The influence of exercise self-efficacy and gender on the relationship between exercise motivation and physical activity in college students

Abstract

This study examined the relationships among exercise motivation, exercise self-efficacy, and physical activity rating, as well as the role of exercise self-efficacy in the above relationship. A survey was conducted among 362 college students (aged 19.0 ± 0.8 years, 185 males and 177 females) from a Chinese college using exercise motivation, self-efficacy, and physical activity assessment scales. The results revealed significant sex differences in exercise motivation, self-efficacy, and physical activity scores. Structural equation analysis revealed that ability motivation had a direct effect, accounting for 41.5% of the total effect, while the mediating effect of exercise self-efficacy accounted for 58.5%. Exercise self-efficacy has been proven to be a key predictor of physical activity in both general analysis and gender stratification analysis. From the results of path analysis, in the overall sample and the male group, the influence intensity of each variable on physical activity is the direct effect of exercise self-efficacy, the mediating effect of exercise self-efficacy and the direct effect of ability motivation. However, different influence patterns were observed in female groups: the direct effect of exercise self-efficacy still ranked first, followed by the direct effect of ability motivation, and finally, the mediating effect of exercise self-efficacy. This finding suggests that gender may be an important variable regulating the relationship between exercise self-efficacy and physical activity. Based on the above findings, this study emphasizes that gender-specific strategies should be adopted when formulating intervention programs for physical activity promotion: for male groups, it should focus on strengthening the cultivation of exercise self-efficacy, while for female groups, it should also focus on the improvement of ability motivation and the establishment of self-efficacy. This study provides an important empirical basis for improving the theory of health behavior promotion.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-95704-5.

Introduction

Globally, a lack of exercise and sedentary behavior are common health risk factors for adults. Therefore, improving physical activity and promoting personal physical and mental health are essential issues in public health. Moreover, in the latest general guidelines on physical activity, strengthening physical activity also reduces the duration of a continuous sedentary state¹. Because a person sits for more than 8 h a day, his or her risk of death due to sedentary sitting increases by 20%². According to a survey on the development and future trend of global adult physical inactivity from 2000 to 2020, the prevalence of global adult physical inactivity in 2022 (31.3%) is higher than that in 2000 (23.4%) and 2010 (26.4%). If current trends persist, the global prevalence of physical inactivity will reach 34.7% by 2030. Therefore, improving the level of national physical activity is urgent.

As early as 2020, China Science Daily reported that the number of Chinese college students whose physical health declined due to a lack of physical activity was increasing³. In particular, the new coronavirus outbreak has resulted in significant changes in physical activity. Several studies have shown that college students’ moderate and high-intensity physical activities have decreased, and sedentary behavior has increased significantly^4–8. Lacking of physical exercise leads to a greater risk of obesity⁹ and a lower quality of healthy life¹⁰. Moreover, college students’ lack of physical activity and mental health literacy is related to increased risk of anxiety and depression symptoms¹¹. However, according to the current research, college students do not understand the importance of physical activities in promoting physical and mental health¹². The above situation shows that more attention should be given to college students’ lack of physical exercise. Previous studies have shown that the factors influencing college students’ physical activity are internal exercise motivation^13,14, external motivation^15,16, exercise self-efficacy¹⁷, and the impact of gender¹⁸. However, how the above factors influence college students’ physical activity path is unclear. Therefore, assessing internal and external motivation and the impact of exercise self-efficacy on college students’ physical activity will help guide more effective physical exercise practices in the future.

Gender differences in exercise motivation, exercise self-efficacy and physical activity

Motivation is an implicit behavior between human beings and the outside world, which others can not directly observe, whereas practice is an explicit behavior that others can observe¹⁹. The intrinsic motivation for physical exercise includes the following five aspects: health motivation, appearance motivation, social motivation, ability motivation and fun motivation²⁰. Previous studies have confirmed a significant positive correlation between internal exercise motivation and physical activity^21,22. External motivation refers to the incentives and rewards individuals obtain on the basis of external factors when they are participating in physical activities, such as material rewards, social recognition and external pressure, including institutional motivation and obedience motivation^23,24.

Research shows that fatigue or laziness is a common obstacle to teenagers’ physical exercise^25,26. Boys are more motivated to exercise, whereas girls face more obstacles²⁷. Recent studies have shown that, among internal motivations, the influence of health motivation on Chinese college students’ exercise behavior is significantly greater than that of other motivational factors²⁸. Gender differences in the field of exercise motivation are reflected mainly in men’s greater health motivation²⁹, social motivation, ability motivation and self-directed motivation³⁰, whereas women’s greater interest in appearance motivation^31,32, weight management and health motivation³⁰. Current research has fully confirmed that understanding the motivations that affect exercise participation is crucial for developing interventions to promote a higher level of participation³².

Self-efficacy is the belief and confidence that a person can complete specific activities³³. In recent years, exercise self-efficacy has been used to predict or explain the level of physical exercise or physical activity of various people. Previous studies have shown that male college students’ exercise self-efficacy is significantly greater than that of female college students^34,35. With respect to the level of physical activity or exercise of college students, the intensity of physical activity of men is greater than that of women^30,35. To summarize, we propose Hypothesis 1: gender significantly affects college students’ exercise motivation, self-efficacy and physical exercise level.

