS) was applied to find that physical activity significantly reduces academic anxiety and self-criticism, and it develops self-enhancement. Further, self-enhancement was negatively associated with academic anxiety while self-criticism is directly related to academic anxiety. More importantly, self-enhancement and self-criticism partially mediated between physical activity and academic anxiety of university students in Pakistan. This research recommends governments to support physical activity programs for university students.


Introduction
The beneficial effects of physical activity (PA) on mental health, quality of life and academics have been described in several studies [1,2]. Existing studies emphasized the relationship between physical activity and academic performance [1,3,4], depression [2,5,6], fitness [7][8][9], happiness [10], anxiety [11][12][13] and psychological distress [14]. However, a systemic review conducted by Larun et al. [15] with randomized control trials revealed that physical activity was not associated with anxiety among students. This makes us curious to find out the reasons behind these inconsistent results. There may be a different mechanism playing a role between physical activity and anxiety demonstrating that very little is known about the rapport that regulates physical activity and anxiety or depression [16]. Psychologically, mechanisms repeatedly suggested in the literature include self-esteem [17] and self-efficacy [18] and a few studies have tested these variables [19,20]. Several scholars have considered other possible theoretical frameworks, for instance, cognitive processes might play a role between the two constructs [16,21,22]. Further, instead of these well-known benefits, physical activity in students (children and youths) has not been satisfactory in Pakistan [23]. This is particularly true in developing economies due to the severe lack of physical activity. For instance, a study conducted by Arat and Wong [24] in six middle-income countries indicated unsatisfactory physical activity among people.
The current research has tried to fill this gap because there is little research linking physical activity to academic anxiety through self-enhancement and self-criticism. Therefore, we have designed this study to investigate the mediating effect of two cognitive propensities, self-enhancement and self-criticism between physical activity and academic anxiety (AA) of university students in Pakistan. Moreover, this research has been conducted in Pakistan, a country where 37.6% of students have been found inactive [25] and having almost a few research in the context of physical activity and anxiety. This is particularly true in Pakistan where no student is shown to participate in physical activity or given undivided attention [23]. This issue encouraged the current research team to conduct a study on physical activity with Pakistani students.

Physical Activity and Academic Anxiety
During their academic life, university students face various problems such as irregular sleeping, financial stress, living away from home, academic pressure, depression which have a negative influence on their academic performance [26,27]. Out of these issues, anxiety is reported as one of the major problems among graduate students that negatively affects their quality of life and academic performance [28]. Anxiety demonstrates the fear and worries of students about their academic outcomes. Hence, the reduction of anxiety is crucial for students [29]. Stathopoulou [16] claimed that physical activity is a key action for controlling and reducing anxiety. Further, physical activity is beneficial for decreasing depression, mental stress, anxiety and for boosting mood enhancement [30,31] Therefore, to have a healthy life and minimizing anxiety, students and adolescents need to participate in regular physical activities such as sport, hiking, walking, running and jogging [4,[32][33][34]. Additionally, studies have shown that students can reduce their depression, stress, and anxiety by adopting the habits of regular exercise and physical activity [13,35]. A study conducted in Pakistan by Khanzada [11] showed that students who actively perform physical activity had a low level of depression and anxiety. Considering the evidence presented, we suppose that students who participate in physical activity and do regular exercise would have less academic anxiety compared to those who are not physically active.

