Introduction

Language learning is stereotyped as a feminine domain where females are depicted as having more gifts, engaging more, and obtaining higher academic performance than their counterparts (Espinoza & Strasser, 2020). Such shared beliefs concerning gender differences are gender stereotypes in language learning (Master & Meltzoff, 2020; Plante et al., 2013). In particular, gender stereotypic beliefs prevail in collectivistic settings such as China (Yao, 2011), where traditional gender roles are deeply ingrained. Upon entering university, societal expectations for males and females to conform to different roles become more pronounced than they were in secondary school, exacerbating the prejudice against males in learning English. Male students studying languages at the undergraduate level are constantly scorned as lacking masculinity (Chaffee et al., 2020; Li & McLellan, 2021). Those stereotypes shape students’ beliefs and influence their perceptions of themselves as well as their learning outcomes (Master & Meltzoff, 2020).

In recent years, the pervasive influence of gender stereotypes on language learning has garnered considerable attention (Chaffee et al., 2024). Researchers observed that gender stereotypes determine students’ attitudes toward language learning (e.g., self-efficacyFootnote 1 and anxiety) and language proficiency (Kutuk et al., 2021). Nonetheless, there are still gaps that need to be addressed. On the one hand, previous studies have yielded inconclusive results regarding gender variations in EFL learning. A camp of researchers argued the females’ advantages in EFL learning (Bai & Wang, 2023) while others revealed non-significant gender differences (Truong & Wang, 2019). There is a need to address the contradictory findings regarding gender effect (Voyer & Voyer, 2014).

On the other hand, little attention has been devoted to the underlying mechanisms of female-advantage-in-language stereotypes. Gender stereotypes may translate into explicit beliefs and learning outcomes. Motivational beliefs and emotional factors, among the most essential personal factors, are likely to mirror students’ gender stereotypes and ultimately shape their engagement in language learning (Kutuk et al., 2021). Hence, we surmised that the emotional factors and motivational beliefs of EFL learners may work synergistically in mediating the effect of gender stereotypes on engagement.

In light of the research gaps mentioned above, this study aimed to scrutinize the gender differences in EFL learners’ stereotypic beliefs, motivational beliefs, emotional factors, and engagement in foreign language learning; and to identify the mediating effects of motivational and emotional factors on the relationship between gender stereotypes and learning engagement. The study could provide an in-depth understanding of gender differences in stereotypes, motivational beliefs, emotional factors, and engagement, as well as their nexus. Theoretically, the study attests to the propositions of the social cognitive theory with data from Chinese EFL learners. Pedagogically, this study could be informative for tailoring intervention targets for eliminating or at least mitigating students’ gender differences in EFL learning and directing avenues for promoting an equitable learning environment.

Literature Review

Social Cognitive Theory

Social cognitive theory provides a useful lens to understand the dynamic relationships among personal, environmental, and behavioral processes (Bandura, 1997). It posits triadic reciprocal relations among these processes (Schunk & DiBenedetto, 2020). The personal process encompasses cognitions, beliefs, perceptions, and emotions while the behavioral process includes activity choice, effort, persistence, and academic achievement (Schunk & Usher, 2019). The environmental process incorporates social norms, instruction, and feedback (Schunk & DiBenedetto, 2020).

Given its robustness, social cognitive theory has been extensively utilized to explore language learning processes (Li et al., 2023; Teng, 2024). Researchers who concentrated on gender stereotypes in language learning have also drawn on the social cognitive theory (Kutuk, 2023; Kutuk et al., 2021). However, few studies have examined intricated relationships among gender stereotypes, motivational beliefs, emotional factors, and engagement. In the current study, we sought to offer a nuanced insight into the influences of gender stereotypes through the lens of the social cognitive theory. Specifically, we delved into the potential roles of motivational beliefs and emotional factors in connecting gender stereotypes and EFL learning engagement. In our model, gender stereotypes, along with motivational beliefs and emotional factors, are considered personal processes, while engagement belongs to the behavioral process.