Exercise motivation and physical activity

The physical activity rating scale (PARS) is a psychological measurement scale used to evaluate the motivation for physical exercise and is derived from self-determination theory (SDT) and cognitive assessment theory (CET). Prior research has established the robustness and credibility of the scale^36–38. However, these studies have focused mainly on the relationship between internal motivation and physical activity, and Chinese people’s evaluation and research on external motivation are limited. A study conducted by Chen and Yan investigated Chinese college students and verified the reliability and effectiveness of the scale. The results revealed that the lack of internal exercise motivation significantly limits the physical exercise of Chinese college students³⁹. However, Tao reported that controlling (external) motivation significantly predicts Chinese college students’ participation in physical activities. This finding shows that external motivation is crucial in encouraging Chinese college students to participate in physical activities⁴⁰. The meta-analysis of 46 studies also proves this view. Adolescents’ autonomous motivation is significantly positively correlated with exercise, whereas controlled motivation is weakly negatively correlated with exercise⁴¹. A survey of Spanish college students⁴² and high school students⁴³ revealed that high school students’ autonomous motivation was greater than that of college students and that the intrinsic motivation of college students was significantly positively correlated with the level of physical activity. Li suggested that appearance and ability motivation directly and positively affect college students’ exercise behavior. Moreover, ability directly and indirectly affects exercise behavior through the relationship between appearance and ability⁴⁴. Therefore, we propose Hypothesis 2: Internal exercise motivation and external motivation are significantly correlated with college students’ physical activity levels.

The mediating role of exercise self-efficacy in the relationship between motivation and physical activity

There is a consensus that exercise self-efficacy is crucial in predicting or explaining physical exercise among people of different ages. It can explain the physical activities of different populations through direct effects and predict them through intermediary effects^45,46. For example, self-efficacy is the main factor in predicting the physical activity of Australian teenagers, followed by friend support. This study revealed that people with insufficient self-efficacy can compensate for the impact of physical activities through the support of friends⁴⁷. Research on female college students revealed that improving their exercise self-efficacy and perceived benefits can significantly promote their physical exercise behavior⁴⁸. Our latest research revealed that exercise self-efficacy significantly predicts or explains physical activity and that peer exercise support effectively promotes physical exercise behavior or enhances physical activity¹⁸, which is consistent with the survey results of American college students⁴¹. Ouyang, a scholar from mainland China, reported that appearance motivation was significantly correlated with the self-efficacy and exercise participation of college students in Chongqing. Self-efficacy was significantly correlated with sport participation. The mediating role of self-efficacy significantly impacts the relationship between body image and exercise participation⁴⁹. In addition, another study revealed that the mediating effect of exercise self-efficacy was more significant than the direct effect of exercise motivation⁵⁰. Therefore, on the basis of the above-related research, we propose Hypothesis 3. There was a significant correlation between exercise self-efficacy, motivation, and physical activity. Exercise self-efficacy is a significant intermediary between exercise motivation and physical activity.

Influence of sex on the relationships among exercise self-efficacy, exercise motivation and physical activity score

Owing to social roles, expectations, and physiological differences, gender significantly affects exercise health beliefs, motivation, and physical exercise level/physical activity. A cross-sectional survey of Thai adults⁵¹ revealed that women’s structural equation fit better than men’s structural equation fit does, and gender factors play a regulatory role in the relationship between self-efficacy and physical exercise behavior. A meta-analysis⁵² explored the regulatory role of gender factors in the relationship between autonomous motivation and physical activity level. The random effect model analysis of 27 included studies revealed that the regulatory effect of sex was minimal and could be ignored. Previous studies by our research group have shown that gender not only affects college students’ exercise self-efficacy³⁴ and physical activity level³⁵ but also affects the relationships among college students’ peer exercise support, exercise self-efficacy and physical activity level. Therefore, on the basis of the above-related research, we propose Hypothesis 4: there are gender differences in the direct and indirect effects of ability motivation on physical activity.

Previous studies in this field have focused on economically developed coastal cities in East China^53–56 or the provincial capital in Central China^29,57, and investigations of college students in economically underdeveloped areas are lacking. Moreover, previous research reports on college students in this area lack structural equation models constructed with various factors (such as internal motivation, external motivation and exercise self-efficacy) to analyze the path affecting college students’ physical activity. Therefore, this study fully assessed the relationships among internal exercise motivation, external motivation, exercise self-efficacy and the physical activity of college students in Northeast Sichuan Province and evaluated the role of exercise self-efficacy and gender in the relationships among internal exercise motivation, external motivation and the physical activity. These findings provide theoretical support for better guiding the practice of physical exercise inside and outside the classroom.

Research participants and methods

Research participants

Ethical considerations

The subject in this study met the requirements of the “Declaration of Helsinki” on human experiments and was approved by the Academic Ethics Committee of Sichuan University of Arts and Sciences (Approval No. 2023SASULL- 001). All participants provided informed consent and voluntarily completed the questionnaire as part of their involvement in the study. The collected data do not involve sensitive information such as participants’ names and telephone numbers, and we declare that these data are only used for research purposes. According to the basic principles of sampling methods^58–60, In this study, a sample was drawn from 19,000 college students using a simple random selection procedure. Based on the previous calculation method, the minimum sample size was determined to be 378 people. The sample size was finally determined to be 400 people according to the upward rounding principle to ensure representativeness and statistical effectiveness. This adjustment not only meets the requirement of minimum sample size but also enhances the reliability of the study and the universality of the results.

