56 © 2025 Conscientia Beam. All Rights Reserved. Examining the reliability and validity of self-efficacy beliefs, stress, perceived teachers' support and academic burnout scales using the PLS-SEM approach Jingyuan Li1 Yoon Fah Lay2+ 1Faculty of Social Sciences and Liberal Arts, UCSI University, Kuala Lumpur, Malaysia. Email: 1002161218@ucsiuniversity.edu.my 2Faculty of Psychology and Education, Universiti Malaysia Sabah, Kota Kinabalu; Faculty of Social Sciences and Liberal Arts, UCSI University, Kuala Lumpur & School of Liberal Arts and Sciences, Taylor's University, Kuala Lumpur, Malaysia. Email: layyf@ums.edu.my (+ Corresponding author) ABSTRACT Article History Received: 13 November 2023 Revised: 11 October 2024 Accepted: 31 October 2024 Published: 21 November 2024 Keywords Academic burnout Perceived teachers' support PLS-SEM approach Stress Students' self-efficacy beliefs. Likert scales were used in this study to collect data on measures such as the student academic burnout scale, the student stress scale, the student self-efficacy beliefs scale and the student perceived teacher support scale. The purpose of this study is to examine the four-part scales' reliability and validity used in this research. The number of measurement indicators for the four scales was 11, 13, 12 and 10, respectively. Seventy-five college students from five colleges and universities participated in the research and the partial least squares structural equation modelling (PLS-SEM) approach was applied to analyze the data. Consequently, the internal consistency and reliability of the measures were assessed using Cronbach's alpha (CA) and composite reliability (CR) both of which exceeded the clinical thresholds of 0.6 and 0.7, respectively. The average variance extracted (AVE) was used to evaluate the scales' convergent validity and the reported values were all stated above 0.5. The scales' discriminant validity was also framed within the range of threshold values. As a result, the scales used in this study demonstrated good validity and reliability and can be useful in assessing relationships throughout a range of study situations. Contribution/Originality: This study employed the second-generation statistical method, PLS-SEM, in a novel approach to assess the reliability and validity of the scales used in the study. In terms of research methodology, it offers a new method for evaluating the reliability and validity of research instruments. 1. INTRODUCTION Training talent for societal growth is one of the three responsibilities of contemporary higher education. However, students' academic achievement is now a crucial factor in defining their level of achievement in China's modern university education system. Although this standard is relatively one-sided, the degree of academic scores is representative of evaluating students' academic performance to some extent. Many factors influence college students' academic achievement. Extensive studies have been undertaken on student self-efficacy beliefs, stress, student perceptions of teacher support and academic burnout and it is considered that these elements have substantial impacts on student academic performance. However, outcomes may change when studies are undertaken on different research populations. It is necessary to validate the research instrument before commencing the Humanities and Social Sciences Letters 2025 Vol. 13, No. 1, pp. 56-68 ISSN(e): 2312-4318 ISSN(p): 2312-5659 DOI: 10.18488/73.v13i1.3987 © 2025 Conscientia Beam. All Rights Reserved. mailto:1002161218@ucsiuniversity.edu.my mailto:layyf@ums.edu.my https://www.doi.org/10.18488/73.v13i1.3987 Humanities and Social Sciences Letters, 2025, 13(1): 56-68 57 © 2025 Conscientia Beam. All Rights Reserved. research since the research population in this study consists of regular undergraduate and junior college students in Shandong Province (China). Structural equation modelling is beneficial in discussing complex models with relevant simple appliances (Dash & Paul, 2021) but selecting the appropriate approach takes much work. The two most popular approaches researchers have applied to conduct research are covariance-based structural equation modelling (CB-SEM) and Partial least square structural equation modelling (PLS-SEM). Researchers have determined that PLS-SEM is more useful for assessing the composite-based mode while the CB-SEM appliance is more appropriate for estimating the factor-based model (Dash & Paul, 2021; Hair, Ringle, & Sarstedt, 2011; Henseler, Hubona, & Ray, 2016). Researchers are increasingly using second-generation statistical approaches as opposed to first-generation techniques like factor analysis and regression analysis (Hair, Hult, Ringle, & Sarstedt, 2017). According to Hair et al. (2011), PLS-SEM was highly influential in testing causal models and an increasing number of researchers are using this method to investigate the connections between endogenous and exogenous constructs. Researchers may choose the PLS-SEM