Gender Differences in Language Learning

Gender differences in language learning have garnered significant attention from students, teachers, and policymakers alike. A large body of research has underscored the language learning advantages of females (Denies et al., 2022). As reported by the OECD (2019), males lag behind females in various literacy skills involving reading, vocabulary learning, and writing. A meta-analysis by Voyer and Voyer (2014) declared that females reported better academic performance in language learning than males all over the world from 1950 to 2011. Wucherer and Reiterer (2018), for example, noted that male and female learners appeared to possess varying strengths in different areas of language learning (i.e., grammar and pronunciation learning).

Apart from differences in academic achievement across genders, differences in motivational beliefs and emotional factors have been also reported. A great deal of literature has highlighted the gender differences in motivational beliefs encompassing task values and competence beliefs in foreign language learning (Durik et al., 2006). Researchers noted that females attached greater personal importance to foreign language learning and showed higher competence beliefs than males (Li et al., 2023; Truong & Wang, 2019; Chaffee et al., 2020).

Mixed results were obtained regarding gender differences in foreign language learners’ emotional factors (i.e., enjoyment and anxiety). Some researchers reported that females were more likely to hold positive emotions such as enjoyment and interest, whereas males tended to show greater anxiety, apathy, and boredom in foreign language learning (Carr & Pauwels, 2006). Other researchers reported otherwise (Dewaele et al., 2016). For example, Dewaele et al (2016) indicated that females reported higher levels of anxiety although they experienced more enjoyment in foreign language learning.

Furthermore, a substantial volume of research has also been conducted to understand gender differences in foreign language learning engagement (Oga-Baldwin & Nakata, 2017). For example, Oga-Baldwin and Nakata (2017) noted that males tended to be disengaged while females showed high engagement in language learning activities. Overall, prior research reported females and males differed in diverse aspects of language learning. Following the previous studies, we hypothesized:

Hypothesis 1: Females and males would differ in gender stereotypes, motivational beliefs, emotional factors, and engagement in EFL learning.

Gender Stereotypes and Engagement in Language Learning

Gender stereotypes pertain to conventions that certain personality traits, skills, characters, and preferences are assigned to an individual based on their biological sex (Lips, 2020). Gender stereotypes in language learning typically refer to beliefs that females are more gifted, persevering, and high-achieving than males (Kollmayer et al., 2018). These female-advantage-in-language stereotypes are prevalent in many countries such as Australia (Carr & Pauwels, 2006), Canada (Plante et al., 2013), Chile (Espinoza & Strasser, 2020), Germany (Steffens & Jelenec, 2011), and China (Li & McLellan, 2021). As noted by Carr and Pauwels (2006), many learners believed that language learning did not align with the masculine ideal and viewed it as ‘uncool’ for males.

The stereotypical notion has a considerable impact on language learners’ academic achievement (Carr & Pauwels, 2006; Kutuk et al., 2021; Li & McLellan, 2021). Highlighting the negative influence of gender stereotypes, the stereotype threat effect suggests that members of a particular group may experience apprehensions about confirming unfavorable stereotypes. Consequently, some efforts are inevitably devoted to inhibiting negative emotions, causing performance decrements (Pansu et al., 2016). According to the stereotype threat effect, males, as members of a stigmatized group in language learning, may experience undermined motivation, engagement, and academic performance due to the negative stereotypes (Li & McLellan, 2021). They are more likely to suffer from higher anxiety caused by internal pressure to resist or disprove stereotypes (Steele et al., 2002).

To date, some attempts have been made to explore whether the belief that “language learning is a feminine domain” would affect foreign language learners’ efforts, motivational beliefs, as well as academic performance (Kutuk, 2023; Kutuk et al., 2021; Li & McLellan, 2021). For example, Li and McLellan (2021) inquired into the gender stereotype of EFL learners in Chinese secondary schools. They reported both males and females tended to stereotypically believe that females excel in aptitude, affect, and achievement. Males’ performance was adversely affected by gender stereotypes, and the females showed the opposite result (Kutuk et al., 2021; Li & McLellan, 2021). In another study, Kutuk (2023) found that gender stereotypes had spectrum effects on EFL learning. Despite the great contribution of their studies, the underlying mechanism of how gender stereotypes influenc EFL learning is still underinvestigated.

Engagement refers to the amount and type of effort to participate in learning tasks (Hiver et al., 2021). It is one of the pivotal steps in language learning (Oga-Baldwin, 2019). Consequently, the engagement of EFL learners deserves special attention. However, scant research has probed into to what extent the variance in engagement could be explained by students’ gender stereotypes.

