Abstract
The shift to online education during the COVID-19 pandemic has raised important questions about what factors predict student success in digital learning environments, particularly in demanding fields such as medical education. This study explores the extent to which selected psychological variables—self-efficacy, growth mindset, fixed mindset, extrinsic motivation, behavioral engagement, effort, and satisfaction—are statistically associated with academic performance, as measured by grade point average (GPA), among Saudi medical students enrolled in English-medium online courses. A total of 585 students from one public university completed a self-report questionnaire. Hierarchical multiple regression analyses were conducted to assess the predictive value of psychological variables beyond demographic factors. Results revealed that only behavioral engagement showed a statistically significant positive association with GPA after correcting for multiple comparisons. The findings suggest that, within this context, students’ perceived engagement with online learning activities is modestly but meaningfully related to their academic outcomes. The study contributes to the growing body of literature on psychological correlates of academic performance in non-Western online medical education and highlights engagement as a potential area for targeted instructional support.
Plain Language Summary
This study explored how various psychological factors predict medical students’ performance in online education in Saudi Arabia. It focused on factors like mindsets, self-efficacy, effort, engagement, motivation, and satisfaction. The research found that student engagement was the only factor linked to higher GPAs, especially in the early years. The results highlight the importance of engagement in online learning and suggest that more research is needed to improve student performance and success.
Introduction
Technological advancements have transformed the landscape of teaching and learning since online education, in its various forms, has experienced consistent growth on a global scale (Taylor et al., 2020). An increasing number of educational institutions are shifting from traditional, face-to-face education to online or blended learning approaches. Blended learning involves integrating online learning elements into the traditional classroom environment. As they embrace online education, institutions worldwide have started to adapt to rapid changes, sparking considerable interest among stakeholders such as educators, administrators, researchers, policymakers, and publishers (Palvia et al., 2018). Online and blended learning models have transcended geographical barriers, enabling access for individuals in remote areas and eliminating the need for travel.
Dziuban et al. (2015) summarized the evolution of American online education across four phases: The 1990s witnessed the emergence of internet-driven distance education; the period from 2000 to 2007 saw a surge in the adoption of learning management systems; from 2008 to 2012, the proliferation of massive open online courses (known as MOOCs) gained prominence; and later, enrollment in online higher education surpassed that in traditional education. Online education has become increasingly viable from technological, economic, and operational standpoints (Allen & Seaman, 2017). Universities are incentivized to offer online programs due to financial considerations, including cost savings associated with reduced infrastructure requirements for physical facilities such as classrooms, libraries, and dormitories (Palvia et al., 2018).
Predicting academic achievement has always been a focus of higher education research that aims to improve students’ self-awareness and enhance teaching (Binder et al., 2019). Academic achievement is defined as the degree to which a student has met their short- or long-term educational objectives, and it is assessed by either cumulative grade point average or ongoing assessment (Tadese et al., 2022). For medical students, academic achievement is demonstrated through mastering medical science knowledge and other related fields.
The significance of academic achievement has driven research to investigate factors that predict it. For instance, psychological attributes have been widely discussed in the literature to commonly correlate with academic achievement (Alyahyan & Düştegör, 2020). Such attributes include student interest, attitudes toward studying, anxiety and stress levels, self-control, motivation, and self-efficacy (Alyahyan, & Düştegör, 2020; Narainsamy & Van Der Westhuizen, 2013). Several studies have revealed a significant toll on the education of medical students which is attributed to the psychological burdens they are subjected to during severe circumstances (Abed et al., 2022; Cao et al., 2020; Consorti et al., 2021; Kavvadas et al., 2022).
Little research has addressed the psychological predictors of online medical students’ achievement in Saudi Arabia (Alghamdi et al., 2023). To fill this literature gap, this study aimed to investigate the correlation between a set of psychological predictors: specifically self-efficacy, fixed and growth mindset, engagement and effort, satisfaction, and extrinsic motivation, and medical students’ achievements in Saudi Arabia during the online classes. More specifically, our study aimed to address the following questions:
(1) What are the psychological predictors of medical students’ achievement in online learning?
(2) Are these predictors different for preclinical and clinical students?