approach for various reasons including a complex structural model, a relatively small sample size, a lack of normal distribution, etc. (Ghasemy, Teeroovengadum, Becker, & Ringle, 2020; Hair, Risher, Sarstedt, & Ringle, 2019). The aforementioned reasons support the notion that it is more beneficial to do more research using the PLS-SEM strategy rather than the first-generation statistical method. SPSS software and the CB-SEM method are usually used to validate the internal consistency reliability of the four scales (SSBS, SSS, SPTSS and SABS) from the perspective of content reliability and CA while the scales' convergent validity and discriminant validity were not mentioned. Hence, it is crucial to use the PLS-SEM approach to assess the reliability and validity of the scales used in the study to confirm their internal consistency and reliability, convergent validity and discriminant validity before undertaking the actual research. 2. LITERATURE REVIEW 2.1 Students' Self-Efficacy Beliefs Scale Previous research has shown that students' self-efficacy beliefs (SSB) have a significant and favourable influence on their academic achievement. Specifically, academics have increasingly focused on college students at the higher education level as a prominent demographic. The SSB is a highly influential factor in shaping human behavior and emotions (Hamann, Pilotti, & Wilson, 2020; Viviers, De Villiers, & van der Merwe, 2023). It holds a prominent position as a critical focus of academic research. It is also an essential manifestation of self-efficacy in education. Based on the result expectations and efficacy expectations dimensions provided by Bandura (1977), Chinese researchers (Guo & Su, 2021; Liang, 2000; Xu, Luo, Yu, Tian, & Zhu, 2021) assessed students' self-efficacy beliefs from the perspectives of learning capacity and learning behaviour based on Bandura's (1977) outcome expectations and efficacy expectations dimensions. High self-efficacy students will maintain a positive outlook when they experience difficulties with their learning and build themselves up when they experience failures. Additionally, they believe they can overcome obstacles in the learning process by applying effort. Meanwhile, students who have low academic efficacy are more susceptible compared to their peers (Aftab, Shah, & Mehmood, 2012; Demirel, Türkel, & Aydin, 2020; Hwang, 2021). When encountering obstacles, they could use poor coping strategies, avoid problems or even give up on trying to solve them. Research indicates variations in self-efficacy views among college students and discrepancies in perception across different groups. The study by Basith, Syahputra, and Ichwanto (2020) contradicted this perspective asserting no substantial gender disparities in self-efficacy views among college students. According to Saleh, Camart, and Romo (2017), French university students typically exhibit low levels of self-efficacy. However, male students demonstrate significantly higher self-efficacy levels than their female counterparts. Hence, the researcher's goal for the study should be taken into consideration while choosing a scale. Humanities and Social Sciences Letters, 2025, 13(1): 56-68 58 © 2025 Conscientia Beam. All Rights Reserved. The Students' Self-Efficacy Beliefs Scale (SSBS) used in this research was compiled initially by Liang (2000) and consisted of two dimensions: learning ability efficacy (LAE) and learning behavior efficacy(LBE). Higher scores suggest higher levels of self-efficacy beliefs. The 5-point Likert scale with 22 items in this study was converted to a 7-point Likert scale while keeping the original items to ensure the measures' accuracy. CA values were recorded as 0.820 and 0.752 in the two dimensions on the original scale. Xu et al. (2021) used a consistent measurement scale to evaluate the relationship between SSB and engagement. The findings indicated that CA coefficients for LAB and LBE were 0.780 and 0.710, respectively. Additionally, the overall Cronbach's alpha coefficient for the scale was 0.829. During the same period, Chen and Zhou (2021) compiled a 12-item questionnaire based on SSBS to examine the students' self-efficacy and belief level. The instrument exhibited an overall alpha coefficient of 0.823. Furthermore, the two sub-dimensions of the questionnaire showed CA coefficients of 0.834 and 0.573, respectively. 