Though there is no unified definition of subjective well-being that can be adopted by all institutions (Travia et al., 2020), in consensus, psychological studies on subjective well-being are mainly focused on two perspectives: hedonic and eudaimonic. The hedonic perspective focuses on personal happiness and good mood, which includes life satisfaction and positive effects (Diener et al., 2009). While the eudaimonic perspective focuses on outcomes, such as the purpose of life, personal growth, and meaning in life (Waterman, 1993, 2007). For instance, recent work has found that eudaimonic orientation of university students moderates the happiness benefits of prosocial activities like money donation (Lai et al., 2020); living in truth to one’s true self gives rise to eudaimonia, which certainly leads to subjective well-being (SWB) (Waterman, 2011; Cui et al., 2021). People achieve and maintain hedonic well-being through the eudaimonic process (Lent, 2004), as people are likely to achieve a happy and satisfaction in life by engaging in meaningful activities and working through their personal goals (Mao et al., 2016). Such engagement, according to the father of positive psychology (Seligman, 2002), is actually the optimal experience called “flow.”

Flow describes the affective and cognitive state when people are immersed in the current activities at hand (i.e., academic study), they are completely involved, entirely concentrated, and fully focused with capacity in information processing at the speed of about 110 bytes per second (i.e., vs. 40 b/s when doing academic work while talking), having a sense of time distortion that either time passes quickly or stops (Csikszentmihalyi and Csikszentmihalyi, 1992; Nakamura and Csikszentmihalyi, 2002). For many years, the flow has become an increasingly important asset in a variety of fields (Mao et al., 2016). For example, the flow was found to significantly reduce anxiety of university students through increased academic self-efficacy and promoted self-esteem in higher education (Mao et al., 2020). Flow related to work has been proven to be able to promote positive organizational outcomes, such as work performance, organizational citizenship behavior, and subjective well-being (Bakker, 2008; Fullagar and Kelloway, 2009; Soriano et al., 2021). Interestingly, research has also demonstrated that flow experience is a highly unstable “optimal experience” in all walks of human activities (Csikszentmihalyi, 1975; Mao et al., 2016), as the individuals keep stretching their skills in coping with fluctuated challenges from the outside world on their way of personal growth. Flow is varied greatly at the within-person level, indicating that it changes at an individual’s different stages of life span (i.e., from pre-school to college years) and with its dynamic nature being stressed (Fullagar and Kelloway, 2009; Debus et al., 2014). In this regard, Ceja and Navarro (2009, 2011, 2012) have found that the flow state within human beings tends to follow a disordered pattern, that is, flow experience shows a constant fluctuation during each day and does not go steady over time. What experiencing flow means and what effects flow potentially for university students in the world have been verified by many researchers (Chang et al., 2017; Soulliard et al., 2019). Although the positive impact of flow on subjective well-being is well established, and flow is found to be a universal experience especially in work and study (Bonaiuto et al., 2016; Mao et al., 2016), the precise underlying mechanism of the flow effect on subjective well-being, however, remains unclear. The accumulating works on flow demonstrate that flow moderates the link between quarantine length and well-being in Chinese participants (Sweeny et al., 2020) and many research records on subjective well-being of college students in the time of COVID-19 (Cui et al., 2021; Genç and Arslan, 2021), we accordingly hypothesize that:

Self-efficacy refers to the confidence of an individual to take actions toward his or her goals (Bandura, 1977). When university students do not have the confidence to pursue their desired academic goals, they may lose hope or find no meaning in life, because the study is one of the major daily activities they pursue. Self-efficacy encourages increasing personal skills and abilities via finding ways to manage stress and to control challenging situations on one’s own (Muris, 2002), therefore, it can enhance mental health and one’s well-being (Kim, 2003). When one’s skill competence meets the challenge of outside or external social environment, s/he will experience flow from the activity at hand (Mao et al., 2016). Within the context of higher education, academic self-efficacy is considered to be an important determinant of academic performance of university students and their personal development (Honicke and Broadbent, 2016; Talsma et al., 2018), because academic self-efficacy is positively related to optimal enjoyment of flow brought about by study/learning activities (Joo et al., 2015; Mesurado et al., 2016). For example, when students control their learning situation and completely immerse themselves in their own academic activities, they feel a strong sense of time distortion, and therefore flow tends to occur more frequently (Fullagar et al., 2013). With coping as a mediator between personality and stress outcomes (Knoll et al., 2005), self-efficacy activates the response to be assertive when facing a challenging situation, resulting in the encouragement of one’s well-being (Eskin, 2003). When university students feel that they have the confidence to deal with stressful and uncomfortable situations, they manage the situations and move on (Bandura, 1997). A study about academic adjustment and life satisfaction in Portuguese college students via a longitudinal design has found that academic self-efficacy and environmental support are predictive of goal progress and academic adjustment, of which academic adjustment is predictive for global life satisfaction of students, and academic self-efficacy and positive affect are found to be reciprocally related to one another (Lent et al., 2009). Academic self-efficacy is also an important factor in predicting academic resilience of Latino college students (Cavazos et al., 2010). Taken together, given that an individual with higher academic self-efficacy can better cope with difficult situations like trying to alleviate anxiety (Mao et al., 2020), it is much more likely that academic self-efficacy can be positively associated with subjective well-being.