Motivational Beliefs and Emotional Factors as Mediators

Motivational beliefs such as task values and competence beliefs were identified as pivotal correlates to EFL achievement (Li et al., 2022; Ma et al., 2018; Wang & Sun, 2020). Task values refer to the quality of a task and contribute to the probability that an individual will select the task (Eccles, 2005). Competence beliefs concern an individual’s beliefs about their competence in doing a particular task (Durik et al., 2006). They are two major components of expectancy-value theory (Wigfield & Eccles, 2000) and operate synergistically. Task values and competence beliefs determine goal-setting, task persistence, and self-regulation in goal-pursuing (Schunk & Pajares, 2009). When language learners value a task or obtain higher competence beliefs, they become more engaged in the language learning task, persist longer, and achieve higher academic performance (Eccles & Wigfield, 2020; Li et al., 2024).

Students’ emotional factors such as enjoyment and anxiety were also empirically connected to language learning (Zhong et al., 2023). Enjoyment has been recognized as a positive predictor of engagement (Zeng, 2021), while anxiety has been consistently associated with lower levels of engagement (Zhang et al., 2020).

Both motivational and emotional factors have been found as vital precursors of students’ learning outcomes (Denies et al., 2022), however, in much of the literature examining the effects of gender stereotypes on language learning, motivational and emotional factors were overlooked (Li & McLellan, 2021). To the best of our knowledge, Kutuk et al.’s (2021) study is the only one that unraveled the “black box” by considering competence beliefs (i.e., self-efficacy) and anxiety in examining the effects of gender stereotypes on Turkish EFL language learners’ attainment. They demonstrated that gender stereotypes shaped EFL learners’ language proficiency through self-efficacy but not through anxiety. Given past literature, we postulated that:

Hypothesis 2: Gender stereotypes would be related to learners’ engagement indirectly through the mediation of motivational beliefs and emotional factors in EFL learning.

The Current Study

The study aimed to examine (1) differences in gender stereotypes, motivational beliefs, emotional factors, and engagement in EFL learning across genders; and (2) the relationships among gender stereotypes, motivational beliefs, emotional factors, and engagement in EFL learning. The following hypotheses were put forward:

Hypothesis 1: Females and males would differ in gender stereotypes, motivational beliefs, emotional factors, and engagement in EFL learning.

Hypothesis 2: Gender stereotypes would be related to learners’ engagement indirectly through the mediation of motivational beliefs and emotional factors in EFL learning.

Method

Participants

After receiving ethical approval from the University of Macau, a total of 521 students were recruited from mainland China by using the convenience sampling method. Invalid cases (n = 48) were deleted due to incomplete responses. The final sample size was 473 students, among them, 261 were females and 212 were males. Most students were registered in their second and third years of university. Their age ranged from 18 to 24 (M = 20.22; SD = 1.64). Students from different majors were included, with 105 (22.20%) enrolled in literature, 84 (17.75%) in science, 91 (19.24%) in management, 92 (19.45%) in engineering, 76 (16.07%) in education, and 25 (5.29%) in other disciplines. At the time of data collection, students’ average experience in English language learning was 12.60 (SD = 2.60) years. There were 325 students (68.7%) who disclosed that English was the medium of instruction in their English language course.

Measures

Gender Stereotypes. Students’ gender stereotypes were measured with 12 items from two subscales (female domain and male domain), each subscale comprises 6 items (Plante et al., 2009). The female domain subscale and male domain subscale assess the degree to which students agree with the notion that language learning favors females or males. Students were asked to circle from 1 (strongly disagree) to 7 (strongly agree). An example item for the female domain subscale is “Females are more gifted than males in learning English”, and an example item for the male domain subscale is “Males like English more than females”. Both females’ and males’ gender stereotypes were represented by subtracting the total scores endorsed on the male domain from the total scores endorsed on the female domain. A positive score indicates that respondents hold the females’ advantage-in-language stereotypes, whereas a negative score indicates the opposite. Cronbach’s alpha coefficients for the female domain and male domain subscales were acceptable (a = 0.83 and a = 0.83, respectively).