This study recruited 400 college students from a public institution in Western China through simple random sampling, with all participants completing the survey on a voluntary basis. The sample screening exclusion criteria for this study were as follows: (1) only full-time college students, (2) no exercise-related disability or movement disorder. To ensure the high reliability of the information provided by the respondents, before the questionnaire is filled out each time, an administrator of our team, under the guidance of the site, will answer only the questions of the respondents on the questionnaire without providing any inductive tips. The administrator reminded the subjects to complete the questionnaire independently and emphasized that they could not access the data or consult with others. After the respondents fully understood the questionnaire’s contents, they filled it out truthfully, and the administrator collected all the questionnaires on the spot. Finally, 38 questionnaires were abandoned because of incomplete information. Three hundred sixty-two questionnaires were valid, and the effective response rate was 90.5%.

Research methods

Demographic informatics survey

Demographic information, such as height and weight, was included as one of the essential contents of this survey. The specific measurements were performed in the “Laboratory of Sports and Health Promotion.” Monthly household income was also included in this survey (see Table 1). See our previous research report³⁴ for specific evaluation methods.

Table 1.

Basic information of the participants (M ± SD).

Items	Male (185)	Female (177)	P
Age (year)	19.1 ± 0.8	18.9 ± 0.8	0.001
BMI (kg/m2)	21.8 ± 2.4	20.6 ± 2.4	0.001
Monthly income (RMB: Yuan)	4.3 ± 2.6	4.5 ± 2.4	0.587

Exercise motivation survey

This study utilised simplified versions of the Exercise Motivation Scale (Chinese version)²⁹ and the College Students’ External Motivation Scale⁶¹, revised by Chinese scholars. These two scales demonstrated good reliability and validity in a survey conducted among Chinese college students. They were deemed suitable for evaluating the exercise motivation of Chinese college students. The Internal Exercise Motivation Scale (Simplified Chinese version) was composed of five main factors: social motivation, appearance motivation, ability motivation, health motivation, and fun motivation. Three measurement items represented each factor. For instance, the health motivation factor was represented by the statement, " I hope to maintain physical and mental health through physical exercise,” and the fun motivation factor included " I like to participate in physical activities with strong entertainment.” External motivation was categorized into institutional motivation (comprising four factors) and compliance motivation (comprising three factors). For example, the institutional motivation factor included " I participate in physical activity because of institutional requirements.” among the obedience motivation factors, " I exercise only under the pressure of friends or family members.” All data were measured using a five-point Likert scale, with the incentive intensity ranging from “strongly disagree” to “strongly agree.” The Cronbach’s alpha coefficient was employed to assess the reliability of the scale. A coefficient between 0.35 and 0.70 indicated acceptable reliability, a coefficient ≥ 0.70 indicated good reliability, and a coefficient < 0.35 was considered to indicate low reliability, with such dimensions generally being removed⁶². For the initial internal motivation scale, the specific Cronbach coefficients were as follows: social motivation (0.340), appearance motivation (0.124), ability motivation (0.885), health motivation (0.894), and fun motivation (0.805). Based on the aforementioned rules, the two dimensions of social and appearance motivation were removed. Consequently, the Cronbach coefficient of the scale in the study of internal motivation was found to be 0.921, the Cronbach coefficient of the scale in the study of institutional motivation was 0.903, the Cronbach coefficient of the scale in the study of subordinate motivation was 0.851, and the total Cronbach coefficient of external motivation was 0.870. The above data indicated that the internal and external motivation scales were reliable.

Exercise self-efficacy survey

It was revealed by early studies that exercise self-efficacy could significantly predict the physical activity of college students^18,63. The exercise self-efficacy scale was composed of five items, including “It is believed that I can exercise every day.” Likert- 5-point scale, ranging from “strongly disagree” to “strongly agree” (1–5), was used to evaluate students’ attitudes. It has been demonstrated by previous studies that the relationship between peer support and physical activities among college students is significantly mediated by exercise self-efficacy^18,64. The Cronbach coefficient of the scale is 0.859.

Physical activity rating survey

The PARS-3 ⁶⁵ was used to evaluate college students’ physical activity during the previous month. The scale contains three questions, namely, the intensity of physical activities (what is your physical activity or intensity? ), duration (how long do you adhere to the above physical activity intensity? ) and frequency (how many times have you carried out the above physical activities? ). The corresponding score ranges from 1 to 5 physical activity grades = intensity score * (duration score − 1) * frequency score, and the score range is 0 ~ 100. The Cronbach coefficient, which was 0.717, was used to evaluate the scale’s reliability. This shows that the scale is reliable. In addition, the results of this study are consistent with those of previous studies evaluating Chinese college students’ physical activities, as evidenced by the studies of Duan et al., Yue et al., and Zeng et al.,^66–68. These findings further support the reliability and validity of the results of this study.

