There is extensive research, which has focused on the importance of the social context. Vygotsky (1978) sociocultural theory of cognition, for example, places emphasis on the social processes – that a person, for example, would internalize his learning and understanding about Psychology 101 from peers and capable others. Bronfenbrenner (1989) bioecological systems theory, similar to Vygotsky (1978) theory, discusses the importance of situated sociocultural influences. Roorda et al. (2011) meta-analysis of engagement, likewise, details the impact of the social environment. According to the authors, a social environment is perceived as being positive when opportunities arise for proactive social relationships (e.g., teacher-student social relationship: Hawkins et al., 2010; Allen et al., 2013; Gallagher, 2013). Why is this the case? A student is more inclined to have a favorable view when a teacher or an educator is able to provide social support and emotional security. This premise, in part, reflects the nature of attachment theory (Ainsworth et al., 1978; Ainsworth, 1979; Stevenson-Hinde and Verschueren, 2002), which emphasizes a student’s need to seek emotional and social bonding.

From the perspective of schooling and academia, it is poignant for students to perceive positive social experiences. Perceived enriched social experiences, such as a positive teacher-student relationship (TSR) (Roorda et al., 2011) would assist to facilitate school, or academic, adjustment, resulting in improved academic and non-academic performances. In the context of the present study, we define a student’s perceived social experiences, ξ₁, as being a composite, or an amalgamation, of the following: the student’s perception of the social milieu itself (X₁), the student’s ability to socially relate to others at school (X₂), and the student’s perception that there are pathways and means for academic growth (X₃). Positive social experience at school (e.g., supporting friendship with a peer) is effective (Roorda et al., 2011; Whannell and Allen, 2011) as it helps students to adjust and to improve their academic performance outcomes. For example, in line with Vygotsky (1978, 1981) sociocultural theory of cognition, a student may actively interact and seek academic support from a teacher and/or from a peer who is more capable in the subject matter.

Aside from proactive social relationships at school, opportunities, pathways, and/or means for personal growth in academic learning and/or personal well-being (Li et al., 2008; Soutter, 2011; Waters et al., 2017) are also encouraged. For example, in a recent study, which involved university students, Phan et al. (2019c) found that relating to others accounted for an improvement in positive emotional well-being. In a similar study that consisted of secondary school students in Taiwan, Phan and Ngu (2020a) noted that proactive social relationship between teachers and students predicted personal and positive emotional well-being. Peer-peer social relationship, in contrast, predicted personal well-being and personal striving. From our point of view, taking into consideration this line of evidence, positive perceptions of social experiences (e.g., a friend’s willingness to assist) could help to instill motivation for academic learning, whereas negative perceptions (e.g., a student’s perceived sense that she is not being accepted by other students) would demotivate and serve to weaken one’s resolve to engage.

There is also existing research development, which contends the potent influences of different aspects of perceived social experiences on different sub-psychological processes. For example, in a correlational study that focused on academic engagement, Liem and Martin (2011) found that peer relationships (e.g., opposite-sex relationship at school) positively predicted general self-esteem. In a study that consisted of children with dyslexia, Shehu et al. (2015) found a positive association between social relationships and self-esteem. In one of the earlier studies in the late 1990s, Wentzel (1998, p. 202) using multiple regression analysis reported some interesting patterns: peer support was a positive predictor of prosocial goal pursuit, teacher support was a positive predictor of both types of interest and of social responsibility goal pursuit, and parent support was a positive predictor of school-related interest and goal orientations. More recently, as indicated in Table 1, Phan and his colleagues observed the positive effects of perceived social experiences – for example: the effect of relating to others on personal striving (Phan et al., 2019d), the effect of the social milieu on self-efficacy (Phan et al., 2019c), and the effect of peer-peer social relationship on personal striving (Phan and Ngu, 2020a).