Literature Review
Self-Efficacy
Bandura (1986) defined self-efficacy as a person’s beliefs in their own abilities to successfully accomplish a specific goal. Self-efficacy is connected to one’s perceptions and beliefs, not actual capabilities (Mills et al., 2007). The core of self-efficacy theory is that the initiation and continuity of behaviors and actions are primarily based on expectations and judgments of a person’s skills and abilities and the probability of coping with difficulties and demands (see Maddux, 1995). In addition, self-efficacy theory acknowledges that these factors intervene in psychological adjustment and unhealthy behaviors and also have a role in curative interventions addressing emotional and behavioral issues. Self-efficacy expectations are perceived across three dimensions: magnitude, strength, and generality. The quantity of steps of escalating complexity or risk that an individual thinks they can complete is known as their sense of self-efficacy. Strength of self-efficacy pertains to the determination of an individual’s certainty of their ability to carry out a specific behavior. Furthermore, strength of self-efficacy has been linked to endurance against anger, pain, and other impediments to performance. Generality of self-efficacy expectancies pertains to the extent to which experiences whether holding success or failure impact self-efficacy expectations through behaviors, or if changes in self-efficacy extend to comparable behaviors and situations (see Maddux, 1995, for various applications).
Self-efficacy is an important factor influencing student motivation and academic behaviors (Burgoon et al., 2012). Research shows that self-efficacy positively correlates with students’ academic performance in non-medical education research (Graham, 2011; Hsieh & Kang, 2010; Mills et al., 2006, 2007). Students with high self-efficacy tend to succeed in their studies and achieve the intended outcomes. Some studies have examined the relationship between self-efficacy and students’ achievement in medical education (Tiyuri et al., 2018; Zheng et al., 2020). To further examine the role of self-efficacy on medical students, Tiyuri et al. (2018) carried out a study on 320 Iranian postgraduate students and found that their self-efficacy was acceptable and significantly correlated with their academic performance. On the other hand, Zheng et al. (2020) investigated self-efficacy in a flipped classroom environment using structural equation modeling on 146 undergraduate medical students’ responses in a U.S. Midwestern university, and found that self-efficacy was a direct predictor of their academic achievement.
Fixed and Growth Mindset
A potential psychological factor that may play an essential role in medical students’ performance is implicit theories (i.e., mindsets) about one’s abilities (Dweck, 1999). A mindset is the extent to which one perceives qualities such as intelligence or personality as being fixed or changeable (Dweck & Molden, 2005). Mindsets can shape the different ways individuals think, feel, and act even if the situation is identical (Dweck & Molden, 2005). While a student with a growth mindset perceives ability as being malleable, a student with a fixed mindset considers ability as inborn. Therefore, growth mindset students consider setbacks as an essential part of learning, while fixed mindset students perceive any failure as a threat to their self-esteem (Lou & Noels, 2019; Yeager et al., 2016). Individuals may hold different mindsets in different domains such as intelligence, personality, academic performance, social relationships, and sports (Dweck, 1999). Previous research has also shown that mindsets are dynamic and are modifiable (Yeager et al., 2016). For instance, after experiencing success or failure, a person may affirm a fixed mindset upon receiving praise or criticism of intelligence or a God-given ability. In comparison, if the person receives specific and genuine feedback that is directed toward effort, strategies, and skills, they may develop a growth mindset. Teachers can therefore foster different mindsets among their students through how they react to failures and mistakes (Cimpian et al., 2007; Rattan et al., 2012, 2015).
Several studies in various educational fields have found that a fixed mindset leads to avoidance of challenges and setting lower goals, which leads to the risk of the learner becoming demotivated over time (Lou & Noels, 2017, 2019; Mercer, 2011). The role of students’ mindsets has been investigated in various learning domains such as math, science, and language learning (Albalawi & Al-Hoorie, 2021; Blackwell et al., 2007; Sun et al., 2021). Some of this research has investigated medical students’ mindsets and the role those mindsets play in their academic performance. For medical students, however, Kustritz (2017) has found no significant association between the growth mindset of veterinary medicine students and their academic performance in traditional learning.