Theoretically, the higher the level of one’s flow (or optimal experience), it is more likely that one can set a clear goal, the steadier control one has over his task, the stronger immersion and pleasure one can feel (Csikszentmihalyi, 1975, 1990). As an important part of the flow, one’s independence or freedom when engaging in activities is repeatedly found to be able to increase positive affect (Saavedra and Kwun, 2000) and motivation (Fried and Ferris, 1987), which in turn promotes academic self-efficacy. As academic self-efficacy is reflected widely in the daily life of university students especially during the challenging time of COVID-19 (Alemany-Arrebola et al., 2020), an individual who has more frequent flow experience may feel increased confidence and a loss of self-awareness during academic activities, in this sense, we predict that:

Self-esteem refers to the extent to which individuals like, value, accept, and respect themselves at a general or global level (Rosenberg, 2015). It is defined as a positive attitude one holds toward him- or herself and is a self-assessment of one’s own value (Jordan et al., 2020). Self-esteem, as a sense of self-worth, is closely associated with subjective well-being and a number of other adaptive outcomes (Seligman and Csikszentmihalyi, 2000; Diener and Diener, 2009; Diener et al., 2015). Prior research has indicated that an individual with higher self-esteem enjoys more positive feelings, in comparison with those who hold lower self-esteem (Martín-Albo, 2007). Numerous research records have dated that one’s self-esteem can strongly predict positive mental health, such as life satisfaction and well-being (Mann et al., 2004; Lehtinen et al., 2006; Wang and Fowler, 2019). Besides, self-esteem is also important for maintaining objective physical health (Stinson et al., 2008). Self-esteem begins to shape in childhood and continues to change and develop throughout the lifespan (Rubin et al., 2002). Studies from a large national sample of young adolescents in the United States (aged between 12 and 16 years) demonstrate that lower level of self-esteem is associated with a number of modifiable risk factors, such as a lower level of team sports participation and poor school performance, and vice versa (McClure et al., 2010). In contrast, a higher level of self-esteem helps to cope with stressors by accumulating available coping resources that support mental health (Crabtree and Rutland, 2001; Taylor and Stanton, 2007).

As for the relationship between one’s self-esteem and his or her flow experience, contradictory findings coexist in research records. For instance, some have argued that self-esteem is the antecedent of flow: greater self-esteem encourages an increased level of flow when engaging in digital games (Choi and Kim, 2013). Whereas others have suggested that it is the lower level of self-esteem (instead of higher self-esteem) that allows people to have a higher level of flow experiences when surfing on the Internet, playing computer games, and immersing in mobile phones (Khang et al., 2013). Understanding such a paradox needs to consider the nature of game experiences based on flow theory (Csikszentmihalyi, 1975, 1990): if the activity itself is to be carried out with a meaningful goal toward personal growth, self-actualization and harmonious integration of the individual are achieved in the social context, then higher self-esteem brings about higher flow, which is consistent with the eudemonic value of well-being. If, however, playing digital games is to avoid negative feedback and unpleasant situations from others and the outside world, then, the activity itself is reckoned as an anti-social activity, thus engaging in such anti-social activity is to express low self-esteem, though one may still have a sense of optimal enjoyment with hedonic pleasure (Delle Fave, 2013). To date, the above concrete evidence supporting a link between self-esteem and flow is generated from the digital activities over the social media, however, whether such a paradox relationship applies to academic activities within a university context, needs further investigation, though the flow concept has been introduced in designing educational software/games.

The previous section has indicated that self-esteem is the antecedent of flow (Khang et al., 2013). However, it can be the other way around, because the flow has important potential in promoting personal development, such as developing one’s character strength (Wells, 1988), that being said, experiencing more frequent flow in everyday life brings about higher self-esteem (Asakawa, 2010). Recent findings have indicated that because of the impact of COVID-19 lockdown, youngsters who are exposed to more social network sites show lower self-esteem (Vall-Roqué et al., 2021). Flow experienced more and frequently during academic activities at university is associated with increased self-esteem (Mao et al., 2020), which, in turn, mediates the association between positive psychological functioning (i.e., optimism and social support) and subjective well-being (Kong et al., 2013; Duy and Yıldız, 2019). However, little is known about whether the challenge of COVID-19 that brought about in-campus lifestyle changes of university students, such as a sudden shift to online learning due to the pandemic quarantine (Anderson et al., 2020; Pagliaro et al., 2021), may pose an impact on their experience of both positive functioning of flow and self-esteem, which may in turn affect subjective well-being. We therefore assume that:

Previous sections have provided supports on the positive association of flow with academic self-efficacy and of flow with self-esteem. Prior empirical work has also supported a positive association between self-esteem and self-efficacy beliefs (e.g., D’Amico and Cardaci, 2003). However, there are inconsistent findings about the association between self-esteem and self-efficacy, yet no records can be traced in the literature to address their joint contribution to well-being. Specifically, there are inconsistent results about the association between self-esteem and self-efficacy. In fact, self-esteem concerns one’s judgment of his or her self-worth, whereas perceived academic self-efficacy is concerned with the judgment of personal competencies and capabilities in academic study. For example, one may judge himself inefficacious in a certain activity without investing his self-worth in that activity, thus, he will not suffer from any loss of self-esteem for being inefficacious (Bandura, 1997). Considering the academic activities in higher educational context at university, students may perceive themselves with increased self-esteem if they achieve academic goals by stretching capabilities with increased skills (Delle Fave and Bassi, 2016), and such positive self-esteem may in turn induce them to perceive their academic capabilities as more positive (D’Amico and Cardaci, 2003). However, although cognitive ability and conscientiousness have been found to predict performance, less is known about whether and when certain mediating variables help explain these relationships (Chen et al., 2001). There are no studies, to our best knowledge and insofar, on the roles of academic self-efficacy and self-esteem played in uncovering flow and subjective well-being relationship. Nevertheless, a recent work may provide an implicit clue that flow can predict academic self-efficacy, which in turn predicts self-esteem, and subsequently predicts reduced anxiety (Mao et al., 2020).

Other indirect evidence can also support such a sequential mediation. For instance, Stupnisky et al. (2013) proposed that if students can maintain both a higher sense of control and self-esteem at university, they will experience more positive well-being and will perform better academically afterward. Odaci (2013) found a positive (but not significant) correlation between the belief of postgraduate students in academic self-efficacy in research and their self-esteem. Data from the German Aging Survey revealed that programs aiming at increasing optimism, self-esteem, and self-efficacy might be helpful to maintain subjective well-being (Hajek and König, 2019). Taken together, all these evidences lead us to propose our assumption that:

The proposed model is therefore indicated below in Figure 1.

Subjective well-being was assessed by the 5-item Satisfaction With Life Scale (SWLS; Diener et al., 1985), which has been widely used for testing the core component and the cognitive aspect of subjective well-being (e.g., Zanon et al., 2014; Jovanović and Brdar, 2018; Chaves et al., 2021; Kong et al., 2021). Since the first item is similar to the second item under Chinese context in our pilot test, while the factor loading for the fourth item was pretty low, these two items were excluded in further analysis. The sample item was “The conditions of my academic life in this university are excellent.” We averaged responses to create an overall individual subjective well-being index, in which a higher score on this measure indicated greater subjective well-being (Arrindell et al., 1999). Cronbach’s alpha coefficient for the present sample was 0.800.

The 7-item Swedish Flow Proneness Questionnaire (SFPQ; Ullén et al., 2012) was designed as a self-report measure of how frequently the participant has flow experiences pertained to each of the three typical divisions of activities (work, maintenance, and leisure time). SFPQ has been used in numerous studies (Mosing et al., 2012; De Manzano et al., 2013; Gyurkovics et al., 2016; Niksirat et al., 2019; Liu and Csikszentmihalyi, 2020). Three items were deleted because of the low factor loadings, therefore, yielded four items representing typical flow characteristics for a single dimension that was pertained to academic study in a university context for the frequency of flow. The sample item was “I often feel that my skills/abilities completely match what I do no matter how difficult it is.” Scores of each participant including reversed items were coded, higher scores represented higher frequency of flow experience. The reliability of the scale was good (Cronbach’s α = 0.813) in the present sample.

We adopted an 8-item Academic Self-Efficacy Scale (Stagg et al., 2018) that has been used previously by Mao et al. (2020) in a Chinese adolescent sample with a reported good internal consistency (Cronbach’s α = 0.837). However, in the present work, we deleted three items after the pilot test due to its lower factor loadings conducted from confirmatory factor analysis. Thus, the responses of each participant were registered based on five finalized items of academic self-efficacy scale, which covered the three principal facets: learning efficiency, examination, and learning processes (e.g., I am confident in my ability to manage my time effectively in study). The higher the score, the greater the level of the academic self-efficacy. For the present sample, the reliability was good (Cronbach’s α = 0.879).

Self-esteem was measured based on the well-recognized 10-item Rosenberg Self-Esteem Scale (RSES; Rosenberg, 1965) on which the participants indicated the extent to which they felt themselves to possess good qualities, to accept their own characteristics, and to have achieved personal success or experienced failure. This scale has been widely used (e.g., Mao et al., 2020). Because of the low confirmatory analysis (CFA) factor loadings yielded on three items, data were analyzed based on seven items. The sample item was “I take a positive attitude toward myself.” The greater the accumulated score, the greater the self-esteem. Cronbach’s alpha coefficient was 0.911.