Analysis of existing research suggests then that there is clear and consistent evidence, which affirms the central role of perceived social experiences as an antecedent of different types of adaptive outcomes (Wentzel, 1998; Roseth et al., 2008; Umberson and Montez, 2010; Raufelder et al., 2013; Shehu et al., 2015). This proposition is poignant as it emphasizes the explanatory nature of perceived social experiences, which have both positive or negative connotations – for example: I feel that this school is not supporting me at all and that, importantly, people here are not too “welcoming” (i.e., this is a negative perception of social experience). A positive social experience, in this analysis, is more potent as it would help students cope with their academic learning, seek emotional and/or social support, and develop enriched well-being experiences (e.g., positive emotional well-being). By the same token, of course, there is credence to contend that a person’s perceived social experiences are intricately linked to his/her sub-psychological processes for effective learning (e.g., a person’s self-belief). In the context of the present study, we postulate that perceived positive social experiences in educational contexts (e.g., the perception that there are opportunities and pathways for personal growth) would positively influence different types of psychological and educational variables.

One notable outcome of perceived positive social experiences is the enactment of internal psychological processes, ξ₂, which espouse three comparable sub-processes: personal resolve (X₁), effective functioning (X₂), and academic striving (X₃). This proposed psychological processes concept, from our conceptualization, coincides with the paradigm of positive psychology (Seligman and Csíkszentmihályi, 2000; Seligman et al., 2009; Seligman, 2010), which focuses on the notion of “positivity and proactivity” of human agency. One notable distinction of positive psychology, in particular, relates to the understanding that there are sub-psychological processes (e.g., a state of resilience) that may facilitate a person’s state of functioning. For example, within the context of academic learning, there is evidence to indicate the positive effects of positive psychological sub-processes such as a state of buoyancy (Martin et al., 2013; Collie et al., 2015).

Our conceptualization considers three comparable constructs, which could potentially optimize a person’s state of cognitive functioning: personal resolve, effective functioning, and academic striving. An analysis of the information detailed in Table 1 suggests that, in general, personal resolve, effective functioning, and academic striving are similar in terms of their nature and characteristics. One notable distinction in similarity lies in the facilitation and enhancement of a person’s positive state of functioning, such as his/her achievement of optimal best in an academic subject matter (Fraillon, 2004; Martin, 2006; Phan et al., 2016c). For example, research development has noted consistent evidence, which shows the positive effect of personal resolve on the achievement of optimal best (Phan et al., 2019d; Phan and Ngu, 2021a). This line of empirical validation contends that a state of decisiveness and unwavered focus without any uncertainty is likely to associate with and/or instill a perceived of confidence, resulting in a person’s self-determination and subsequent performance outcome. Indecisiveness, in contrast, is more aligned with a person’s sub-optimal experiences, reflecting in his/her underachievement in a subject matter.

Examining the nature of sub-psychological processes of personal resolve, effective functioning, and academic striving, which may operate in tandem with each other, is insightful as this would provide clarity into their combined and/or individual predictive effects on different types of adaptive outcomes. Specifically, as shown in Figure 1, it is plausible to consider two contrasting explanatory and predictive effects: the predictive effect of the latent representation of psychological processes on an adaptive outcome [i.e., ξ₂ → O, where ξ₂ = psychological processes, O = adaptive outcome (e.g., a state of motivation), → = predictive effect] versus the predictive effects of the three measured indicators on the same adaptive outcome [e.g., X₁ → O, where X₁ = personal resolve, O = adaptive outcome (e.g., a state of motivation), → = predictive effect]. This consideration, we contend, reflects methodological innovation, providing a basis for us to gauge into comparative predictive effects of psychological processes. By the same token, structurally, it is also possible for us to test and identify different types of information, which could assist to account for the formulation of personal resolve, effective functioning. For example, in a recent study that involved university students, Phan et al. (2018) found that a student’s current level of knowledge and understanding of a subject matter, termed as “realistic best practice,” would positively influence a state of effective functioning and his/her personal resolve. Interestingly, with secondary school students, Phan and Ngu (2020a) noted that the formulation of academic striving, defined as a person’s effortful attempt to seek out a realistic and/or an ambitious endeavor for accomplishment, is shaped by different psychosocial factors – academic self-efficacy, global self-esteem, peer-peer social relationship, personal well-being, and belief of optimal best.