2.2. Student Stress Scale Student performance in colleges and universities is also influenced by factors related to self-identity, interpersonal relationships, academics and future development in addition to self-efficacy beliefs (Acharya, Jin, & Collins, 2018; Li, Yang, Zhou, Zhao, & Liu, 2022; Othman, Ahmad, El Morr, & Ritvo, 2019; Satpathy, Siddiqui, Parida, & Sutar, 2021). Academic stress is typically regarded as the stressor that has the most considerable effect on students among these stressors (Li et al., 2022; Satpathy et al., 2021). According to the former researchers (Alduais et al., 2022; Deng et al., 2022), stress can negatively affect college students' mental and physical health, ultimately affecting their academic performance. According to the cognitive appraisal theory of stress, there are four ways in which stress can impact an individual: prospective stressors, cognitive evaluation, coping and reaction (Lazarus & Folkman, 1984). The China College Student Psychological Stress Scale (CCSPSS) developed by Liang and Hao (2005) served as a model for the Student Stress Scale (SSS) which measures the stress levels of college students in Shandong Province. It maintains the four main sources of stress that students face: academic stress, interpersonal stress, future development stress and student life stress. In the original scale, the number of items for the four dimensions was 18, 18, 10 and 28 and the alpha values reported for the four sub-scales were 0.860, 0.840, 0.810 and 0.870, respectively. The overall Cronbach's alpha was 0.960. In the study investigating the correlation between stress and adaptation in college students, Ma, Qu, Yan, and Fu (2017) evaluated the precision and uniformity of the scale used in the research to examine the relationship between stress and adaptability in college students. They discovered that the overall CA of the scale evaluated three times surpassed 0.950. Two retests also demonstrated a retest reliability exceeding 0.700. The researchers determined that the China College Student Psychological Stress Scale (CCSPSS) exhibited both reliability and validity. An and Pei (2017) used the CCSPS to create the student stress questionnaire which was designed to investigate the relationship between academic stress and emotional intelligence in university students. As a result, the questionnaire proved to be highly reliable and valid with an internal consistency reliability value of 0.90, a composite reliability value of 0.95, and an average variance extracted value of 0.61. 2.3. Student Perceived Teachers' Support Scale Social support theory (Cullen, 1994) focuses on the adaptation of human beings to society and the utilization of social resources and teachers' support belongs to the perspective of micro-level support. Researchers often employed the analytical dimensions of emotional support, learning support and competency support to measure students' perceptions of teachers' support (Abdullah, Shamsi, Jenatabadi, Ng, & Mentri, 2022; Liu, Du, & Lu, 2022). Humanities and Social Sciences Letters, 2025, 13(1): 56-68 59 © 2025 Conscientia Beam. All Rights Reserved. Therefore, teachers are regarded as the closest partners for students in the learning process. Students benefit greatly from the assistance that professors offer them, whether they are learning in-person or virtually (Brandisauskiene et al., 2021; Frazier, Gabriel, Merians, & Lust, 2019) just as it does to help students reduce uncertainty and insecurity (Abdullah et al., 2022) as well as stimulate students' learning participation (Liu et al., 2022). In particular, the support provided by teachers plays an irreplaceable role in improving students' learning environment (Abdullah et al., 2022), stimulating students' participation in learning (Liu et al., 2022) and affecting college students' academic mood (Hao, Cui, & Chiu, 2018). The Student Perceived Instructors' Support Scale (SPTSS) was developed by OuYang (2005) and has three dimensions: learning support (LS), emotional support (ES) and capacity support (CS). This allows researchers to analyse how college students perceive support from their teachers. Simultaneously, values of the composite reliability for the three dimensions were reported as 0.840, 0.730 and 0.790. The questionnaire's overall reliability was 0.870. Chen and Ma (2022) used the student perceived teachers' support behaviour questionnaire (SPTSBQ) to evaluate the relationship between students' information literacy skills and the perceived support from their teachers. The study produced a CA coefficient of 0.87 indicating a high degree of internal consistency. Furthermore, the three sub-dimensions of the construct demonstrated satisfactory reliability with alpha values of 0.84, 0.73, and 0.79, respectively. Chen and Tu (2019) revealed that the CA values for each of the three sub-dimensions were 0.929, 0.897 and 0.889 in a different investigation that also used this scale. The total alpha coefficient of this study was 0.937. All of which indicate that the questionnaire has high reliability and validity. 