Task Values. Six items were derived from the subscale of the Motivated Strategies for Learning Questionnaire by Pintrich et al. (1991). Students reflected on their beliefs and indicated how much they value English learning by using a 6-point Likert scale, with 1 representing “strongly disagree” and 6 representing “strongly agree”. An example of items measuring task values is “I think I will be able to use what I’ve learned in the English course in other courses.”. The internal consistency of students’ responses to this scale was 0.84.

Competence Beliefs. As the conceptually analogous nature of self-efficacy and competence beliefs, we employed self-efficacy to represent competence beliefs (Eccles & Wigfield, 2020). Twelve items were adopted to measure students’ competence beliefs in learning English (Wang & Bai 2017). It contains four subscales: listening (e.g., “I can understand stories told in English”), speaking (e.g., “I can introduce myself in English”), reading (e.g., “I can understand English news on the Internet”), and writing (e.g., “I can send emails in English”). Each subscale has 3 items. Students were invited to evaluate their competence in performing certain English language learning tasks by using a 7-point Likert scale, with 1 indicating “I cannot do it at all” and 7 indicating “I can do it”. The internal consistency of students’ responses to this scale was 0.91.

Enjoyment. Students’ enjoyment in language learning was measured with the Short Form of the Foreign Language Enjoyment Scale which was developed by Botes et al. (2022). It consists of 9 items from 3 dimensions: personal enjoyment (e.g., “I enjoy learning English”), teacher appreciation (e.g., “My English teacher is encouraging”), and social enjoyment (e.g., “We form a tight group in English class”). Students scored on a 5-point Likert scale, with 1 representing “strongly disagree” and 5 representing “strongly agree”. The internal consistency of students’ responses to this scale was 0.82.

Anxiety. Five items were employed to measure students’ anxiety in learning English (Botes et al., 2022)2. An example of items measuring language learning anxiety is “Even if I am well prepared for English class, I feel anxious about it”. Students responded on a 6-point Likert scale, with 1 representing “strongly disagree” and 6 representing “strongly agree”. The internal consistency of students’ responses to this scale was 0.89.

Engagement in Language Learning. Students’ engagement in language learning was assessed with 12 items from 3 aspects, including behavioral engagement (e.g., “I listen very carefully in English class”), emotional engagement (e.g., “I enjoy English classroom activities”), and cognitive engagement (e.g., “I try to understand the foreign language”). These 12 items were from Engagement Scale Items for Language Classrooms by Oga-Baldwin (2019). Items were rated from 1 (strongly disagree) to 6 (strongly agree). The internal consistency of students’ responses was 0.91.

Data Analysis

Preliminary analysis. Data analyses were conducted using Amos. Confirmatory factor analyses (CFAs) were performed for items measuring task values, competence beliefs, enjoyment, anxiety, and engagement separately. A series of goodness-of-fit indices were used to represent the fit of models to data, involving standardized root mean square residual value (SRMR), root means square of approximation (RMSEA), comparative fit index (CFI), and the Tucker-Lewis Index (TLI). To indicate a satisfactory fit of models to data, SRMR and RMSEA should not exceed 0.08, and CFI and TLI should be higher than 0.90 (Byrne, 2010). After the psychometric properties of instruments were confirmed, means (M), standard deviations (SD), and Cronbach’s alpha coefficients were calculated and reported. Pearson correlation analyses were used to reveal correlations among gender stereotypes, task values, competence beliefs, enjoyment, anxiety, and engagement. Furthermore, measurement invariance tests which evaluate whether the scales functioned similarly in different groups were carried out to ensure meaningful comparisons between females and males. By imposing additional constraints in sequence, we tested the configural (i.e., factor structure), metric (i.e., factor loading), and scalar (i.e., factor mean) invariance of the scales. When △CFI ≤ 0.01, △RMSEA ≤ 0.015, and △SRMR ≤ 0.03, the metric invariance was established. When △CFI ≤ 0.01, △RMSEA ≤ 0.015, and △SRMR ≤ 0.01, the scalar invariance was established.

Main analysis. Multivariate analyses of variance (MANOVAs) were used to reveal mean differences in females’ and males’ gender stereotypes, motivational beliefs, emotional factors, and engagement. We used Tukey’s pairwise testing method to perform Post hoc analyses. Guiding by the thresholds set forth for research in second language learning by Plonsky and Oswald (2014), partial eta squared equal to 0.06, 0.16, and 0.36 corresponds to small, medium, and large effect sizes, respectively.