Self-Enhancement as a Mediator
According to the theory of positive illusions, self-enhancement means possessing positive views of the self [36]. Skayannis et al. [37] state that people can enhance their quality of life, positive feeling, lifestyle, wellbeing, motivation and personality mobility through PA. Also, PA should be promoted among adolescents to keep them healthy.
Physical education and physical fitness facilitate students to develop their self-enhancement and self-confidence while a lack of participation in physical education classes reduces enhancement among students [38]. Individuals often strive for high status, fitness and functional abilities that can be gained through performing moderate to vigorous physical activity [39,40]. Timo et al. [41] reported that PA exercise helps in the achievement of an effective lifestyle by motivating and cultivating enhancement among individuals. They further scrutinized that young people and adults can gain their self-enhancement and quality of life through PA. Daily exercise is considered a significant factor for mental strength and a favorable construal leading individuals to feel better [42]. PA and exercise increase happiness, build positivity among youths and reduce the adverse concerns which in turn stimulate enhancement [43]. PA diminishes mental and physical stress among individuals and increases social interaction and support in the community [44]. In general, individuals who participate in PA and exercise are physically and psychologically healthy and their self-enhancement is developed [45].
Physically active students frequently feel positive about themselves, have high self-esteem and see themselves in a positive status which in turn facilitates them in getting desirable academic results and decreasing anxiety [46]. Physical activity provides health-related benefits out of which self-development and self-enhancement are remarkable and may facilitate better academic performance [47]. Considering the potential benefits, many individuals incline to participate in exercise and sports activities. In turn, these benefits (e.g., self-enhancement) encourage their social interaction and reduce the anxiety level [48,49]. PA and exercise enhance self-esteem, motivation, self-efficacy, positive emotion, positive body image and social interaction that are the attributes of self-enhancement. In turn, these abilities and the positive consequences reduce anxious thinking among students and increase academic performance [50,51]. Consequently, daily exercise is a significant element of high satisfaction in life, a high level of happiness, high self-esteem that can result in low depression and anxiety among individuals [52]. We can also argue that regular exercise is a vital factor of general health and encourages individuals to take responsibility for well-being and self-enhancement [12]. In addition, self-enhancement and self-esteem diminish anxiety among people [53]. Furthermore, it is claimed that the relationship between physical activity and academic outcomes are mediated by psychosocial factors and characteristics of the students (e.g., self-esteem and depression) [2]. In general, it is believed that daily exercise and physical activity maintains self-enhancement that can provide support in the reduction of anxiety, stress, and depression [54]. Hence, self-enhancement could be a potential mediator between PA and AA.

Self-Criticism as a Mediator
Self-criticism (SC) is a cognitive tendency to have negative views of self [55]. According to Halamová and Kanovský [56], self-criticism should be reduced as it negatively influences the daily routine activities and mental health of the individuals. For instance, it is argued in an experimental study that regular exercise and physical activity significantly reduces criticism among individuals [57]. In this perspective, Vancampfort et al. [58] also scrutinized that self-criticism, lack of motivation and lack of confidence can be solved (e.g., removed) with the help of regular exercise and physical activity because one who rarely or occasionally participates in sport, hiking, dance, and exercise often face self-criticism [59]. Students who regularly exercise and participate consistently in physical activity have a high level of awareness, tolerance and general information. Hence, they do not accept criticism without a reasonable argument. Conversely, inactive students have internal conflict, a low level of self-confidence and self-alteration that can result in "being cheap with critic statements" [60]. Individuals who exercise daily have a low level of depression and often avoid criticism [61]. A study reports that lack of exercise, lack of fitness and imbalance eating often lead to self-criticism that can be mitigated through regular physical activity [62]. Physically active adults, due to their activeness in daily exercise have a low risk of health issues. Therefore, they maintain their motion and have less inclination toward self-criticism [63,64]. A study of Pakistani samples scrutinizes that physically active individuals are more satisfied with their personal life and positively interact with their family members. They usually do not demonstrate symptoms of depression and negative thinking [65]. Physical activity protects individuals from depression and in turn, they perceive a low level of self-criticism [66]. Hence, we hypothesize that physically active students have less self-criticism and less academic anxiety than physically inactive students. So, SC is expected to mediate between PA and AA.

Current Study
Considering the above presented theoretical and empirical argument, we can argue that physically active students doing physical activity intend to possess high self-enhancement and a low level of self-criticism that in turn reduces their academic anxiety. We have formulated the following hypotheses for the study: Hypothesis 1 (H1). Physical activity significantly reduces academic anxiety.
Hypothesis 7 (H7). Self-criticism mediates the relationship between physical activity and academic anxiety.
The research model is shown in Figure 1.
mediate between PA and AA.

Current Study
Considering the above presented theoretical and empirical argument, we can argue that physically active students doing physical activity intend to possess high self-enhancement and a low level of self-criticism that in turn reduces their academic anxiety. We have formulated the following hypotheses for the study:

Hypothesis 7 (H7). Self-criticism mediates the relationship between physical activity and academic anxiety
The research model is shown in Figure 1.