2.4. Student Academic Burnout Scale Scholars initially used the term "burnout" to evaluate the mental health of individuals at work due to excessive demands on energy, strength and resources. Later, the researchers noticed that students had diverse levels of burnout in the learning process which were influenced by factors such as age, gender and even parental education (Chahid, Ahami, Chigr, & Najimi, 2018; Hyytinen, Tuononen, Nevgi, & Toom, 2022) and displayed features of various types of academic burnout. Students' levels of academic burnout eventually have a detrimental effect on their academic performance, independent of the form of academic burnout, emotional weariness, sense of cynicism towards school or inadequacy (Asikainen, Salmela-Aro, Parpala, & Katajavuori, 2020; Madigan & Curran, 2020). The extent of academic burnout and its effect on academic performance similarly demonstrated an upward trend over time (Asikainen et al., 2020; Madigan & Curran, 2020; Raisanen, Postareff, & Lindblom-Ylanne, 2021; Yu, Yin, Zhao, & Xin, 2020). Lian, Yang, and Wu (2005) developed the undergraduates' learning burnout scale (ULBS) to assess college students' academic burnout in learning contexts and the resulting burnout behavior. It contains three dimensions: emotional exhaustion (EE), improper behavior (IB) and low sense of achievement (LSA) and 8, 6 and 6 items comprise each dimension. According to Lian et al. (2005), the overall Cronbach's alpha was 0.865 and the values for the three dimensions were 0.812, 0.704 and 0.731, respectively. The scale applied in the current study is compiled from USBS by altering the 5-point Likert scale to a 7-point Likert scale and 1 to 7 were used to signify the degree of student assent to the observed indicator, from strongly disagree to strongly agree. Chen and Zhou (2021) conducted a study to examine the relationship between academic self-efficacy, academic stress and learning burnout. They found the total alpha coefficient for SABS was 0.750 and Cronbach's alpha of the sub-dimensions was 0.817 for EE, 0.676 for IB, and 0.326 for LSA. Wu, Yu, An, and Li (2021) applied the same scale to investigate the effect of college students' academic burnout emotions on disciplinary competitions using LBS. In the reliability test of the scale, the total Cronbach's alpha was 0.887, 0. 867 for EE, 0.766 for IB and 0.745 for LSA. Liu, Zhou, Li, Wang, and Teng (2018) used the CSABS to understand the level of academic burnout and Humanities and Social Sciences Letters, 2025, 13(1): 56-68 60 © 2025 Conscientia Beam. All Rights Reserved. its relationship with college students' academic stress and psychological toughness. An internal consistency coefficient of 0.750 from the factor analysis confirmed the questionnaire's good structural validity. 3. METHODOLOGY 3.1. Research Instruments Four scales were used to collect data for this investigation. First, there is the Students' Self-Efficacy Beliefs Scale (SSBS), a 22-item measure that assesses students' LBE and LAE (Liang, 2000). The scale used to measure the stress of college students during college life and study is the Student Stress Scale (SSS). The four facets of primary student stress examined in the study are student academic stress (SAS), student life stress (SLS), student interpersonal stress (SIS) and student further development stress (Liang & Hao, 2005). Meanwhile, the Student Perceived Teachers' Support Scale (SPTSS) was the primary tool used to assess the level of support that students perceived from teachers in their academic and personal lives in the domains of "learning support" (LS) and "capacity support" (CS) (OuYang, 2005). Finally, the Student Academic Burnout Scale (SABS) was used to measure the level of academic burnout among college students (Lian et al., 2005). Researchers obtained the student profiles for inappropriate behaviour (IB), emotional exhaustion (EE) and low sense of accomplishment (LSA). Appendix 1 presents the content of some of the question items on the scales used in the study. 3.2. Procedure The quota and random sampling methods were employed in this investigation. The researcher divided the 153 universities in Shandong province (China) into four groups according to their types before starting the sampling process: private colleges and universities, provincial undergraduate universities, provincial higher vocational colleges and colleges and universities under the central ministry's authority. In the first stage, five colleges and universities were selected from the 4 clusters using a quota sampling method. Fifteen college students from each university were randomly selected by simple random sampling from the 4 clusters in the second stage. Shandong University was chosen to represent the central ministry's colleges and universities, Linyi University was picked to represent the provincial undergraduate universities, Binzhou Polytechnic was determined to represent the provincial higher vocational colleges, QiLu Institute of Technology and Qilu Medical College were assigned to represent the Shandong Province's private colleges and universities. There were a total of 75 respondents employed for this investigation. The researchers initially explained the goal of the study and the support required from teachers and school administrators at the five colleges and institutions they had chosen in order to collect data for the study. Simple random selection was used to choose 75 students after getting the schools' consent. This was followed by briefing the students on the purpose of the survey and the precautions to be taken in completing the questionnaire such as that the survey was voluntary, they were allowed to leave in the middle of the survey, the questionnaire was submitted anonymously, and the data obtained would be used only for conducting educational research. A teaching staff member was present to help with any problems. 