Learning another language is not a one-dimension processing but it is a multifaceted process in which a combination of various emotional and non-cognitive factors are also called for Oga-Baldwin et al. (2019). This has opened the gate for a plethora of studies investigating different affective factors such as self-efficacy (Young Kyo, 2022), burnout (Xing, 2022), resilience (Chaigneau et al., 2022), foreign language teaching anxiety (Aydin, 2016), self-esteem (Gaybulloyevna, 2022), boredom (Pawlak et al., 2021), psychological wellbeing (Jonas and Hall, 2022), and academic optimism (Lelieur et al., 2022). Also, key subjects of the inquiry in individual differences are beliefs and attitudes, physical and mental behaviors reflections and actions, learning strategies and styles, move and motivations, talent, mindsets, personality traits, and self-regulation (Dörnyei and Skehan, 2003). The results of the abundance of such papers led to a large number of ideas governing new rules in the field of L2 acquisition (Dewaele, 2009).

However, recent studies seem to have suppressed the old ones which used to focus only on threatening factors by shifting the idea from negative psychological factors toward positive ones (MacIntyre et al., 2019). This movement has paved the way for scholars to examine the role of individual differences and emotions under the new scope of positive psychology which has led to the emergence of numerous studies in this regard (e.g., Seligman et al., 2009; Gregersen et al., 2014; Mercer et al., 2016; Dewaele et al., 2019; Wang et al., 2021). The results of the previously mentioned studies illuminate that the process of L2 teaching can be more fruitful and L2 learning might be fostered significantly when emotions and individual differences are taken into account (Dewaele and Li, 2020).

The construct of mindsets or implicit/lay theories was introduced by Dweck (2006). Mindsets refer to one’s rudimentary belief systems about different human’s traits such as personality factors (Mercer and Ryan, 2010). It is widely recognized that teachers’ mindsets play a crucial role in their job achievements (Dweck, 2014). Furthermore, these mindsets tend to form and reform teachers’ beliefs of failure and success about their approach to coping with different challenges in L2 education (Cacali, 2019). Since the learning process itself is a complex one; therefore, the construct of mindset is of high significance in L2 education (Nguyen et al., 2022). Mindsets are divided into two categories of fixed and growth mindsets (Dweck et al., 1995). A fixed mindset is referred to as a belief that one’s skills, attributes, personality, and intelligence are fixed and cannot be changed whereas a growth mind is concerned with a belief that the aforementioned factors are malleable and can be enhanced (Haimovitz and Dweck, 2017). Growth mindsets refer changeable beliefs of a person of his/her own abilities when coping with setbacks (Schroder et al., 2019) which are systematically associated with endeavors, attitudes, aims, and learning strategies (Molden and Dweck, 2006). Language mindset pertains to the personal opinions about the changeability of language learning (Shirvan et al., 2021).

The studies on humans beliefs about whether they have the capacity to learn a second or foreign language have recently received much attention among L2 researchers who are interested in investigating if a person considers one’s attributes as something fixed or changeable (Lou and Noels, 2019). Mindsets are interrelated with many affective factors such as motivational factors (Lou and Noels, 2019), autonomy (Henriksen et al., 2018), wellbeing and perseverance (Zeng et al., 2019), willingness to communicate (Zarrinabadi et al., 2021), and speaking anxiety (Ozdemir and Papi, 2021). Research on learners’ mindsets seems to be well-documented and it is well acknowledged that these affective factors play a significant role in learners’ achievement; however, this topic still needs under-investigated for teachers since the literature is quite fragmented in this regard (Bostwick et al., 2020). According to Yeager and Dweck (2020), it can be argued that growth mindset’s impacts are replicable, fruitful, and theoretically well-documented for both teachers and learners. With regard to the significance of this construct in educational contexts, Zhang et al. (2017) synthesized the articles published between 1998-2017 on the relationship between learners’ academic achievement, their mindsets, and teachers’ mindsets. It was revealed that students’ mindsets play a crucial role in determining their academic achievement; however, teachers’ mindsets were found to have played a mediating role, not a determining a role. In another study, Derakhshan et al. (2022) investigated the relationships between classroom climate, growth mindset, and students’ engagement where boredom acted as a mediator. The findings indicated that learners’ growth mindset indirectly impacted students’ engagement and also, classroom climate and boredom were the significant predictors of learners’ engagement.