Data analysis

SPSS 27.0 and Amos 23.0 were used for data analysis. Participants’ demographic characteristics, exercise motivation, self-efficacy, and physical activity levels were summarized using descriptive statistics, presented as means ± standard deviations. Independent-samples t-tests were conducted to assess the differences in exercise motivation, exercise self-efficacy, and physical activity rating between male and female college students, and Pearson correlation analysis was used to evaluate the relationships among the variables. SPSS Amos 23.0 was used to evaluate the fitness of the hypothetical model and construct a structural equation based on exercise motivation, self-efficacy, and physical activity ratings. See our previous research report¹⁸ for the specific steps. The CMID/DF, GFI, AGFI, RMSEA, TLI, IFI, and CFI were used to test the model’s fitness. Amos 23.0 was used to calculate the direct and mediating effects of ability motivation on physical activity ratings, and a multigroup comparative analysis model was used to compare whether gender has a moderating effect; The data were stratified for analysis, and the direct and indirect influences of ability motivation on physical activity ratings were examined. Model 4 in process 3.5 evaluates whether exercise self-efficacy is significant and whether it is a complete or partial mediation effect. Differences were considered significant at P < 0.05.

Results

Basic demographic characteristics

In this survey, the proportion of male college students was 51.1%, whereas the proportion of female college students was 48.9%. Compared with male college students, female college students presented lower BMI and ages, but no significant difference in family monthly income was observed.

Physical activity rating characteristics

Female college students have a lower physical activity rating. Specifically, the average physical activity of male college students is 42.4, the maximum is 100, the minimum is 0, and the median is 48; the average physical activity of female college students is 19.4, the maximum is 100, the minimum is 0, and the median is 12.

Gender differences in exercise motivation and self-efficacy

In terms of internal motivation, male college students presented significantly higher levels of ability, health, and fun motivation than females did (Fig. 1A). With respect to external motivation, females presented higher institutional and obedience motivation than did male college students (Fig. 1B). Furthermore, the exercise self-efficacy of male college students was significantly greater than that of female college students (Fig. 1C).

Fig. 1.

Differences in internal motivation (A), external motivation (B), and exercise self-efficacy (C) between male and female college students.

Relationships among exercise motivation, self-efficacy, and physical activity

The interrelationships among variables were examined using Pearson correlation analysis. Among the demographic factors, only BMI was significantly correlated with exercise self-efficacy (r = 0.196, P < 0.01) and physical activity rating (r = 0.166, P < 0.01) (see Table 2). The ability, health, and fun components’ internal motivation significantly correlate with exercise self-efficacy and the physical activity. In contrast, institutional compliance and external motivation are significantly and negatively correlated with self-efficacy. Exercise self-efficacy was significantly correlated with the physical activity (see Table 3). Among them, ability motivation in internal motivation is negatively correlated with institutional motivation and obedience motivation, and the correlation coefficient between exercise self-efficacy and the physical activity is greater than 0.5. The indicators most closely related to the physical activity are ability motivation and exercise self-efficacy.

Table 2.

Relationships between demographic factors and exercise motivation, exercise self-efficacy, and physical activity (n = 362).

Items	Age (years)	BMI (kg/m2)	Monthly income (RMB)
Ability motivation	0.094	0.092	0.065
Health motivation	0.102	0.097	0.019
Fun motivation	0.060	0.086	0.069
Institutional motivation	– 0.048	– 0.068	– 0.031
Obedience motivation	– 0.026	0.056	0.019
Exercise self-efficacy	0.054	0.196**	0.043
Physical activity	– 0.002	0.166**	0.075

Note: * p < 0.05, * * P < 0.01, *** P < 0.001.

Table 3.

A relationship exists between exercise motivation, self-efficacy, and physical activity (n = 362).

Items	Health	Fun	Institutional	Obedience	Self-efficacy	PAR
Ability	0.671***	0.616***	− 0.240***	− 0.167***	0.592***	0.518***
Health	–	0.730***	− 0.102	− 0.067	0.496***	0.304***
Fun	–	–	− 0.117*	− 0.056	0.393***	0.269***
Institutional	–	–	–	0.435***	− 0.296***	− 0.322***
Obedience	–	–	–	–	− 0.154***	− 0.245***
Self-efficacy	–	–	–	–	–	0.592***

Note: * p < 0.05, * * P < 0.01, *** P < 0.001. PAR: physical activity rating.

The structural model of exercise motivation and self-efficacy affecting physical activity ratings: the mediating role of exercise self-efficacy and the influence of sex