Participants
The model of this research is tested on the empirical evidence collected from Pakistani students. To gain useful insights, students from bachelor (13 and 14 years of education), master (15 and 16 years of education), MS/M Phil (17 and 18 years of education) and Ph.D programs have been studied. It was difficult to get the number of enrolled students in each program e.g., social sciences, computer, and engineering, natural and applied sciences. Hence, we used a

Participants
The model of this research is tested on the empirical evidence collected from Pakistani students. To gain useful insights, students from bachelor (13 and 14 years of education), master (15 and 16 years of education), MS/M Phil (17 and 18 years of education) and Ph.D programs have been studied. It was difficult to get the number of enrolled students in each program e.g., social sciences, computer, and engineering, natural and applied sciences. Hence, we used a convenience sampling technique asking participants to sign informed consent before participation.

Data Collection
We distributed 700 questionnaires in all universities after taking necessary permissions. Five hundred and seventy-seven (577) questionnaires were received back with a response rate of 82.43%. We detected 509 usable for analyses after scrutiny but 91 cases were also removed after Mahalanobis distance calculated. Finally, 418 useable cases were entered into analyses. Further details are presented in the flow chart in Figure 2. We distributed 700 questionnaires in all universities after taking necessary permissions. Five hundred and seventy-seven (577) questionnaires were received back with a response rate of 82.43%. We detected 509 usable for analyses after scrutiny but 91 cases were also removed after Mahalanobis distance calculated. Finally, 418 useable cases were entered into analyses. Further details are presented in the flow chart in Figure 2.

Measurement of Variables
Structured questionnaires were used for data collection and a hard copy approach was followed because data collection through email provides a lower response rate [67]. We used the English version of the questionnaires as the official and institutional language in Pakistan is English. In the cover letter of the questionnaire, it was mentioned that the data of this study would be used for research purposes only.

Measurement of Variables
Structured questionnaires were used for data collection and a hard copy approach was followed because data collection through email provides a lower response rate [67]. We used the English version of the questionnaires as the official and institutional language in Pakistan is English. In the cover letter of the questionnaire, it was mentioned that the data of this study would be used for research purposes only.
The measures of the current study were physical activity, Self-enhancement, Self-criticism, and Academic Anxiety. All have been taken as mono-factorial constructs. The control variables were age, educational background, kind of study, and gender.

Physical Activity
There are a few valid and reliable measures of physical activity which are described in prior studies [1,34,61,66,68]. In the present research, we have used the five items Physical Activity Questionnaire used by Cho [69] who Validated it in the Asian culture and we have adopted for Pakistan. Physical Activity Questionnaire is a five items scale collecting data about type, frequency, intensity, duration, and length of physical activity which are measured on a 5-point Likert scale. Example questions are "during a week, how often do you participate in the activity in your free time?" and "how intensely do you participate in the activity?" The range of alpha in the Cho study was 0.61 to 0.78. While the test-retest reliability of the tool was ranged between 0.61 to 0.91 [69]. In the current study, we presented data for the revalidated in Pakistani setting as the tool has never been used in Pakistan before. All items were loaded on one single factor when exploratory factor analysis was applied by using varimax rotation. The results for exploratory factor analyses, validity and reliability can be seen in Table 1. All results show acceptable, reliability, convergent and discriminant validity for the scale in the Pakistani context.

Self-Enhancement
Self-enhancement and self-protection strategies scale by Hepper et al. has been used in the study [70,71]. It is a 20-item scale developed for measuring self-enhancement strategies of students in a comparative study of Chinese and U.S. students [71]. Students have responded on a five-point Likert scale. The tool has already been used in eastern and western cultures, yet we have revalidated the tool in the current study for Pakistani culture. A sample item to measure self-enhancement is "Asking for feedback when you expect a positive answer". All the values for convergent validity, discriminant validity, and reliability show that the tool is valid and reliable in the Pakistani setting as reported in Table 1, and the measurement model for the scale is given in latter tables.