Another potential outcome of perceived positive social experiences, and also that of the enactment of psychological processes, is a perceived state of motivation (Franken, 2007). The study of motivation within the field of Educational Psychology is extensive, with different theoretical perspectives offered and research inquired undertaken. Maslow (1968, 1970) humanistic theory of motivation, for example, focuses on a person’s motive to achieve self-fulfillment of different needs (e.g., a need to self-fulfill and experience of love), whereas the cognitive perspective of motivation focuses on theoretical understanding of attribution (e.g., attribution of success that reflects one’s own personal ability) (Weiner, 1972, 1986), self-determination (Deci and Ryan, 1991, 2008), and value-expectancy beliefs (Eccles et al., 1983; Wigfield and Eccles, 2000). Researchers and educators have over the years proposed different inquiries for research development, focusing on antecedents and consequences of a state of motivation and, likewise, a state of demotivation. Research undertakings into the complex nature of motivation (Sansone and Harackiewicz, 2000; Pintrich and Schunk, 2002; Franken, 2007; Schunk et al., 2008) are effective and beneficial, providing theoretical insights and fruitful information for the purpose of practicality – for example, what is the most effective strategy to instill extrinsic motivation?

Our conceptualization of motivation, ξ₃, as a distinct construct, depicts three comparable and interrelated constructs, interest in learning (X₁), motivation toward learning (X₂), and self-efficacy for learning (X₃). The stipulation of X₁, X₂, and X₃, in this case, enables us to provide a definition of motivation, which we consider it as being “a multifaceted and positive entity that encompasses one’s personal interest and a heightened state of self-belief to accomplish enriched learning experiences for different types of intrinsic reasons”. This consideration, for us conceptually, contends that the nature of motivation is intrinsic, wherein a person’s intent and purposive act is to learn and accomplish for intrinsic reasons, such as enjoyment, fulfillment of an inner desire to achieve mastery, and to improve personal self-belief of competence. From our proposition then, a person’s low level of interest and motivation, as well as his/her inefficacy for academic learning (i.e., a weakened state of self-efficacy) would undermine, weaken, and result in underperformance in a subject matter. Self-efficacy and high levels of interest and motivation, in contrast, would improve and/or facilitate a student’s learning experience.

There is a plethora of research that has yielded clear and consistent evidence, highlighting the potent role of motivation as a facilitator and predictor of different types of adaptive outcomes (Pintrich and Schunk, 2002; Franken, 2007; Schunk et al., 2008). What is innovative and relatively unique about our conceptualization, however, relates to the stipulation of contrasting predictive effects of both latent factor (i.e., ξ₃ → O, where ξ₃ = motivation, O = adaptive outcome, → = predictive effect) and measured indicators of motivation (e.g., X₁ → O, where X₁ = interest in learning, O = adaptive outcome, → = predictive effect) on different types of adaptive outcomes. For example, evidence of a positive effect of self-efficacy on achievement of an adaptive outcome would provide theoretical insights into the development of a teaching strategy and/or an educational program that could enhance and foster motivation, resulting improved academic performances and enriched learning experiences.

Our conceptualization also allows us to examine different types of antecedents that could account for the formulation of motivation, in general. What causes a student to feel motivated to engage in a particular course of action, academically and/or non-academically? In relation to self-efficacy (Bandura, 1997), for example, a number of correlational studies have attested to the effect of enactive learning experience such as successful accomplishments (Lent et al., 1991; Pajares et al., 2007; Liem et al., 2008; Usher and Pajares, 2008). In a similar vein, existing research has found numerous factors that could influence a person’s interest for learning and proactive engagement (Senko and Miles, 2008; Upadyaya et al., 2011; Walkington, 2013). We contend, as shown in Figure 1, that the two latent variables of perceived social experiences, ξ₁, and psychological processes, ξ₂, and their respective measured indicators could influence motivation and its respective measured indicators.

Seven hundred and thirty-nine undergraduate students (N = 442 women, 297 men) from three private universities in Taiwan participated in this study. The dataset, collected in 2019, forms part of our larger research project, involving secondary and university students from Australia, Malaysia, and Taiwan. The study reported in this manuscript was approved by the University of New England Research Ethics Committee. Because of the fact that the participants were all adults (i.e., over the age of 18), we decided to use a simpler approach for recruitment, which the university approved – that is, to verbally seek permission at the onset and anyone who did not want to take part to inform us. This method of verbally seeking participatory consent, used by us on previous research undertakings, is more convenient logistically.