To test our hypothesis further and analysis the relationship between internal and external exercise motivation and exercise self-efficacy on college students’ physical activity, we used a structural equation model for further testing. We included all the factors related to the level of physical activity in the structural equation model. Through step-by-step analysis, we found that only exercise self-efficacy and ability motivation are suitable for establishing a structural equation model with a physical activity. The model fitness parameters are as follows: CMIN/DF = 3.29, NFI = 0.96, RFI = 0.94, IFI = 0.97, CFI = 0.97, TLI = 0.96, RMSEA = 0.08, P = 0.000, indicating that the model is acceptable. When Model 4 in process 3.5 is used for further analysis, the regression coefficient of the mediating effect is 0.260, and the 95% confidence interval is [0.198, 0.328], excluding the number 0. Therefore, the mediating effect of exercise self-efficacy between ability motivation and physical exercise grade is a partial mediator. The direct impact of exercise self-efficacy is greater than that of ability motivation on the level of physical exercise, and the mediating effect of exercise self-efficacy is greater than that of ability motivation (0.67*0.47 = 0.31 > 0.22) (see Fig. 2). The direct effect of ability motivation accounted for 41.5% of the total effect, and the mediating effect of exercise self-efficacy accounted for 58.5%. When we construct the structural models of male and female college students, the regression coefficient of the mediation effect is 0.177, and the 95% confidence interval is [0.096, 0.262], excluding the number 0. Therefore, the mediating effect of exercise self-efficacy on the relationship between ability motivation and physical exercise grade is partial. We found that male college students’ exercise self-efficacy had a greater direct effect on physical activity scores than did ability motivation, including direct and indirect effects (0.58 * 0.40 = 0.23 > 0.16). The influence of male ability motivation accounted for 41.0% of the total influence, and the mediating effect of exercise self-efficacy accounted for 59.0%. For female college students, the regression coefficient of the mediation effect is 0.203, and the 95% confidence interval is [0.109, 0.292], excluding the number 0. Therefore, the mediating effect of exercise self-efficacy on the relationship between ability motivation and physical exercise level is partial. The direct effect of college students’ exercise self-efficacy on physical activity score is greater than that of ability motivation, but the mediating effect of exercise self-efficacy (0.62*0.36 = 0.22 < 0.28) is less than that of ability motivation on physical activity. Female ability motivation accounted for 56.0% of the total effect, and the mediating effect of sport self-efficacy accounted for 44.0%.The structural equation model demonstrated a superior fit for female college students compared to male college students (see Figs. 3 and 4 for details). It is suggested that female college students pay attention to ability training and improve exercise efficiency.

Fig. 2.

Structural equation of the relationships among college students’ ability motivation, exercise self-efficacy, and physical activity rating. Model fitness parameters: CMIN/DF = 3.29, NFI = 0.96, RFI = 0.94, IFI = 0.97, CFI = 0.97, TLI = 0.96, RMSEA = 0.08, P = 0.000.

Fig. 3.

The relationship between ability motivation and physical activity rating among male college students: the mediating effect of exercise self-efficacy. The model fit parameters are CMIN/DF = 2.87, NFI = 0.93, RFI = 0.90, IFI = 0.95, CFI = 0.974, TLI = 0.95, RMSEA = 0.08, and P = 0.000, which are acceptable for the model.

Fig. 4.

The relationship between ability motivation and physical activity rating among female college students: the mediating effect of exercise self-efficacy. The model fit parameters are CMIN/DF = 1.92, NFI = 0.95, RFI = 0.92, IFI = 0.97, CFI = 0.97, TLI = 0.96, RMSEA = 0.07, P = 0.000, indicating good model fit.

Discussion

Main findings of the survey

The primary insights derived from this survey can be encapsulated as follows: (1) There are significant differences in exercise motivation, self-efficacy, and physical activity ratings between male and female college students. (2) Internal and external motivations for exercise are significantly correlated with college students’ exercise self-efficacy and physical activity ratings, and exercise self-efficacy is positively correlated with physical activity ratings. (3) Ability motivation and exercise self-efficacy significantly predict the level of physical activity. Exercise self-efficacy significantly mediates the relationship between ability motivation and physical activity rating. In general, the mediating effect of exercise self-efficacy is more remarkable than the direct effect of ability motivation. However, in the female group, the direct effect of ability motivation on the physical activity rating was greater than the mediating effect of exercise self-efficacy.

Influence of gender on exercise motivation, self-efficacy, and physical activity rating

Prior research has indicated that gender and age exert differential impacts on exercise motivation, self-efficacy, and levels of physical activity^69–71. For example, people over 60 years of age mainly exercise to prevent disease and cultivate a good mood. Women pay more attention to appearance motivation, social motivation and weight management, whereas men pay more attention to fitness motivation and ability motivation⁷⁰. However, a previous study showed that gender did not significantly affect exercise motivation. These findings emphasize that being unmarried and having no chronic metabolic diseases are necessary for improving exercise motivation⁷². This may be because unmarried people and people without chronic metabolic diseases have more free time, more energy and better physical quality. Our study revealed that there were significant sex differences in exercise motivation. Compared with female college students, male college students have greater health motivation, ability motivation and fun motivation, whereas female college students have greater motivation. A previous studies by scholars⁷³ have shown that social support is crucial for sedentary people to start exercising and that understanding the benefits positively impacts motivation. Therefore, future research must further evaluate the relationship between perceived benefits and exercise motivation.

A previous study revealed that the exercise self-efficacy of male college students was greater than that of female college students¹⁸. The results of this study are consistent with our previously reported conclusions. According to a survey of college students majoring in music, those with stronger motivation, higher perceived ability to overcome obstacles and a greater sense of self-efficacy reported a greater frequency of exercise participation. In contrast, female college students reported greater efficiency in overcoming exercise obstacles⁷⁴. Research has shown that self-efficacy improves the level of physical activity of black and white girls, which suggests that self-efficacy should be encouraged as a targeted intermediary variable in intervention measures to increase girls’ physical activity⁷⁵.