Structural equation modeling (SEM) were carried out to detect the mediating effect of motivational beliefs and emotional factors. The bias-corrected bootstrapping (2000 bootstrapping sample) with 95% confidence intervals was adopted to examine the direct effects and indirect effects in the model. Furthermore, to test whether the mediation model varies across genders, we employed multi-group SEM with two steps. First, the “unconstrained” model, or the freely estimated model with no equality constraints across genders was tested. Second, we further performed the “constrained” model by adding the constraint that the path coefficients be equal.

Result

Results of CFA indicated that the structure of all questionnaires fit the data well (Table S1). Table 1 displays descriptive statistics including means, standard deviations, internal consistencies of students’ responses to the questionnaires, and correlations among gender stereotypes, motivational beliefs, emotional factors, and engagement for males and females separately. Results of Pearson correlation analysis showed that most variables were statistically significantly correlated, with r ranging from − 0.22 to 0.34 for females and − 0.49 to 0.44 for males. As shown in Table 2, configural, metric, and scalar invariance across genders were achieved for all variables, ensuring the comparability of the females’ and males’ responses to the scales (Fig. 1).

Table 1 Descriptive statistics
Full size table
Table 2 Measurement invariance across genders
Full size table
Fig. 1
figure 1

Mediating roles of motivational beliefs and emotional factors for the relationship between gender stereotypes and engagement. ***p < .001, **p < .01, *p < .05; Coefficients for females before the dash and males after

Full size image

Differences in Gender Stereotypes, Motivational Beliefs, Emotional Factors, and Engagement

Results of MANOVAs indicated differences in engagement, task values, competence beliefs, and enjoyment between females and males, except for gender stereotypes and anxiety. First, both males and females conceived foreign language learning as a domain that is more suited for females. No statistically significant differences were noted between their stereotypic beliefs, F (1, 471) = 0.26, p > 0.05, partial η2 = 0.001. Second, the gender difference in English learning engagement was prominent, F (1, 471) = 13.02, p < 0.001, partial η2 = 0.03. Females were more engaged in learning English compared to males. Third, significant gender differences in students’ perception of task values, F (1, 471) = 35.26, p < 0.001, partial η2 = 0.07, and self-perceived competence, F (1, 471) = 8.64, p < 0.01, partial η2 = 0.02 were found. Precisely, females valued learning English more and obtained higher competence beliefs compared with males. Finally, females showed more enjoyment in English, F (1, 471) = 24.17, p < 0.001, partial η2 = 0.05, but gender differences related to EFL learning anxiety were not statistically significant F (1, 471) = 1.20, p > 0.05, partial η2 = 0.003. In conclusion, compared with females, males valued less on language learning, obtained lower competence beliefs, enjoyed less in foreign language learning, and had a lower engagement. Nevertheless, females and males did not differ in their gender stereotypes and anxiety in foreign language learning.

Mediating Roles of Motivational Beliefs and Emotional Factors

Table 3 displays the path coefficients of the models. As expected, most paths were statistically significantly different from zero, indicating that the mediating roles of task values, competence beliefs, and anxiety between gender stereotypes and engagement in language learning were confirmed. For females, the direct effect of gender stereotypes on students’ language learning engagement was statistically significant (β = 0.20, p < 0.001). The total indirect effect of gender stereotypes on students’ language learning engagement through task values, competence beliefs, enjoyment, and anxiety was also statistically significant (β = 0.13, p < 0.001). Table 3 also depicts the mediating effects of task values (β = 0.06, p < 0.001), competence beliefs (β = 0.02, p < 0.01), and anxiety (β = 0.02, p < 0.05), on the relationship between gender stereotypes and engagement.

Table 3 Path coefficients of the SEM model
Full size table

In terms of males, the direct effect of gender stereotypes on students’ language learning engagement was statistically significant (β = − 0.21, p < 0.001). The indirect effect of gender stereotypes on students’ language learning engagement through task values, competence beliefs, enjoyment, and anxiety was statistically significant (β = − 0.33, p < 0.001). More specifically, males’ gender stereotypes affected their engagement through task values (β = − 0.04, p < 0.01), competence beliefs (β = − 0.07, p < 0.001), enjoyment (β = − 0.16, p < 0.001), and anxiety (β = − 0.07, p < 0.001).