Our sampling was convenient as it was logistically difficult to seek permission from students in other universities and colleges to take part in the present research study. Aside from this difficulty, limited resources also deterred us from attempting to expand on our data collection. The participants voluntarily took part in the study, knowing that there were no incentives. We administered the questionnaires in class, using the traditional paper-format. Overall, answering the Likert-scale measures took approximately 35–40 min to complete. The front page of the questionnaires contained demographic information – for example, which university a student attended, and what degree program he/she was studying. Overall, all the participants were full-time students and came from 40 different departments across the three universities.

The medium of formal instruction at school and in university in Taiwan is Chinese Mandarin. In 2016, we translated the questionnaires from English to Chinese Mandarin using a three-step methodological procedure: (i) Step 1: to translate the questionnaires from English to Chinese Mandarin (E₁ → CM₁), (ii) Step 2: to translate the translated questionnaires from Chinese Mandarin to English (CM₁ → E₂), and (iii) Step 3: compare the original English version of the questionnaires (i.e., E₁) with the translated version (i.e., E₂) (e.g., see Phan et al., 2019d for an in-depth description:). We have since then used the questionnaires with different cohorts, and results published (Phan et al., 2019c, d, 2020e) show sound psychometric properties (e.g., factorial structures).

Overall, from the preceding sections, there are 12 variables, which involved the use of 12 corresponding subscales, measured on a Likert-scale with five ratings – for example: 1 (Complete Not True), 2 (Not True), 3 (Neutral), 4 (True), and 5 (Completely True).

A four-factorial structure is interesting and may, in this case, provide evidence of factor loadings and the potential interrelations between the four latent factors. We used the statistical software package MPlus 8.5 (Muthén and Muthén, 1998–2012) to assist with our analyses of a four-factor model. Likewise, SPSS 25 was used for descriptive statistics (e.g., data screening). Per guidance (Bollen, 1989; Kline, 2011), we performed an initial data screening test to ensure multivariate normality and the justification of using maximum likelihood (ML) estimates to test our hypothesized model [e.g., kurtosis values ranging from −0.06 (Academic Striving) to 1.02 (Optimal Best Practice), Std error = 0.18; skewness values ranging from −0.61 (Current Best Practice) to 0.06 (Academic Striving), Std error = 0.09]. From previous research undertakings (e.g., Green et al., 2012; Liem et al., 2012; Phan et al., 2019d), we considered the following indices to assist in the gauging of appropriate model fits: the Comparative Fit Index (CFI) (i.e., CFI value > 0.95), the Tucker Lewis Index (TLI) (i.e., TLI value > 0.95) the Root Mean Square Error of Approximation (RMSEA) (i.e., RMSEA value < 0.07), the χ² test statistic, and an evaluation of parameter estimates were used in the present research to assess model fit (Liem and Martin, 2011).

Correlations between mean scale scores are shown in Table 3. Our CFA undertaking for the four-factor model showed an appropriate model fit, as indicated by the following goodness-of-fit index values: CFI = 0.94, TLI = 0.92, RMSEA = 0.08, p < 0.001 (Lo90 = 0.071, Hi90 = 0.090), and χ²/df = 5.77. From Table 4, the factor loadings ranged from 0.52 to 0.77 for perceived social experiences, 0.70 to 0.75 for psychological processes, 0.63 to 0.70 for motivation, and 0.53 to 0.86 for adaptive outcomes, p < 0.001. associations also existed between the four latent factors: r = 0.95, p < 0.001 for psychological processes and perceived social experiences, r = 0.92, p < 0.001 for motivation and perceived social experiences, r = 0.89, p < 0.001 for motivation and psychological processes, r = 0.77, p < 0.001 for adaptive outcomes and perceived social experiences, r = 0.82, p < 0.001 for adaptive outcomes and psychological processes, and r = 0.77, p < 0.001 for adaptive outcomes and motivation.