The influence of gender factors on physical activity is multifaceted and involves biological differences, social and cultural expectations, psychological attitudes and environmental factors. For example, men usually have high muscle content and cardiopulmonary endurance, which makes them perform better in some physical activities⁷⁶. In terms of social and cultural expectations, men can be encouraged to participate in competitive and team sports, whereas women may be more inclined toward aerobic exercise and fitness activities⁷⁷. Men and women may have different attitudes and self-efficacy toward physical activities, affecting their physical exercise participation. Men may be more confident and competitive, whereas women may be more concerned about the social and health benefits of physical activities⁷⁸.

Relationships among exercise motivation, self-efficacy, and physical activity rating

A prior research has consistently highlighted the critical role of intrinsic motivation and the enjoyment of physical activities in sustaining long-term engagement in physical exercise⁷⁹. Teenagers from the UK⁸⁰ and middle school students from Hong Kong, China⁸¹ reported a significant positive correlation between autonomous motivation and physical activity. They suggested that initiative is more critical in extracurricular physical exercise. The possible mechanism is that when individuals feel that the three psychological needs of autonomy, ability and relationships are met, they are more likely to generate intrinsic motivation and actively participate in physical exercise⁸². A cross-sectional survey of college students in Shanghai, China, revealed that the impact of health motivation on exercise behavior was significantly greater than that of other motivation factors². In contrast to previous studies, our study revealed that the influence of ability motivation on exercise behavior was more significant than that of other motivation factors and that health motivation ranked second. Internal motivation exhibited a significant positive association with both exercise self-efficacy and physical exercise performance. A significant negative correlation between institutional motivation and obedience motivation and exercise self-efficacy, and a positive correlation between exercise self-efficacy and physical exercise performance. Therefore, although the respondents are all college students from China, the difference may be that the former⁸³ is mainly from economically developed Shanghai (China’s coastal cities), whereas our respondents are mainly from economically underdeveloped Sichuan Province (China’s western province).

College students majoring in physical education have greater cognition and sports skills than ordinary college students do. They have greater motivation to enjoy physical exercise, and their intrinsic motivation is positively correlated with physical activity⁸⁴. This finding shows that the motivation to enjoy physical exercise is more important than exercise self-efficacy in predicting physical activity⁸⁵. Compared with intrinsic motivation, our study revealed that exercise self-efficacy is the most critical factor in explaining physical activity. The order of the relationships between intrinsic motivation and physical exercise performance is ability motivation, health motivation and fun motivation. This finding aligns with prior research findings⁸⁶, indicating that extrinsic motivation (supervision or system) may be the critical determinant for undergraduates to increase exercise and improve the utilization of campus entertainment facilities. Because college students have more independent time to control and there are no system restrictions, their physical activities have difficulty meeting the standards recommended by the World Health Organization.

Boosting self-efficacy and intrinsic motivation significantly increased women’s intention to engage in future physical exercise. Groups with high self-efficacy maintain their interest in and participation in exercise behavior⁸⁷. Even in the UK, the most active patients with osteoarthritis reported higher levels of motivation and self-efficacy⁸⁸. Scholars suggest that compulsory courses or health and physical education plans should help students understand the self-efficacy barriers of physical activities before graduation to maximize their enthusiasm for physical activities⁸⁹.

Exercise self-efficacy can not only predict the exercise motivation of American urban minority middle school students⁹⁰ but also help improve the exercise self-efficacy of American college students by increasing physical activity⁴¹. Successful exercise experience can improve personal self-efficacy, whereas failure experience may reduce self-efficacy⁹¹. People with high exercise self-efficacy are more likely to set practical exercise goals and improve their self-efficacy after they achieve these goals⁹². When facing difficulties or challenges, individuals with higher levels of exercise self-efficacy tend to persist in their physical activities rather than abandoning them⁹³. People with high self-efficacy may obtain more fun from exercise because they feel that they can gain more control and a sense of achievement through physical exercise⁹⁴.

The relationship between college students’ exercise motivation and physical activity rating: the mediating role of exercise self-efficacy and the influence of gender

The level and type of motivation directly affect the frequency, intensity and duration of individual participation in physical exercise. When people take part in exercise because they like exercise themselves, they are more likely to exercise continuously and regularly. Prior research has identified a link between intrinsic motivation, higher levels of commitment, and reduced dropout rates⁹⁵. Exercise driven by external rewards, social recognition, or punishment avoidance may not last as long as intrinsic motivation. However, extrinsic motivation can catalyze the start of a movement⁸². Setting specific exercise goals can improve motivation and the likelihood of achieving these goals. Goal-oriented motivation is associated with increased exercise compliance⁹². Self-efficacy or confidence in one’s physical ability can enhance exercise motivation because individuals believe that they can complete tasks⁹⁶.

Self-efficacy plays a crucial role in regulating the relationship between planning and behavior⁹⁷, as people with high self-efficacy benefit from planned intervention. If a person has low self-efficacy, the initial plan may not effectively promote physical activity. Early studies have shown that exercise self-efficacy mediates exercise motivation and physical activity⁹⁸. In recent years, scholars have investigated the psychological factors that prevent the reduction of physical activity in the early stage of the new coronavirus pneumonia pandemic and reported that intrinsic motivation may not predict physical activity because the new coronavirus pneumonia limits the opportunity to exercise in the way of generating positive emotional experiences (i.e., intrinsic rewards)⁹⁹. A survey of Japanese adults revealed that exercise self-efficacy regulates the relationship between goals and physical activities in health promotion programs¹⁰⁰. This study further shows that exercise self-efficacy regulates goal setting and increases physical activity. The development of a workplace in an environment that promotes physical activity may be more conducive to improving physical activity.