Additionally, it is argued that the mindset can cause shifts in one’s emotions concerning various scopes which are classified into growth mindset and fixed mindset (Yeager et al., 2022). Flexibility in teachers’ mindset is referred to as a growth mindset, whereas inflexibility in their mindset is defined as a fixed mindset. Teachers holding a growth mindset consider personality traits and personal teaching skills as a changing phenomenon (Heggart, 2015). On the other hand, those who hold a fixed mindset believe personality attributes and skills as something unmalleable, while growth mindset holders believe that smartness and intelligence can be fostered, whereas fixed mindset holders consider it as an unchanging construct (Zeng et al., 2019). Also, Frondozo et al. (2020) hypothesized that a teacher’s mindset alongside their emotions can function as a perdictor of teachers’ work engagement. The outcomes indicated that teachers with a growth teaching mindset showed a higher work engagement with teaching enjoyment as a mediating factor. In another study by Patrick and Joshi (2019), it was hypothesized that teachers’ prior ideas about the learners and learning have an impact on teachers’ mindset. It was revealed that teachers’ growth mindset is significantly correlated with students’ mindset about their own skills, personality, and attributes. The authors also found that teachers’ mindset was able to enhance or deteriorate students’ beliefs about their intelligence or skill. Also, Gutshall (2014) conducted a study on the mindset of pre-service teachers since the beginning of their practicum. The findings indicated that their mindset remained stable over their preparation program. Irie et al. (2018) also explored teachers’ mindset on their own profession. They found that teachers cared more about their own technical knowledge rather than interpersonal skills since these skills were regarded as natural.

The interest in positive psychology has paved the way for investigating positive emotions as well as the negative ones that foreign language learners and teachers might experience (Dewaele et al., 2018). This picture has changed the frame of many studies into positive psycho-affective factors (Hagenauer and Hascher, 2014). For instance, the concept of enjoyment was a reaction towards the negative factor of anxiety (Zeng, 2021). Foreign language enjoyment (FLE) can be considered as an example of positive emotions (Pekrun, 2006). It is defined as a general good feeling after overcoming an obstacle in a stressful situation (Li et al., 2018). Those who are concerned with enjoyment feel safer in various contexts and do their tasks in a controlled manner (Pekrun et al., 2007). In this vein, learning enjoyment can be defined as the joyful experience of learning through which one can develop a controlled and safe learning environment (Mierzwa, 2019). The construct of enjoyment is a multidimensional construct and includes five underlying components: emotion, movement, cognition, eloquence, and physiology (Hagenauer and Hascher, 2014). Among these components, emotional or the affective one refers to the pleasure feeling and the sense of joy in learning contexts, hence it fosters affective and cognitive factors in the learning environment that ultimately rises learners’ language achievement (Mierzwa, 2019).

Recently, foreign language teaching enjoyment was introduced to the literature from the studies on foreign language enjoyment by Proietti Ergün and Dewaele (2021). This construct is defined as the broad feeling of positive emotions experienced by teachers in spite of existing obstacles in the context of foreign language teaching (Noughabi et al., 2022). In this regard, Mierzwa (2019) conducted a mixed-methods study among 89 Polish teachers of English. Web questionnaires were used for data collection. It was discovered that teachers experience not only learning enjoyment, but also foreign language teaching enjoyment regardless of exterior factors such as age, residential area, or gender. In another study, Proietti Ergün and Dewaele (2021) the role of well-being and resilience as predictors of FLTE was examined among 174 Italian teachers of English as a foreign language. Their findings indicated that resilience was a strong predictor of FLTE. In another study, Xiao et al. (2022) tested a structural model of work engagement, self-efficacy, and teaching enjoyment among 315 EFL instructors. Their findings demonstrated that both predictors affected teachers’ work engagement, although teacher self-efficacy turned out to be a more powerful predictor than teaching enjoyment. Also, employing 296 English language teachers, Fathi and Naderi (2022) examined the roles of teacher resilience and FLTE in influencing teaching enjoyment. Their results indicated that FLTE was a significant predictor of teacher work engagement.

Wei et al. (2019) probed the interplay between grit, L2 performance, and L2 anxiety among 832 middle-school learners in China. The findings revealed that the learners’ grit fosters L2 performance and indirectly enhances FLE. In addition, Lee (2022) tried to shed light on the relationship between grit, classroom enjoyment, and willingness to communicate. The data analyses showed that grit and classroom enjoyment predict willingness to communicate significantly, which supports the idea that positive psychology plays a significant role in fostering the outcome of the class. Moreover, Liu and Wang (2021) discovered that grit, FLE, and L2 performance were significantly correlated with each other. Also, enjoyment was mediated by grit significantly. Moreover, Wei et al. (2019) concluded that grit impacts L2 performance greatly and enjoyment mediated between grit and L2 performance. Also, Jiang and Dewaele (2019) carried out a mixed-methods study to examine FLE and L2 anxiety of 564 EFL learners outside China. Regression analysis showed that FLE was predicted by the variables related to the teachers, while L2 anxiety was related to student-related factors.