Initial investigations into mindsets primarily occurred in laboratory environments. Interventions carried out in non-laboratory settings yielded varying outcomes. A meta-analysis of such interventions in real-world settings indicated that altering mindsets resulted in minimal enhancements in academic performance (Sisk et al., 2018) or no improvements at all (Foliano et al., 2019). Nonetheless, interventions targeting mindsets appear to offer significant benefits to students facing academic challenges, particularly those from disadvantaged socioeconomic backgrounds. Additionally, recent nationwide initiatives involving high school students in the United States demonstrated the effectiveness of mindset interventions when peer attitudes aligned with the intervention’s messages (Yeager et al., 2016).
When it comes to language acquisition, Mercer (2011) made a parallel argument, maintaining that individuals with a fixed mindset regarding language learning tend to shy away from challenges and set lower aspirations, thereby increasing the likelihood of becoming demotivated over time (see also Lou & Noels, 2017, 2019). Similarly, recent research by Lou and Noels (2020) revealed that migrants embracing a growth-oriented mindset toward language acquisition reported reduced anxiety, increased language use, and enhanced proficiency levels, even after controlling for baseline proficiency.
Engagement and Effort
Student engagement was defined by Gunuc and Kuzu (2014) as “the quality and quantity of students’ psychological, cognitive, emotional and behavioral reactions to the learning process as well as to in-class/out-of-class academic and social activities to achieve successful learning outcomes” (p. 588). Within educational psychology, engagement is a broad term that encompasses students’ diverse motivational, cognitive, and behavioral characteristics (Appleton et al., 2008; Baron & Corbin, 2012; Fredricks et al., 2004; Sharma & Bhaumik, 2013). Engagement is also closely related to the concept of action and the level of a student’s mental/physical active participation in a learning task (Skinner & Pitzer, 2012).
Engagement has always been regarded as being connected to effort. Some researchers have argued that effort is a key ingredient of engagement and numerous studies have found that effort is related to academic performance (Johnson et al., 2001). Therefore, research into engagement has depended largely on effort measures such as the completion of homework, attentiveness, and preparedness (Carbonaro, 2005). Within traditional learning, it refers to all activities that the learner participates in regardless of the type or amount of engagement in the learning process (Hiver et al., 2024). Kassab et al. (2015) has examined the role of effort regulation in Bahraini medical students’ achievement at the blended learning classes. The results revealed that regulating effort directly affected student achievement.
In medical education, interaction is part of how medical students gain medical skills and knowledge via both instructors and patients. Interaction is thus also a key indicator for academic achievement in the medical context (Holzinger et al., 2009). While Holzinger et al. (2009) showed that simulation software is beneficial for undergraduate medical students learning complex concepts, Leksuwankun et al. (2022) demonstrated the importance of extracurricular activities on the GPA of pre-clerkship students. Both studies supported the role of engagement and interaction in traditional learning. Moreover, Kay and Pasarica (2019) observed differences between preclinical and clinical medicine student engagement and found that the preclinical students attended both completed their online sessions and assignments more regularly than did clinical students. Although reviewing the literature shows that student engagement is related to the high quality in learning outcomes (Krause & Coates, 2008) and it explains some of the learning motives (Hiver et al., 2024), further investigation is needed to examine the role of engagement and effort in online learning within the medical educational field.
Satisfaction
Students’ satisfaction is a complex concept that could be defined as the students’ evaluation of their educational experiences and outcomes (Elliott & Shin, 2002). It consists of various dimensions and combines different elements including, for example, workload, teaching methods, non-educational services (e.g., housing and computer facilities), and overall climate (Harvey, 1995; Marzo-Navarro et al., 2005; Richardson, 2005). The concept can be classified into three levels depending on the type of relationships involved: personal, interpersonal, and organizational (Elliott & Healy, 2001). Students’ satisfaction is considered a core element in higher education for determining the quality of different areas of educational programs such as teachers’ styles, curriculum presentation, and institutions’ services (Dziuban et al., 2015). While students’ relationship with the educational environment falls under the organizational domain and is influenced by the teachers’ numbers, patients, teaching methods, and learning mode, the relationship between students and clinical faculty falls under the interpersonal domain in the medical context (Ziaee et al., 2004).