A survey of 164 college students with social support in Korean social culture revealed that changes in physical activity directly affect physical activity and that the self-efficacy of physical activity indirectly affects physical activity through changes in physical activity¹⁰¹. This study investigated 30 healthy adults (12 males and 18 females) and revealed that intrinsic motivation and schedule self-efficacy were significantly correlated with energy consumption and moderate and severe physical activity duration. These findings indicate that the relationships among motivation, self-efficacy and physical activity may change quickly. It may be necessary to regularly evaluate motivation and self-efficacy in the intervention plan to promote healthy adults’ continuous participation in physical activity¹⁰².

Our research revealed that exercise self-efficacy plays an essential intermediary role in the relationship between ability motivation and physical activity score and that exercise self-efficacy is the most critical predictor of physical activity, which is consistent with previous research conclusions¹⁰³. Among the internal and external factors of exercise motivation, only ability motivation can significantly predict the level of physical exercise, which is inconsistent with the findings of previous studies. Regarding the overall sample size and the findings for male college students, the mediating effect of exercise self-efficacy was more pronounced than the direct influence of exercise motivation on physical activity levels, this observation corroborates the conclusions drawn in prior research⁵⁰. Our results do not indicate that gender affects the relationship between ability motivation and physical activity level, which is consistent with the findings of the previous studies⁵². In a previous study, a survey of college students in Spain revealed that internal exercise motivation mediates the relationship between gender and exercise¹⁰⁴. In the present study, the structural equation model of female college students revealed that the mediating effect of exercise self-efficacy on the level of physical activity was less than the effect of ability motivation, which was one of the new findings of this study.

Research has found that in the overall sample, the mediating effect of self-efficacy is greater than the direct effect of ability motivation. However, in gender grouping, the mediating role of self-efficacy still dominates in the male group. At the same time, the direct effect of ability motivation on physical activity is more significant in the female group. This gender difference may be related to physiological, psychological, and socio-cultural factors¹⁰⁵. Specifically, men typically have higher muscle strength and endurance physiologically¹⁰⁵, which may make them more reliant on self-efficacy stimulation during exercise¹⁰⁶. Self-efficacy can enhance their confidence in their athletic ability, promoting physical activity through indirect pathways such as increased motivation. In contrast, women experience physiological changes such as increased body fat, accelerated bone growth, and hormonal fluctuations after puberty, which may impact their athletic performance¹⁰⁵. Therefore, women may rely more on direct motivation (such as external incentives or health goals) to participate in physical activities, rather than solely relying on self-efficacy.

From a psychological perspective, men may be more inclined to obtain a sense of achievement and satisfaction through self-efficacy¹⁰⁷, which can effectively motivate them to participate in physical activities. Women may be more concerned with external motivations such as social, health, or appearance improvement. Research has shown that women’s motivations are more external and hostile, such as to lose weight or gain recognition from others. This external motivation may drive their participation in physical activities rather than mediating through self-efficacy. Socio-cultural factors also have an impact on this gender difference. Society has higher expectations for men’s exercise abilities, which may encourage them to focus more on improving their self-efficacy. For example, men may be more inclined to gain social recognition by showcasing their athletic abilities in physical activities. Women may be more concerned about exercise’s social and health value than their physical abilities. In addition, women may be more influenced by environmental factors such as exercise facilities and social support¹⁰⁷.

In terms of practical significance, intervention strategies should be designed for gender differences. For men, it is important to focus on improving their self-efficacy by setting reasonable goals and providing positive feedback to enhance their confidence in exercising¹⁰⁸. For women, more attention should be paid to stimulating external motivation, such as through social activities, health education, or improving the exercise environment, to increase their participation in exercise directly. Secondly, optimize the cultivation of exercise motivation. For men, enhancing self-efficacy indirectly increases their motivation to exercise, forming a virtuous cycle. For women, directly stimulating external motivation, such as through health lectures and sports clubs, may be more effective in helping them establish positive exercise habits. Finally, in terms of promoting mental health, men can enhance their self-efficacy, self-confidence, and sense of achievement, thereby promoting mental health. Women can indirectly improve their mental health by satisfying external motivations, such as improving their appearance or obtaining social support.

In summary, the role of self-efficacy and ability motivation in physical activity varies by gender. Men rely more on the mediating role of self-efficacy, while women rely more on the direct role of ability motivation. This difference may be related to physiological, psychological, and socio-cultural factors. Therefore, when designing exercise intervention programs, gender differences should be fully considered to improve intervention effectiveness.