The results established from the factorial structure analysis substantiated our SEM undertakings, which consisted of a baseline model. Overall, from the hypothesized a priori model shown in Figure 1, there are four latent factors (i.e., perceived social experiences, psychological processes, motivation, and adaptive outcomes) and 12 measured indicators. The base-line model, denoted as Model M₀, consisted of six structural paths for statistical testing: the structural path from perceived social experiences to psychological processes, the structural path from psychological processes to motivation, the structural path from perceived social experiences to motivation, the structural path from motivation to adaptive outcomes, the structural path from psychological processes to adaptive outcomes, and the structural path from perceived social experiences to adaptive outcomes. The results of this base-line model, using covariance matrices as correlation matrix analysis is known to entail potential problems (e.g., producing incorrect goodness-of-fit index values (Byrne, 1998; Jöreskog and Sörbom, 2001) are relatively modest in terms of fit – for example: χ²/df = 5.77, CFI = 0.94, TLI = 0.92, RMSEA = 0.080. From a researcher’s point of view, existing theorizations and research development (e.g., Schumacker and Lomax, 2004; Kline, 2011; Byrne, 2012) would recommend the finalization and discussion of the results of Model M₀. Having said this, we are interested to test different iterations of Model M₀, via modification fit indices.

We are interested, in this case, to explore the relationships between both latent factors and measured indicators. Statistical software packages such as MPlus 8.5 (Muthén and Muthén, 1998–2012) provide modification fit indices to assist a researcher in his/her quest to achieve an optimal fit for an a priori model. Having said this, researchers have cautioned the use of modification fit indices, contending that respecification of an a priori model depends on rationale and/or strong theoretical and/or empirical grounding (Byrne, 1998; Kline, 2011). Our testing and comparison of different a priori and a posteriori models is interesting, as it enables us to establish theoretical understanding into L₁ (e.g., perceived social experiences → psychological processes) and L₂ (e.g., relating to others → psychological processes) relationships. Table 5 shows the testing of seven a posteriori models. Goodness-of-fit index values are also provided to help finalize and determine the optimal fit model for discussion.

The respecification of a base-line model and, subsequently, a comparison of two competing models (e.g., Model M₀ versus Model M₁) require the use of the Δχ² test, as well as an inspection of the goodness-of-fit index values. The principle of parsimony, in this case, advocates for the acceptance of a less restricted model. Overall, as shown in Table 5, the results (e.g., the use of the Δχ² test as well as various goodness-of-fit index values) indicate a progression in improvement of model fit from Model M₀ to Model M₇. From the modification fit index values, we respecified Model M₀ to include seven additional structural paths, which we freed one at a time:

As shown in Table 5, the Δχ² tests showed an improvement in model fit, progressively, from Model M₀ to Model M₇. Aside from the Δχ² tests, an inspection of the goodness-of-fit index values also supported the testing of different iterations of Model M₀, and the acceptance of a less restrictive model. Overall, from analyses of different a posteriori models, we accept the results of Model M₇ for discussion (e.g., χ²/df = 2.56, CFI = 0.98, TLI = 0.97, RMSEA = 0.046). The solution of Model M₇ is depicted in Figure 3 and, indeed, shows some interesting evidence in terms of the associations between the four latent factors: (i) that perceived social experiences positively influenced psychological processes (β = 0.81, p < 0.001), motivation (β = 0.61, p < 0.001), and adaptive outcomes (β = 0.36, p < 0.01), (ii) the positive effect of psychological processes on motivation (β = 0.31, p < 0.01), and (iii) that both psychological processes and motivation did not influence adaptive outcomes. Furthermore, of the four latent factors, evidence established indicated the potentiality for psychological processes to as a central mediator between perceived social experiences and motivation: the total effect from perceived social experiences on motivation is 0.87, p < 0.001 in which the decomposition is follows: direct effect of perceived social experiences on motivation is 0.61, p < 0.001, whereas the direct effect, mediated via psychological processes is 0.26, p < 0.01.