Furthermore, results of multi-group SEM indicated that both the unconstrained model and the constrained model had an acceptable fit to the data (Unconstrained model: RMESA = 0.02 [0.01, 0.03], SRMR = 0.05, CFI = 0.98, TLI = 0.98; Constrained Model: RMESA = 0.03 [0.03, 0.04], SRMR = 0.08, CFI = 0.95, TLI = 0.95). In comparing the constrained model to the unconstrained model, △χ2 = 154.47, △df = 25, p < 0.001, evidencing that the model differs by gender.

Discussion

The study aimed to delve into differences in language learning stereotypes, and other related learning outcomes across genders. We also investigated the mediating roles of motivational beliefs and emotional factors in the relationship between gender stereotypes and engagement in EFL learning. We did not find evidence for differences in language learning stereotypes but identified significant differences in engagement, task values, competence beliefs, and enjoyment across genders. Additionally, we confirmed the impacts of gender stereotypes on engagement through motivational and emotional factors in both genders.

Differences in Gender Stereotypes, Motivational Beliefs, Emotional Factors, and Engagement

Group-mean differences in engagement (small effect size), task values (small to moderate effect size), competence beliefs (small effect size), and enjoyment (small effect size) were all statistically significant across genders, with females reporting higher values (supporting H1). These echo previous findings and add empirical evidence to the scarce literature on gender differences in EFL learning (Dewaele et al., 2016).

Surprisingly, differences in gender stereotypes were not statistically significant, which means that females and males uniformly regarded English as a feminine domain. The endorsement of the gender stereotypes is possibly due to the females’ superiority in foreign language learning, which reinforced students’ gender stereotypes in learning English. Another explanation might be the pervasive influence of cultural norms emphasizing gender roles, making female-advantage language learning stereotypes widely acknowledged and internalized regardless of gender (Master & Meltzoff, 2020).

In addition, their anxiety did not differ significantly either, denoting that both females and males worried about making mistakes and experienced a similar level of anxiety in the foreign language classroom (Jiang & Dewaele, 2019). Perhaps, this can be attributed to the utilitarian atmosphere in the collectivistic settings, wherein both female and male students are under great pressure and are expected to succeed in learning English (Hiver et al., 2021).

Mediating Roles of Motivational Beliefs and Emotional Factors

Gender stereotypes were indirectly associated with students’ engagement through the mediation of task values, competence beliefs, and anxiety in language learning (supporting H2). Such parallel mediating effects indicate that EFL learners’ engagement is contingent on the internalization of gender stereotypes reflected by their task values, competence beliefs, and anxiety. Notably, the mediation effects of task values were most potent for females, while enjoyment had the strongest mediation effect for males. This demonstrates the importance of cultivating task values for females and fostering enjoyment for males in the influences of gender stereotypes on students’ engagement in learning English.

An intriguing finding is that females’ gender stereotypes were positively associated with their task values, competence beliefs, enjoyment, and engagement, and negatively related to their anxiety, yet this pattern was reversed in the males’ group. This suggests that males tend to hold low motivation, show negative feelings, and make less effort in language learning when they regard themselves as members of the disadvantaged group. Instead, when females believe that foreign language learning is a feminine domain, they are more likely to demonstrate high motivation, positive feelings, and engagement. In other words, males’ engagement in learning English was indirectly hampered by socially negative stereotypes towards them through motivational beliefs and emotional factors, whereas females exhibited the opposite result (Kutuk et al., 2021; Li & McLellan, 2021; Steele et al., 2002). These findings confirm the stereotype threat effect, which emphasizes that the members of negatively stereotyped groups’ performance could be impaired (Pansu et al., 2016). Considering the significant influences of gender stereotypes, it is necessary to let both instructors and students be aware of the impact of gender stereotypes on language learning. Indeed, motivational beliefs and emotional factors should be taken into account in boosting students’ engagement in foreign language learning, especially for males whose enjoyment and motivational beliefs were undermined. For example, strategies that enhance students’ beliefs in their competence, improve their values on tasks, increase their enjoyment of foreign language learning, and ease anxiety should be integrated into classroom instructions (Truong & Wang, 2019; Wang & Bai, 2017).