A significant positive correlation between medical student satisfaction and behavioral engagement was observed in traditional education by Xia et al. (2023). Furthermore, Al-Omairi and Hew (2022) administered a survey to 564 Omani students who enrolled in online courses and concluded that student–instructor and student–student interactions influenced students’ satisfaction the most. It is worth noting that previous research in the medical context has demonstrated that using blended and e-learning approaches in teaching courses like anatomy and physiology reported higher student satisfaction levels, especially with content delivery (Gray & Tobin, 2010; Ruiz et al., 2006). Although Bean and Bradley (1986) reported a positive influence of student satisfaction on students’ academic achievement in a traditional learning context, relatively little is known about satisfaction and academic performance in online learning, especially in the medical context.
Motivation
Student motivation combines the personal and environmental factors that shape the learning process. Personal factors relate to intrinsic motivation, students’ goals, or any factors related to what is pushing the student to learn, making the student perceive the learning as valuable. Environmental factors relate to extrinsic motivation, which stems from features of the social environment such as rewards and social responsibilities (Ryan & Deci, 2020).
To further investigate the role of motivation in medical students’ traditional learning, Wu et al. (2020) conducted a study in which 1,930 medical students, enrolled in 10 universities, responded to an electronic questionnaire. The results revealed that male students had significantly higher intrinsic motivation but lower academic performance than female participants. They concluded that academic performance was positively associated with intrinsic motivation. On the other hand, Sreeramareddy et al. (2007) found that the high parental expectations were a source of stress for medical students in Nepal. These family pressures and social issues affect students’ motivation negatively. To further investigate the stressors of medical students in Saudi Arabia, Soliman (2014) conducted a cross-sectional study on 319 medical students and concluded that the parents’ high expectations was one of the stressors that medical students faced, especially in the first year. Overall, previous studies highlight the need for motivation in the medical education field.
To date, there has been very little research that directly investigated psychological predictors of academic achievement/performance in the medical educational field in the Saudi context. The present paper therefore aims to address this research gap by examining the relationship between several psychological predictors (mindset, motivation, satisfaction, self-efficacy, engagement) and the academic performance of medical students in a Saudi higher education context.
Method
Participants
A total of 585 Saudi medical students (55% female) volunteered to take part in this study. They were at various levels of their studies (10.9% first year, 18.5% second year, 24.4% third year, 21.0% fourth year, 13.8% fifth year, 10.8% sixth year, and 0.5% seventh year). About 88% were in the age range of 18 to 24, with the remaining being older. Of these participants, 423 were preclinical students while 135 were clinical. The participants were full-time education students at a public Saudi university that follows a six-year curriculum.
Instruments
A questionnaire adopting 6-point Likert scales (1 = strongly disagree, 6 = strongly agree) was designed and piloted by the authors on a sample from the target population. These scales were self-efficacy, fixed and growth mindsets, extrinsic motivation, behavioral engagement, effort, and satisfaction (see Appendix). These psychological factors included in this study were selected based on their established importance in literature as significant predictors of the academic performance of the medical students (Gray & Tobin, 2010; Leksuwankun et al., 2022; Zheng et al., 2020). The study focused on these specific factors due to their empirical implications, and theoretical grounding in Saudi medical context. The second part of the questionnaire consisted of four demographic questions in which the students were asked about their age, the year of studying, gender, and GPA. Finally, participants were asked to evaluate themselves regarding resilience against challenges/difficulties encountered in online learning.
In the current study, GPA was selected as a proxy for students’ achievement and the dependent variable. GPA is commonly used in educational research as a standardized and quantifiable measure of overall academic performance. While GPA reflects the ability of the students to master course content and meet the academic requirements, it may not fully capture all aspects of achievement in the online learning environment, such as skill acquisition. Thus, our use of GPA should be considered as one representation of academic performance, rather than a comprehensive measure of achievement.
According to Mackey and Gass (2005), using the participants’ first language in such instruments helps remove concerns about their proficiency that may affect the validity of the collected data. Thus, to avoid any language interference, the study was administered in Arabic, the participants’ first language.
Procedure
The questionnaire included an initial pool of items that were reviewed by experts to examine its validity. Then, based on the experts’ feedback, the items were modified and translated. A pilot study was also conducted with a group of 161 participants to check the reliability and validity of the questionnaire. The results of this pilot study were used to improve the clarity of scale items. The reliability of validity of the main study is reported in the “Results.”
This study used a purposeful sampling technique, as only specific participants were selected to participate (Creswell, 2012). The study was conducted during the COVID-19 pandemic in 2020, when all classes were delivered virtually. IRB approval was obtained from the Local Research Ethics Committee at Tabuk University (No. UT-339-176-2024).