Self-Criticism
Self-criticism was measured with a 22-item scale (see Table 1). The items 6, 8, 11, 12, 16, 20 and 21 are reverse coded items. These items were adapted from the study of Thompson and Zuroff [72] who have tested and validated the English version of the scale. A representative item is "I have a nagging sense of inferiority". The scale has already been used for university students in Pakistan [73]. However, it was revalidated by applying exploratory factor analyses and confirmatory factor analyses. Reliability, factor loadings, and validity results have been presented in Table 1. Results show that the tool is highly valid and reliable. Further, the measurement model presents a good fit.

Academic Anxiety
This research used the 6-item state and trait anxiety scale [74] adapted from a previous study [75] in which it confirmed the validity and reliability of the English version of the tool. It is a widely used tool. Sample items of the academic anxiety scale were "I feel upset", 'I am relaxed". The tool has been revalidated in Pakistan as well (see Table 1) confirming the monofactorial structure as explored by the former authors. Table 1 displays the validity and reliability of the tool in terms of the Pakistani context. Further, the measurement model in Table 4 exhibits a good fit for the scale.

Control Variables
We have controlled age, educational background, a degree in which they study, and gender to reduce the spurious results of the research. For the categorical variables, we have performed group difference analysis in AMOS. Age and educational background are controlled in the structural model. We have found no significant difference between males and females, and the same was the case for study majors. Therefore, we have excluded them from the analysis. However, age and education have generated mix results that have been discussed in the structural models.

Exploratory and Confirmatory Factor Analyses
Exploratory and Confirmatory factor analyses have been performed to find out the reliability and validity of tools (see Table 1) and Cronbach's alpha was calculated in SPSS ver. 20 (IBM, Armonk, NY, USA) [76]. Then, we have executed convergent validity by taking a square of the standardized Sustainability 2020, 12, 3595 7 of 17 estimate of each item and then dividing the average of squared values by the number of items for each factor. This is the Average Variance Extracted (AVE) which is greater than 0.5 and less than composite reliability (CR). After that, discriminant validity was calculated by comparing maximum shared variance (MSV) and the square root of AVE with inter-construct correlations. It can be seen that AVE > MSV and √ AVE > the inter-construct correlations. Findings indicated that the tools are reliable and valid [77,78].

Common Method Bias
Results derived from a cross-sectional data set might have threats of common method bias [79]. To check this issue, we have applied the Harman One factor test using principal component analysis as an extraction method in SPSS. All the items were included in the process (un-rotated factor solution). The results indicated four factors with eigenvalues greater than one of which the first factor explained 27.20% variance. Additionally, we have also executed analysis by adding a common latent factor to the model. We have reached to the conclusion that the data are free of common method bias as there is no difference between the regression weights of two models (model with CLF and model without CLF).

Descriptive Statistics
Descriptive statistics, mean (M), standard deviation (SD) and normality of the data, are reported in Table 2 Table 3 shows the Pearson correlation that has been executed in SPSS. We have found a significant negative correlation of physical activity with academic anxiety (r = −0.281, p < 0.01), and self-criticism (r = −0.115, p < 0.05); a significant positive relationship between physical activity and self-enhancement (r = 0.347; p < 0.01); a significant positive relationship between self-criticism and academic anxiety (r = 0.248, p < 0.01); and a significant negative correlation between self-enhancement and academic anxiety (r = −0.310, p < 0.01).

Model Fit Statistics
We achieved satisfactory model fits in term of Chi-square/degree of freedom(χ2/df) ≤ 5, Standardized Root Mean Square Residual (SRMR) ≤ 0.05, Root Mean Square Error of Approximation (RMSEA) ≤ 0.05, Tucker-Lewis Index (TLI) ≥ 0.90, comparative fit index ≥ 0.90, as recommended [77,78]. Model fits for all instruments, measurement model and structural model have been shown in Table 4.

Direct Effects
The structural model approach was considered to test the hypotheses of the study. In the first structural model (Figure 3), we have examined the direct influence of physical activity on academic anxiety. The results indicate that physical activity has significant impacts on academic anxiety (β = −0.28, p < 0.001) that supported H1. R2 indicates that physical activity explains a 9% change in academic anxiety when controlled for age and education. In Figure 4, the direct effect of PA on SE is positive (β = 0.35; p < 0.001), while the direct effect of PA on SC is negative (β = −0.12; p < 0.05) supporting H2 and H3. Conversely, direct effect of SE on AA is negative (β = −0.24; p < 0.001), while SC on AA is positive (β = 0.22; p < 0.001) supporting H4 and H5.