Potential confounding factors and how they affect gender differences

Cultural expectations, socio-economic status, social support, and environmental factors play important roles. Cultural expectations profoundly influence gender participation patterns in physical activities. Traditional beliefs often encourage men to participate in physical activities that showcase strength and competitiveness. At the same time, women may tend to engage in low-intensity activities and even be expected by society not to overly emphasize physical abilities¹⁰⁹. This cultural expectation not only shapes the differences in exercise motivation and self-efficacy between men and women but may also lead to men being more actively involved in physical activities. In addition, socio-economic status (SES), as a key factor, significantly affects individuals’ opportunities to participate in physical activities. Research has shown that individuals with high SES typically have more resources and opportunities, while individuals with low SES may face resource constraints¹¹⁰. More complexly, the interaction between gender and SES may further exacerbate this difference, and for example, high SES women are often better able to overcome cultural barriers and actively participate in physical activities than low SES women. In addition, social support and environmental factors such as family, peer, and school support also play important roles in gender differences. Research shows that men typically receive more social support to participate in physical activities.

In contrast, the availability and safety of exercise facilities significantly impact women’s participation in physical activities. Based on these factors, the interpretation of the research results shows that in the male population, the direct and mediating effects of exercise self-efficacy are relatively strong, which may be related to men’s greater emphasis on ability performance and competitiveness in physical activities. In the female population, the direct impact of ability motivation is particularly prominent, indicating that women are more inclined to participate in physical activities based on intrinsic motivation (such as health or social) rather than external performance¹¹¹. In addition, the mediating role of exercise self-efficacy in the female population is relatively weak, which may reflect more external barriers women face when participating in physical activities, including insufficient cultural expectations and social support. These factors constitute a complex background of gender differences, providing an important theoretical basis for further research.

Limitations of this study

Although this study validated the relationship between ability motivation and physical activity through a cross-sectional quantitative survey and revealed the mediating role of exercise self-efficacy, some limitations remain. Firstly, the sample of this study was only from a public college in Sichuan Province, western China, so the applicability of the conclusions may be greatly limited and cannot be extended to private institutions in eastern China. In order to improve the universality and representativeness of research conclusions, future studies should consider collecting data from different geographical regions and types of universities, increasing the sample size, and achieving sample diversity. Secondly, the data collection in this study mainly relies on self-report and questionnaire surveys, which may, to some extent, lead to subjectivity and memory bias in the information. To overcome this issue, various data collection methods, such as observation and physiological indicator measurement, can be used in future research to reduce subjective bias and improve the objectivity of research results. In addition, due to the observational design of this study, determining causal relationships is relatively complex. Although mediation analysis provides some clues, the dose-response relationship between exercise self-efficacy, motivation changes, and physical activity remains unclear. Therefore, future research can track and investigate the same population at different time points through longitudinal design, more clearly revealing the causal relationships between these variables and providing a more profound perspective for understanding the complex relationship between ability motivation and physical activity. By integrating different samples and adopting a longitudinal study design, future research will be able to better comprehensively examine the relationship between various types of motivation and exercise behaviour, thereby laying a more solid theoretical foundation for developing effective intervention strategies. Fourthly, although established theoretical frameworks and mature scales were used, new perspectives or variables were not introduced to deepen the research content. Although this study examined gender differences, it did not further explore other potential moderating variables (such as social support, psychological stress, and health beliefs). Future research should consider introducing moderating variables such as social support, psychological stress, and health beliefs and adopt multidimensional, cross-cultural, and longitudinal research designs to explore gender differences and their influencing mechanisms in-depth and focus on specific groups or backgrounds, such as investigating physical exercise behavior in specific groups (such as sports majors, obese students) or special situations. Fifth: This study only uses quantitative research methods, which have the advantages of precise data analysis, reproducible results, and broad applicability. However, their limitations lie in deeply understanding participants’ subjective experiences and complex situations. Therefore, combining qualitative research methods can provide a more comprehensive perspective and enrich the understanding of research phenomena.

Conclusion

This study shows a significant correlation between college students’ physical motivation, institutional motivation, obedience motivation, physical activity rating and exercise self-efficacy, especially a significant positive correlation between exercise self-efficacy and physical activity level. The study revealed that exercise self-efficacy is a stronger mediator than direct ability motivation in influencing exercise participation. In addition, gender difference analysis showed that the mediating effect of exercise self-efficacy in the male group was significantly more substantial than the direct effect of ability motivation. In contrast, for the female group, the direct effect of ability motivation was significantly more significant than that of self-efficacy. These show that the direct or indirect effect of disability motivation is less than the direct effect of exercise self-efficacy on physical activity, which emphasizes the importance of improving college students’ exercise self-efficacy in promoting their physical activity level. Based on these findings, the future practice of physical exercise inside and outside the classroom should pay more attention to cultivating and improving exercise self-efficacy. By designing various physical activities and exercises, students’ exercise skills and physical quality can be enhanced, and their sense of self-efficacy can be effectively improved. In addition, teachers should help students overcome challenges with self-confidence through positive feedback, encouragement, and support to stimulate their enthusiasm for physical activities further. Therefore, physical exercise inside and outside the classroom is not only a means to improve students’ physical quality but also an important way to enhance exercise self-efficacy.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Jiazhi Sheng wrote the main manuscript text, and Indang Ariati Binti Ariffin and Jacquline Tham provided academic guidance. All authors reviewed the manuscript.

Funding

This research was funded by the Zigong Key Research Base of Philosophy and Social Sciences, Sports and Health Innovation Research Center Project (YDJKY23-29) and Taji Talent Project Funding.

Data availability

Data is provided within the manuscript or supplementary information files.

Declarations

Competing interests

The authors declare no competing interests.