Interestingly, while enjoyment did not mediate the relationship between females’ gender stereotypes and engagement, it significantly mediated this link among males. This means that males’ gender stereotypes exert a greater impact on enjoyment in learning English and consequently play a more important role in influencing engagement compared with female students. This might be attributed to the potential conflicts with notions of masculinity and female-advantage-in-language gender stereotypes. Due to this conflict, males’ enjoyment may be compromised by their gender stereotypes. On the contrary, females’ enjoyment might be more affected by other factors such as the relevance of teaching materials and relatedness satisfaction.

Limitations and Directions for Future Research

Several limitations have to be acknowledged. First, data were collected at a single time point, hence, how students’ gender stereotypes change over time and how these changes interact with motivational beliefs, emotional factors, and engagement are yet to be investigated. Future researchers are recommended to use longitudinal approaches such as the cross-lagged modeling approach to observe changes and to unfold the dynamic interplay among these factors. In addition, intervention studies aimed at alleviating or reducing gender stereotypes in learning English merit further consideration. For instance, future researchers could explore training male students to adopt more positive attitudes toward learning English, introducing successful male role models, and fostering a positive growth mindset among students.

Second, we solely relied on self-report questionnaires in the current research, albeit with high reliability and validity. Future researchers may integrate some qualitative measures such as interviews and classroom observations. For example, organizing interviews can inspect how females and males perceive and internalize gender stereotypes, while performing classroom observations can reveal how these stereotypes manifest in real-time interactions and behaviors. By integrating the quantitative and qualitative approaches, a more in-depth understanding of gender stereotypes and their associated factors can be unveiled.

Third, the study only investigates the role of the personal process (i.e., gender stereotypes, motivational beliefs, and emotional factors) and the behavioral process (i.e., engagement). The environmental process was not incorporated into the model. Future research could benefit from encompassing environmental factors such as expectations from significant others, teachers’ gender stereotypes, school climate, and peer relationships.

Implications

Despite the above-mentioned limitations, this study has some implications. Theoretically, the study adds empirical evidence to the social cognitive theory by ascertaining the relationship between personal (i.e., gender stereotypes, motivational beliefs, and emotional factors) and behavioral (i.e., engagement) processes among Chinese EFL learners. The interplay of gender stereotypes and EFL learning outcomes showcases the robustness of social cognitive theory in explaining the complexity of the language learning process.

Pedagogically, the study shed light on the profound impact of gender stereotypes on EFL learning among Chinese learners. Our findings call for special attention to the prevalent stereotypic beliefs about foreign language learning among students. The teacher is recommended to promote a more equitable language learning experience for students. Particularly, they should adopt strategies to minimize this stereotypic conception, which is detrimental to males’ foreign language learning engagement via impairing their beliefs of competence, values endorsement, and enjoyment in foreign language learning. To buffer the effect of stereotypic beliefs favoring females in EFL learning, males should be given support to boost their competence belief by offering positive verbal encouragement. By affirming their abilities and potential for success in learning English, teachers can help counteract the impact of gender stereotypes. Teachers can also consider emphasizing the practical benefits of learning English, increasing enjoyment, and easing anxiety (Li et al., 2023). For example, teachers are advised to help students understand the utility of English in job hunting and global communication (Ma et al., 2023). They can organize engaging activities, incorporate interactive tasks, utilize multimedia materials, and harness advanced language learning technologies to enhance students’ enjoyment. Reducing students’ fear of failure, adopting formative assessment, and providing scaffolding to ease EFL learning anxiety would also be beneficial (Feng et al., 2023; Hwang et al., 2024).

Conclusion

This study revealed the differences in gender stereotypes and their correlates entailing motivational beliefs, emotional factors, and engagement among Chinese university students. Moreover, the study uncovered that gender stereotype in EFL learning was associated with engagement through heightened or decreased motivational and emotional factors, respectively. The findings of this study underscore the significance of addressing males’ gender stereotypes in foreign language learning and highlight the necessity of making concerted efforts to bolster task values, competence beliefs, and enjoyment, and eliminate anxiety.