An inspection of Figure 3 indicates a number of pathways, or trajectories, that are statistically significant, accounting R² = 0.60 for the concept of adaptive outcomes, which is made up of a student’s current best practice, optimal best practice, and academic experience. An important question that we often ask, as educators, is how we would encourage and foster students’ educational experiences in school or university contexts. One possible approach to this is to consider pathways, opportunities, and/or means that could assist to improve, strengthen, and/or facilitate different students’ learning and non-learning experiences. Interestingly, as summarized below, there are a number of statistically significant pathways:

^∗Outcome is a latent concept – that is, a measured indicator → latent concept, where → = predictive effect.

From the above, there are five notable findings that are of significance: (i) the role of perceived social experiences as an important antecedent of psychological processes, motivation, and adaptive outcomes, (ii) the positive effect of psychological processes on motivation, (iii) the central role of psychological processes as a predictor as well as a mediator (Baron and Kenny, 1986; Trafimow, 2015), (iv) explanatory accounts and predictive effects of both latent factors (e.g., perceived social experiences → psychological processes) and measured indicators (e.g., relating to others → self-efficacy) level, and (v) empirical validation of a four-factor structure, depicting factor loadings (e.g., psychological processes consisted of personal resolve, effective functioning, and academic striving) and interrelations between the four latent factors.

What we can conclude from the established findings, as detailed in the preceding sections? There are three overarching aspects for consideration, which may advance our theoretical understanding into the positivity and proactivity of human agency. One achievement of human agency in educational contexts, in this case, relates to a student’s autonomy and motivation to achieve optimal learning experiences (Phan et al., 2019b). The conclusion derived from SEM analyses contends the following:

The present study, aside from empirical and theoretical contributions, has also provided enriching insights for the purpose of daily relevance and practical application. This acknowledgment recognizes the importance of the nexus between research and quality teaching and learning experiences. Our focus of inquiry, differing from a deficit approach (e.g., a focus on school failure or task disengagement) (Liem et al., 2008; Henry et al., 2012), is significant for its emphasis on the proactivity and positivity of human behavior. From the perspective of schooling, in general, it is a valuable feat to focus on the promotion and encouragement of enriched student experiences, academically and/or non-academically – for example, a secondary school student may attend and enjoy school for various reasons (e.g., to partake in extracurricular activities), or a university student who enjoys her learning and seeks to engage in mastery. There are a number of educational implications that are of value for us to consider. For the purpose of conciseness, we have included Table 6, which surmises our construction of different propositions and recommendations for consideration.

An analysis of the propositions and recommendations shown in Table 6 indicates one commonly theme – namely, to consider the use of effective pedagogies, programs, institutional policies, etc., that could promote the development of psychosocial (e.g., positive social relationship at school) and/or psychological experiences (e.g., a heightened state of personal resolve), which in turn would improve engagement of different types of adaptive outcomes. Moreover, Table 6 is significant for its depiction of practical emphasis, which may differ in nature and varieties – for example, the use of verbal discourse (e.g., encouraging feedback) as opposed to exposing students to different types of metacognitive strategies. Aside from this description, what else can we consider for practicality? Recently, for example, our research inquiries have delved into the topical theme of life and death education (Chen, 2009, 2012; Huang, 2014; Phan et al., 2020c) from the perspective of Taiwanese Education. One aspect of life and death education, reflecting both quality teaching and research development, is the incorporation of Buddhist teaching (Yeshe and Rinpoche, 1976; Master Sheng, 2010; Thanissaro, 2015), which may encompass Buddhist mindfulness and meditation practice (e.g., “walking meditation”). The underlying premise of Buddhist teaching, in accordance with the study of life and death education, is related to the notion of “spiritual cultivation” (Phan et al., 2020c, 2021c). Buddhist meditation, in this sense, may instill and cultivate a “purified mindset,” directing and assisting a person to feel more enlightened and spiritual. An interesting question then is whether and/or to what extent a spiritual mindset, via means of engagement in Buddhist meditation, say, could promote and/or predict a person’s psychological processes (e.g., a person’s heightened state of personal resolve).