As previously highlighted, various studies have examined affective factors in the literature of L2 education (Dewaele, 2020). One of these psycho-affective factors that have received much attention is grit (Sudina et al., 2021). Grit, conceptualized as consistent interests and perseverant efforts for fulfilling long-term aims, can play a crucial role in learning another language (McCain, 2017). Duckworth and Quinn (2009) asserted that grit has the same level of significance as aptitude in predicting learners’ success, raising the likelihood of achieving more than the natural ability (Duckworth and Quinn, 2009). This positive social-affective trait was introduced by Duckworth and her colleagues in the University of Pennsylvania (Oxford and Khajavy, 2021). Recently, its debate has gained momentum among researchers of various disciplines (Wei et al., 2020). Applied linguistics has not been an exception and some papers (Khajavy et al., 2021; Sudina and Plonsky, 2021; Sudina et al., 2021) have investigated L2 grit among learners and teachers. This construct has been acknowledged as an important construct affecting L2 learning and teaching (Teimouri et al., 2020). Since mastering another language requires much endeavor, persistence, and enduring interest in the process of learning, grit is viewed to be a vital variable as the learners may face numerous failures (Khajavy et al., 2021). In the same vein, a teacher’s profession is full of tension and stress that leaves various impacts on teachers’ emotional state (Griffith et al., 1999). Under the pressure of teaching, teachers may either show persistence in pursuing their goals or demonstrate a lower motivation; therefore, teachers’ grit can help teachers to sustain their passion and perseverance in achieving their aims (McCain, 2017).

Concerning teachers’ grit, some studies have been conducted. For instance, Argon and Kaya (2018) explored whether the potential effects of personal factors on teachers’ grit. Their findings revealed that only gender played a pivotal role in affecting grit and the other demographic factors failed to exert significant effects on teachers’ grit. They reported low levels of teacher grit for their participants as they believed that teachers’ needs, wishes, and expectations were not met. However, a mixed-methods study on African-American male teachers conducted by Yates et al. (2015) yielded a different outcome. It was illustrated that there was a significant relationship between a teacher’s grit, age, GPA, and their life partner. In the qualitative phase, three themes were also found to be dominant, which included: (a) family, (b) circumstances, and (c) spirituality.

In a recent study, Sudina et al. (2021) tried to develop and validate a questionnaire for measuring grit among teachers. For this, 202 English language teachers were recruited as the participants. Their findings led to a valid and reliable L2-specific grit scale. Also, Azari Noughabi et al. (2022) modelled the relationship between EFL teachers’ L2 grit, engagement, and immunity. Having distributed the electronic versions of the scales among 369 EFL teachers, they found that L2 grit and engagement impacted immunity substantially. Faravani (2022) also explored teacher’s self-efficacy, grit, and continuing professional development. 204 EFL teachers were selected through convenience sampling. SEM results indicated that grit and self-efficacy acted as powerful predictors of continuing professional development. In another study, Fabelico and Afalla (2020) tested a model of teachers’ self-efficacy, grit, and burnout. Their findings indicated that grit was associated with other variables and significantly impacted teachers’ educational attainments.

Also, Nazari and Alizadeh Oghyanous (2021) investigated the relationship between job’s stress, turnover intention, wellbeing, and grit among EFL teachers. The findings revealed that these factors were significantly correlated but the associations were stronger for novice teachers. Also, the results of interviews showed that career’s stress affected emotions and perseverance which were under the impact of institutional and socioeconomic status. However, in another mixed-methods study, McCain (2017) examined the role of teachers’ grit in affecting learner’s grit and their reading achievement. Survey, observation, and interviews were used to collect the data. The findings demonstrated that there was no significant relationship between the aforementioned factors.

Work engagement is related to a pleasant mindset about one’s own profession characterized by vigor, dedication, and absorption (Schaufeli and Bakker, 2004), which is associated with reflection, creativity, willingness to help colleagues, and respect towards career (Bakker, 2008). In educational contexts, teacher work engagement is defined as an affective variable indicating instructors’ voluntary devotion of physical, cognitive, and emotional resources to teaching-specific practices (Klassen et al., 2012). Scholars have introduced various models of work engagement (Choochom, 2016). For instance, Saks (2006) proposed the engagement of the employees is characterized as job and institutional involvement that might have antecedents (institutional support and critiques) and also consequences (the outcomes of work engagement). Given this, some studies (Rothmann and Hamukang’andu, 2013; Iyer, 2016; Burić and Macuka, 2018) have investigated this psycho-affective factor in educational contexts. Moreover, it is argued that more engagement on the teachers’ side contributes to the further engagement and academic achievement of the learners (Zhang and Yang, 2021).