Data Analysis
The data were analyzed over two stages. At the first stage, the validity of the scales was investigated. The unidimensionality of the scales was tested statistically using parallel analysis and visually using a scree plot. Unidimensionality is essential for the interpretability and soundness of the results. At the second stage, hierarchical multiple regression was conducted to find out what variables predicted GPA. In the first step, gender and year of study were entered. In the second step, the remaining variables were entered. This was intended to find out which variables were able to predict achievement over and above gender and year of study. The same analytical procedure was then applied to preclinical and clinical students separately. Data were screened for outliers, and a few values exceeding z = ±3.0 were excluded. The Bonferroni correction was applied to reduce the risk of Type I error inflation given the multiple psychological predictors tested simultaneously in the regression model. We note that this is a conservative approach that adjusts the threshold for statistical significance in proportion to the number of comparisons made—in our case, adjusting the standard alpha level (p < .05) by dividing it by the number of predictors included.
Results
Scale Validation and Descriptive Statistics
In order to establish the validity of the questionnaire scales uses in this study, they were entered into exploratory factor analysis using JASP 0.14.1 (JASP Team, 2020). Parallel analysis was implemented with direct oblimin with loadings below 0.30 suppressed. As shown in Table 1 and the scree plot in Figure 1, the results support the presence of seven factors underlying the data. A few cross-loadings were found, but all were small (<0.40).
Exploratory Factor Analysis Results.
Note. Loadings below 0.30 were suppressed.
Scree plot.
Table 2 presents the reliabilities. All reliabilities were over .80, with the exception of Growth Mindset (.68). Table 2 also shows that the correlation between Growth Mindset and Fixed Mindset was rather small (−.18). The strongest correlation was between GPA and Engagement.
Reliabilities, Means, Standard Deviations, and Correlations Among the Scales.
p < .10. *p < .05. **p < .01. ***p < .001.
Hierarchical Multiple Regression
To answer RQ1, a two-step multiple regression analysis was then conducted. In the first step, Gender and Year of Study were entered into the equation, while the other variables were entered into the second step. Enter method was applied in order to use all variables in the dataset. The results are presented in Table 3. Gender was a significant predictor, indicating that female students outperformed male students. The second step accounted for 12% of the variance in GPA. Out of the seven predictors in this step, only Engagement turned out to be significant, which remained significant even after using Bonferroni adjustment. Another unexpected finding is that Growth Mindset was negatively associated with GPA. However, the magnitude of this association was negligible (−0.10) and its significance value did not remain significant after the Bonferroni correction. Thus, it seems safe to state that this finding might not be meaningful until it is replicated.
Multiple Regression Results.
To answer RQ2, the analysis was conducted to compare preclinical and clinical students (Table 4). The results showed very similar results, with the exception that Engagement remained a significant predictor of GPA only for preclinical students.
Multiple Regression Results for Preclinical and Clinical Students.
Discussion
The aim of this study was to investigate the psychological factors involved in medical students’ achievement in online classes. Specifically, the factors investigated were self-efficacy, extrinsic motivation, growth and fixed mindsets, engagement, effort, and satisfaction.
The most robust finding to emerge from this study was that engagement turned out to be a significant predictor of academic achievement. This result is in line with those of Holzinger et al. (2009) and Leksuwankun et al. (2022) who found that engagement and interaction activities were correlated positively with medical students’ GPAs in traditional learning. While Holzinger et al. (2009) highlighted the importance of collaborative learning and peer discussion, Leksuwankun et al. (2022) emphasized the role of group activities to foster motivation and academic achievement. The current study provides more support that engagement has a significant impact on the academic performance of medical students in online learning. These findings also suggest that participation during online classes and engagement with the lecturers and other students has a positive impact on medical students’ academic achievement. Additionally, interaction and involvement in group discussions may eventually have a positive impact on GPA.
The results also indicated that engagement was particularly relevant in the early years of medical school. A possible explanation is that preclinical students need to attend classes more regularly to be engaged with their peers and lecturers. During this formative phase, the degree of student engagement hinges on the institution and lecturers working collaboratively to create a supportive environment. On the other hand, in the later clinical phase, students are more focused on practicing skills with the guidance of their supervisors. They tend to assume greater responsibility for their learning, potentially reducing their reliance on external support resources. The current study adds further evidence that medical student engagement is associated with their academic achievement in online learning.