The present study has provided some important methodological insights into the measurement and assessment of psychosocial and psychological concepts. One interesting aspect of our undertaking involved the use of Likert-scale measures, situated within a non-experimental context. Non-experimental data, especially cross-sectional are extremely restricted, limiting a researcher from making casual inference and/or personal inference (Rogosa, 1979; MacCallum and Austin, 2000). By all account, the use of self-reporting is inadequate in terms of providing robust and stringent information, which could capture and/or illustrate the underlying nature and intricacy of internal psychological processes. Self-reporting is descriptive and, in this case, provides evidence of perception, judgment, feeling, and belief, all of which may indicate inaccuracy. Pajares and his colleagues researching the topic of self-efficacy for academic learning in the 1990s alluded to an interesting phenomenon, which they termed as “miscalibration” (e.g., Pajares and Miller, 1994; Pajares and Kranzler, 1995; Pajares, 1996a). Miscalibration, in this case, considers two contrasting phenomena: a state of overconfidence versus a state of underconfidence. In the context of the present study then, it is plausible to contend that a university student may miscalibrate and report, for example, a high level of personal resolve or effective functioning. In a similar vein, a secondary school student may have some form of grievance toward his teacher, resulting in a biased report that indicates very little, if any, academic and/or social support at school (i.e., a biased indication of the indicator of pathways, means, and opportunities). On this basis, we argue for the use of alternative methodological designs, which could address the aforementioned shortcomings.

An interesting observation that we note, which has potential methodological relevance relates to in situ observations may complement the use of “meditative reflection” and subsequent self-reporting of personal feelings and experiences (Phan et al., 2021c). Meditative reflection and self-reporting, as a whole, is internal and arises from within a student (i.e., subjective in nature), whereas a teacher’s in situ observation of the student is external (i.e., objective in nature). This consideration of using multiple strategies of data collection, we contend, is more effective as this could provide a balanced, well-rounded viewpoint and/or testament (i.e., subjective versus objective) of a person’s behavior, psychological thoughts, emotions, experiences, etc. As such, contrasting methodological strategies may add credence to assist in the validation and elucidation of a process, relationship, etc.

We posit that the hypothesized relationships depicted in Figure 1 and/or the obtained solution shown in Figure 2 are somewhat complex and as such, as we previously mentioned, using cross-sectional and non-experimental data is inadequate. We encourage the use of longitudinal data as these could offer a more robust methodological approach in validating and/or substantiating long-term flow and personal growth. As an example, which may support our rationale, the underlying nature of academic striving (e.g., Phan and Ngu, 2020a, 2021a; Phan et al., 2020e) is complex and would, in this instance, require longitudinal data for accurate assessment and theoretical understanding. When a person aspires and strives to achieve an optimal state of functioning (Fraillon, 2004; Liem et al., 2012; Phan et al., 2016c), he/she would require time, effort, the availability of resources, etc. As such, in terms of appropriateness, we advocate for the use of multiple time points in data collection as this longitudinal consideration would offer a logical and more accurate capture of a student’s personal resolve, striving, etc., – for example, a student may indicate to us the following (T₁): “I want to strive and achieve Honors in Psychology by the time I graduate” (Date of indication, for example: 13th May, 2020). The question then is whether this mentioning of aspiration and personal striving would eventuate in 2 years’ time, when the student would have completed her academic studies (T₂).

Finally, one of our reviewers provided some interesting methodological insights, which we deem may play a critical role in future research development. The study of perceptions (e.g., a perception of positive social experience versus negative social experience), especially when it comes to diversity and sampling requires formal recognition and acknowledgment of potential selection bias and limitations. For example, as the reviewer highlights, our participants came from different universities and departments and consequently, from this variation and/or diversity, their individual differences were not well controlled. Thus, from this observation of “dissimilar” backgrounds, we urge readers to read and interpret our results with caution. In a similar vein, of course, our own observation recognizes that the sample was relatively modest in size, limiting us from undertaking additional and/or different types of statistical analysis – for example, a statistical undertaking, which involves a factorial variance/invariance analysis, identifying potential equivalency (Byrne, 1998) between different groupings (e.g., the extent to which the final solution depicted in Figure 3 is invariant across both men and women). A larger sample size, situated within hierarchical clusters or structures (e.g., universities × faculties × schools × departments), likewise, may provide opportunities for multilevel modeling (Little, 2000).