Hultell and Gustavsson (2011) examined the factors affecting teacher burnout and engagement among Swedish beginning teachers. Their findings revealed that the demands of the job directly predicted burnout, whereas job resources were positively correlated with work engagement. Also, Hakanen et al. (2006) investigated burnout and work engagement among teachers via employing Job Demands–Resources Model in a sample of 2038 teachers. They hypothesized that there were two predictors of teacher burnout or work engagement: energetical processes (demanding tasks of a job) and motivational processes (work engagement processes). The data analysis with SEM showed that both processes affected teacher burnout or engagement; however, energetical processes predicted teacher burnout more significantly than motivational processes. In addition, motivational processes affected work engagement more significantly. Additionally, Høigaard et al. (2012) examined teacher efficacy and work engagement and their association with burnout, job satisfaction, and desire to quit among teachers with six years of experience or less. 750 questionnaires were administered to the teachers. It was found that teacher efficacy and work engagement positively predicted job satisfaction and negatively affected teacher burnout. In another study, Xiao et al. (2022) examined a structural model of the teacher self-efficacy, work engagement, and work enjoyment among EFL teachers. To this end, 315 EFL teachers completed an online survey. SEM results showed that self-efficacy and teaching enjoyment were substantial predictors of work engagement, although self-efficacy played a more significant role in affecting work engagement.

In another study, Li et al. (2015) investigated the interconnection between work engagement and value congruence. It was hypothesized that controlled and autonomous motivations play the mediating role. To this end, 767 teachers filled out the questionnaires and the data were analyzed with SEM. It was discovered that there was an indirect relationship between value congruence and work engagement and the mediating effects of the two proposed variables were also verified. In a similar study, Bakker and Bal (2010) tested the model of weekly work engagement among 54 teachers. It was hypothesized that job resources in a week are positively associated with levels of work engagement. Also, they hypothesized that momentary work engagement had a positive impact on weekly work engagement. Multilevel analyses confirmed the hypotheses that autonomy was reported as a positive predictor of work engagement. Additionally, exchanging ideas with supervisors and chances for personal growth were predictors of work engagement.

As was mentioned earlier, research dealing with the association between growth mindset, teaching enjoyment, work engagement, and teacher grit still calls for further empirical inquiry. Moreover, despite that the constructs of growth mindset, teaching enjoyment, work engagement, and teacher grit have been, to greater or smaller extents, the scope of various studies carried out in the realm of foreign language education, to our best knowledge, no study has so far sought to examine the relationships between and/or among the constructs in question. Based on the theoretical and empirical considerations discussed above, a structural model, specifying the interconnections between the constructs (i.e., growth mindset, teaching enjoyment, work engagement, and teacher grit) was hypothesized and is depicted in Figure 1.

In light of the findings of Nalipay et al. (2021), it was hypothesized that teaching enjoyment directly impacts teachers’ work engagement; therefore, a path from teaching enjoyment to teachers’ work engagement was drawn. In addition, a path from teaching enjoyment to L2 teacher grit was conjectured based on the empirical study by Elahi Shirvan et al. (2021), in which enjoyment positively affected L2 grit. Also, based on the empirical findings by some studies (e.g., Singh and Chopra, 2018; Azari Noughabi et al., 2022), we hypothesized that teacher grit significantly predicts work engagement, thus, a path from grit to teaching engagement was also drawn. In addition, according to the Broaden-and-Build theory (Fredrickson, 2001) in which not only positive feelings shape emotional resources, but they also build cognitive ones which can serve as pacifiers for long-term negative emotions, and also consistent with the findings of Hochanadel and Finamore (2015), we hypothesized that growth mindset can impact teacher grit. Moreover, in light of the nature of the construct and some previous studies (e.g., O’Neal et al., 2018; Ma et al., 2020; Lan et al., 2021; Lee and Ha, 2022), we hypothesized that grit acts as the mediator for the impact of teaching enjoyment on teaching engagement. Likewise, due to the characteristics of the constructs and based on the findings of some studies (e.g., Özhan, 2021; Kim et al., 2022), it was hypothesized that grit mediates between the impact of growth mindset on work engagement of EFL teachers. Concerning the hypothesized model, the following hypotheses were proposed:

In this study, a total number of 486 Chinese EFL teachers took part as the participants. Among the participants, 189 were male and 297 were female English instructors who were engaged in teaching English at schools, universities, and language academies in Chinese provinces. Given availability, convenient access, and willingness criteria, the participants were selected based on convenience sampling. They were selected based on convenience sampling procedure. Their age range fell between 21 and 44 years (M = 25.03, SD = 6.74) and their teaching experience varied from 2 to 21 years of teaching experience (M = 7.21, SD = 5.37).