Direct Effects
The structural model approach was considered to test the hypotheses of the study. In the first structural model (Figure 3), we have examined the direct influence of physical activity on academic anxiety. The results indicate that physical activity has significant impacts on academic anxiety (β = −0.28, p < 0.001) that supported H1. R2 indicates that physical activity explains a 9% change in academic anxiety when controlled for age and education. In Figure 4, the direct effect of PA on SE is positive (β = 0.35; p < 0.001), while the direct effect of PA on SC is negative (β = −0.12; p < 0.05) supporting H2 and H3. Conversely, direct effect of SE on AA is negative (β = −0.24; p < 0.001), while SC on AA is positive (β = 0.22; p < 0.001) supporting H4 and H5.

Mediation (Indirect Effects)
In the multiple mediation model (Figure 4), we have checked the mediating role of selfenhancement and self-criticism between physical activity and academic anxiety. Model shows that both self-enhancement (β = −0.084, p < 0.05) and self-criticism (β = 0.026, p < 0.05) partially mediate between physical activity and academic anxiety because the indirect influence of

Mediation (Indirect Effects)
In the multiple mediation model (Figure 4), we have checked the mediating role of self-enhancement and self-criticism between physical activity and academic anxiety. Model shows that both self-enhancement (β = −0.084, p < 0.05) and self-criticism (β = 0.026, p < 0.05) partially mediate between physical activity and academic anxiety because the indirect influence of physical activity on academic anxiety remains significant after mediation (β = −0.17, p < 0.001). R square shows that a 19% variation in academic anxiety is explained by physical activity with the mediation of SE and SC after controlling for age and education. Hence, the model supports hypotheses H6-H7.
Sustainability 2020, 11, x FOR PEER REVIEW 11 of 18 physical activity on academic anxiety remains significant after mediation (β = −0.17, p < 0.001). R square shows that a 19% variation in academic anxiety is explained by physical activity with the mediation of SE and SC after controlling for age and education. Hence, the model supports hypotheses H6-H7.

Hierarchical Regression analyses
To verify the validity of results, we have also performed a hierarchical regression for testing the main model. The results of the hierarchical regression (see Table 5) endorsed the structural model developed in AMOS confirming that SE and SC partially mediate between PA and AA.

Hierarchical Regression Analyses
To verify the validity of results, we have also performed a hierarchical regression for testing the main model. The results of the hierarchical regression (see Table 5) endorsed the structural model developed in AMOS confirming that SE and SC partially mediate between PA and AA.