Although student effort is a main factor in their engagement, there was no significant relation with medical students’ achievement in online learning. This could be related to the nature of medical courses, the material of which requires more interaction and engagement with teachers and patients to be understood. This outcome seems inconsistent with that of Kassab et al. (2015) who established a relationship between effort and achievement. This inconsistency may be due to the mode of learning in Kassab et al., which was blended learning. According to Kaur (2013), blended learning gives instructors and students the best of traditional and online learning due to its flexibility and accessibility, and helps students overcome the challenges of these two educational worlds. Another possible explanation for this finding is that the current study aimed to investigate academic achievement in the medical context during COVID-19, during which time students were exposed to a lot of pressure, anxiety, and social challenges.
However, while Bean and Bradley (1986) established a relationship between student satisfaction and academic performance in a traditional learning context, the results of our study indicated that although students may be satisfied with the online learning mode, there was a low positive correlation between student satisfaction and academic achievement which was not significant in the regression model. This suggests that the medical students expressed their interest in online activities, but this was not related to their effort level and performance. This inconsistency may suggest that satisfaction per se may not be a determining factor in achievement, and that what students find satisfactory may not always align with what is actually useful for their learning.
In terms of self-efficacy, the current study found no significant relation between the academic achievement of medical students in online learning and self-efficacy. However, a relationship was found between self-efficacy and three variables: satisfaction, engagement, and growth mindset. Although medical students believed in their abilities and thereby were more satisfied and more engaged in online classes, their performance was not necessarily directly affected by these beliefs. These results are consistent with data obtained by Wu et al. (2020), who could not find a significant direct effect of extrinsic motivation and self-efficacy on the academic performance of medical students in 10 universities and colleges in China. One explanation for this pattern is that although the medical students were focused on their abilities and skills to determine their academic objectives, this was not enough to improve their performance. External effort and institutional engagement activities in online learning may be needed to improve their academic performance.
One surprising variable that was not significantly associated with academic achievement was growth mindset. Although this factor was positively correlated with self-efficacy, it correlated negatively with students’ GPA, but this was not found to be significant following the Bonferroni correction. Replication of this finding is therefore necessary before a claim can be made about this potential relationship. This outcome somewhat aligns with the findings by Kustritz (2017), who found no significant association between the growth mindset of veterinary medicine students and their academic performance in traditional learning. However, fixed mindset correlated positively with extrinsic motivation while growth mindset correlated positively with self-efficacy. Interestingly, these relationships may indicate that students who hold a fixed mindset are more likely to benefit more from extrinsic motivation in order to carry on their learning journey, while students who hold a growth mindset tend to rely more on their self-efficacy. Further research is needed to test these hypotheses.
Conclusion
This study set out to examine the relationship between medical students’ achievement in online education and a number of psychological factors in Saudi Arabia. Student engagement was a significant predictor of GPA, particularly in the early years of medical context. However, no significant relationships were found between self-efficacy, extrinsic motivation, growth and fixed mindsets, and students’ performance. These findings suggest the need for more fine-tuned engagement-related empirical research to investigate how various dimensions of engagement influence enhance students’ performance.
There are certain limitations to this study in terms of its methodology, as the quantitative approach was adopted without integrating any qualitative methods to obtain a deeper understanding of the role psychological factors might be playing in our context. Conducting semi-structured interviews, for example, with students can help teachers to understand the differences between males and females, as well the variations in the preclinical and clinical phases. Additionally, it would be valuable to compare the perceptions of students and teachers to explore factors relevant to online education. Another potential area of future research would be to observe online classrooms and investigate students’ motivation in the online mode. This would provide an opportunity to further compare the current results with the traditional context. Research is also needed to shed light on the causality in the association between student engagement and academic achievement, providing insights into both the presence of this causal relationship and the degree to which it influences outcomes.
Footnotes
Appendix: Questionnaire Items
ORCID iDs
Ethical Considerations
IRB approval was obtained from the Local Research Ethics Committee at Tabuk University (No. UT-339-176-2024) on February 4, 2024.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
The datasets used or produced in this study can be obtained from the corresponding author upon reasonable request.