As the first step to collect the data for this non-experimental study, an electronic survey which included a section for demographic information as well as another section for the four scales was constructed using WeChat application. Creating the electronic version of the questionnaires helped the researchers to collect data from various regions in China. The link of the questionnaires was shared on a large-scale with various teachers in online space. Also, some colleagues and friends cooperated with the first researcher in data collection. As the participants were English teachers and possessed an adequate command of English proficiency, the English versions of the scales were used and there was no need to translate them into Chinese. The necessary explanation on how to complete the items was included in the beginning of the electronic survey. Informed consents were obtained from all participants who were also ensured about the confidentiality of their data. The participation was also announced to be voluntary and the participants could give up at any stage that they wished. Overall, the data collection started in March 2022 and took about 9 weeks to collect all the data.

SPSS 22 and AMOS 23 were used to analyze the collected data in this study. Then SEM was employed to test the hypothesized model in this study. Prior to performing SEM, confirmatory factor analysis (CFA) was used to test the measurement models in order to verify the construct validity of the used scales. Regarding the evaluation of the model fit, a number of goodness-of-fit indices were used: Chi-square divided by degree of freedom (χ²/df), Comparative Fit Index (CFI), Tucker–Lewis Index (TLI), and Root Mean Square Error of Approximation (RMSEA). A model is regarded to have acceptable fit in case χ²/df < 3, CFI and TLI ≥ 0.90, and RMSEA ≤ 0.08 (Kline, 2011).

As the initial screening, the collected data were checked for missing values, outliers, and non-normal data. Concerning the missing data, we employed Expectation-Maximization algorithm as a data imputation technique (Kline, 2011). Univariate outliers were determined by converting all the scores to standard scores for each construct. Moreover, we used Mahalanobis D² to determine multivariate outliers. Skewness and kurtosis indices were used to examine the data normality and values falling out of ±2.0 were treated as non-normal data (Kunnan, 1998). The outliers and non-normal values were discarded prior to running CFA and SEM. Table 1 depicts the number of valid cases for each construct.

To verify the validity of the scales used in this research, the measurement models were tested via conducting CFA. Table 2 indicates the goodness of fit indices for each construct.

As some models failed to demonstrate adequate fit to the data, some modifications were made to the models, in which included the omission of two items of work engagement, two teacher grit items, and one item of teaching enjoyment scales as these items’ factor loadings were low. The revised model indicated good fit to the data (see Table 3).

Concerning the internal consistency of the questionnaires, the computed coefficient alphas were all higher than 0.70, approving the acceptability of their reliability (Hair et al., 1998) (see Table 2). Then, descriptive statistics and correlations among the constructs were calculated (Table 4).

Employing the maximum likelihood technique and variance-covariance matrices as input, we tested the hypothesized model via AMOS program. The calculated fit indices demonstrated a good fit of the model to the data, thereby confirming all the hypotheses in the final model (see Figure 2). Also, effect size (ES) (Cohen’s f²) was calculated to gain a better understanding of the obtained results.

As illustrated in Figure 2, teaching enjoyment, grit, and growth mindset were the directly predicted work engagement of EFL teachers. Among these constructs, teaching enjoyment was the strongest direct predictor of work engagement (β = 0.44, R² = 0.19, f² = 0.23, large effect size). Growth mindset (β = 0.37, R² = 0.13, f² = 0.15, medium effect size) and teacher grit (β = 0.32, R² = 0.10, f² = 0.11, medium effect size) also directly predicted teachers’ work engagement. In addition, teaching enjoyment influenced work engagement indirectly through the mediation of teacher grit (β = 0.41 × 0.32 = 0.13, R² = 0.001, f² = 0.001, small effect size). Likewise, growth mindset impacted work engagement indirectly through teacher grit (β = 0.36 × 0.32 = 0.11, R² = 0.01, f² = 0.01, small effect size). Also, it was revealed that teaching enjoyment (β = 0.41, R² = 0.16, f² = 0.20, large effect size) and growth mindset (β = 0.36, R² = 0.12, f² = 0.14, medium effect size) had significant effects on teacher grit.