Discussion
This study presents the direct influence of physical activity on academic anxiety as well as the mediating role of self-enhancement and self-criticism. Data of the current study had normal distribution which is an important assumption for conducting analyses and structural models in AMOS graphics [81]. Next, validity and reliability for all tools have been assessed. Results revealed that all tools are valid and reliable in the Pakistani context. The findings are consistent with previous studies where these tools have been tested for validity and reliability in eastern cultures [69][70][71]73,75]. Highlighting the major findings, a key objective of the study was to investigate the direct effect of physical activity on academic anxiety. A negative association of physical activity and academic anxiety of university students in Pakistan was revealed. Current findings are in agreement with a previous study conducted by Khan [25] where a significant inverse relation has been found between physical activity and academic anxiety and stress of university students in Pakistan. Another study has reported a negative association of physical activity and stress of university students in Pakistan [2]. Our finding is also following Khandaza et al. [11] where they claim a strong association between more exercise and low frequency of anxiety and depression among adults in Pakistan.
It is described in the literature that increasing physical activity of students has been the focus of attention among scholars and practitioners because of its positive impacts on physical and psychological health [82]. Our study adds up in the body of knowledge by examining the association between PA and AA in the context of Pakistani students at the university level. Our findings are in line with the previous work [34] in which it was argued that students can reduce their anxiety and can achieve satisfactory academic success through regular physical activity such as sport, hiking, walking, and jogging.
Another aim of our study was to identify the direct relation of physical activity with self-enhancement and self-criticism. It is inferred from the results that there is a positive relation of physical activity with self-enhancement and negative with self-criticism. That is, more physical activity leads to an enhanced self and less self-criticism. It is deducted from previous research [25] that PA is positively linked with the quality of life in Pakistan. Physical activity significantly develops self-enhancement which in turn decreases academic anxiety. It supports the hypothetical notion that the greater the amount of physical activity is, the enhanced the self of an individual will be. The more enhanced one's self is, the lesser will be the academic anxiety.
Other studies support our findings reporting a negative link of self-esteem with anxiety [83,84]; and depression with physical activity [2] of university students in Pakistan. It is in line with the finding that positive feelings, lifestyle, high status, motivation, personal mobility and high quality of life can be achieved through regular physical activity [37]. Further, it has been pointed out in past studies that physical activity helps individuals to sustain their motivation, enhancement, and status [41]. Therefore, enhancing self of students would be a proficient technique to be used by professors for getting improved outcomes and reducing students' stress, anxiety, and pressures at academic work.
To explore the direct relation of self-enhancement and self-criticism with academic anxiety of Pakistani university students, we have found that self-enhancement leads to reduce academic anxiety while self-criticism leads to increase academic anxiety. Although there is less research in Pakistan regarding self-enhancement and self-criticism especially about the academic anxiety of students, there have been less related studies conducted with other samples. For instance, a few studies have reported that low self-esteem leads to depression and anxiety [85,86]; while higher self-concept is associated with lower anxiety of university students in Pakistan [87]. Similarly, a study conducted with the general population in Pakistan [83] claims that self-esteem is negatively related to anxiety. Conversely, despite its positive effects on physical and mental health, self-criticism creates negative thinking, anxiety, and depression [88]. That is why it increases anxiety among students. This research has confirmed that physical activity facilitates students in the reduction of their self-criticism, which leads to reduce anxiety. Indeed, self-criticism negatively influences daily routine activities and can disturb individuals' health [56]. Therefore, one has to exercise regularly to reduce his/her self-criticism. Similarly, Jamil and Khalid [65] scrutinized regular exercise as a significant factor for the low level of depression and negative thinking among Pakistani participants. Our research suggests that self-enhancement should be increased as it helps in decreasing anxiety [53] and self-criticism should be avoided as it leads to a high level of anxiety [89].
Moreover, a major objective of this study was to develop a model exhibiting the mediating effect of self-enhancement and self-criticism on the relationship between physical activity and academic anxiety of university students in Pakistan. The results describe that self-enhancement and self-criticism play a partial mediating role between physical activity and academic anxiety. Further, from the above findings and discussions, we can consider physical activity positively related to self-enhancement [25] which itself is linked inversely with anxiety [85]. On the contrary, PA is negatively linked to depression [2] which is a form of self-criticism [55]. Self-criticism, in turn, increases anxiety and depression [90] in a piece of evidence from eastern culture. It shows that self-enhancement and self-criticism are the potential mediators between physical activity and academic anxiety.
Although the study has several strengths, it has limitations. Self-report measures have been used in this study. Therefore, response bias or method effects could be due to the possible inflation of the relations among variables. Objectives measures are suggested for future research. Another limitation is due to the sample of the study, which was selected conveniently affecting the generalizability of the results. In further studies, it would be helpful to adopt different sampling techniques. Besides, participants had been selected only from the universities in two cities (Islamabad and Rawalpindi). It would be interesting to compare the physical activity and academic anxiety of students in different provinces of Pakistan to examine the variability of results across the regions.

Implications
The current study provides a theoretical background for the benefits of physical activity and exercise for reducing the academic anxiety of students and the mediating role of self-enhancement and self-criticism between the two variables. The study is conducted in an Asian country, Pakistan where self-concept plays a different role than in western cultures, as people pay more attention to the group than to the individual in a collectivistic culture. Various studies have been carried out on the role of self-concept in physical activity but most of them are from western societies. Research on physical activity in Pakistan is highly important, as the rate of physical activity participation is extremely low. Moreover, relating personal factors such as self-enhancement and self-criticism to physical activity can be beneficial for increasing the physical activity of students. Hence, educational institutions are