3.3. Data Analysis The current study employed quantitative research, collected data using cross-sectional surveys and used Smart PLS software 4.0.8.7 to assess the scale's validity and reliability. Data screening was first carried out on the self- reports of the returned respondents in PLS-SEM data analysis to ensure that missing values, deviations and suspicious matching patterns were not included in the data and to verify that all the data examined was valid. The questionnaire for this study contained four sub-scales with 46 items. Cronbach's alpha and composite reliability were used to examine the internal consistency and reliability of the scales. Convergent validity for the Humanities and Social Sciences Letters, 2025, 13(1): 56-68 61 © 2025 Conscientia Beam. All Rights Reserved. instruments was mainly assessed by the Average Variance Extracted and three critical criteria for discriminant validity: the Fornell-Larker criterion, cross-loading and the Heterotrait-Monotrait ratio (HTMT). 4. RESULTS 4.1. Internal Consistency Reliability for the Instruments Generally, the construct's CR value and the CA were used as metrics to evaluate the scale's internal consistency and reliability. The former is the most conventional and widely used criterion for assessing the internal consistency reliability of instruments in social science research (Hair et al., 2017) and this way of measuring reliability is proposed based on variable correlations. It is often assumed that in exploratory research, the measurement tool's reliability is adequate when the alpha coefficient hits 0.7 and high when the coefficient is between 0.7 and 0.9. Table 1 illustrates the CA values for the four latent components. Exactly, CA for construct SAB was presumed to be 0.898, 0.921 for SPTS, 0.930 for SS and 0.902 for SSB. All of the results were greater than 0.7. Table 1. The constructs' internal consistency and reliability. Construct Cronbach's alpha Composite reliability (CR) Average variance extracted (AVE) SAB 0.898 0.907 0.527 SPTS 0.921 0.930 0.540 SS 0.930 0.940 0.545 SSB 0.902 0.903 0.509 Note: SAB= Student academic burnout, SPTS=Students perceived teachers' support, SS= Student stress, and SSB=Students' self-efficacy beliefs. Another criterion to evaluate the internal consistency and reliability of the instrument is composite reliability based on the latent constructs. The composite reliability values for SAB, SPTS, SS and SSB were 0.907, 0.930, 0.940 and 0.903, respectively based on the four reflecting constructs of the research as indicated in Table 1. The four CRs for the latent constructs were more significant than the threshold value (0.7). In addition, it was shown that the instrument's internal consistency and reliability had been established by combining CA values with CRs. 4.2. Convergent Validity for the Instruments Table 1 shows the AVE values for the constructs. The AVE values for constructs SSB, SS, SPTS and SAB are 0.509, 0.545, 0.540 and 0.527 as indicated. Hence, all four AVE values represented have exceeded the threshold value (0.5) and the convergent validity has been validated from the perspective of the average variance extracted. 4.3. Discriminant Validity for the Instruments Three criteria are necessary to evaluate the Heterotrait-monotrait (HTMT) ratio of correlations, cross-loading and the Fornell-Larcker criterion in order to determine the instrument's discriminant validity. Cross-loading which concentrates on the indicators is the first criteria to assess discriminant validity. Any item's loading should always be greater than its cross-loadings. Table 2 and Figure 1 illustrate the loading and outer loadings for all the items. Constructs SAB comprises ten items: SAB_EE1, 2, SAB_IB4, 7 and SAB_LSA1, 2, 3, 4, 5, 7. Table 2 shows that all 10 item loadings on the correlation construct SAB are greater than cross-loadings on the other three constructs (SPTS, SS, and SSB). Constructs SS and SSB exhibit cross-loadings on the other constructs. However, the loadings on the particular construct are more substantial similar to those of SAB and SPTS. Loadings for items SS_SAS2, 4, 5, 6, 7, SS_SFDS6, 11, SS_SIS1, 2, 3 and SS_SLS1, 2, 9 are the largest on construct SS. Meanwhile, loadings on SSB construction are the same. Loadings on items SSB_LAE1, 2, 3, 4, 6, 8, 9, 10 and SSB_LBE2, 8, 10 exceed cross- loadings on constructs SAB, SS, and SPTS. The loadings of the indicators on the respective structures are all greater than the cross-loadings on the other constructs. Humanities and Social Sciences Letters, 2025, 13(1): 56-68 62 © 2025 Conscientia Beam. All Rights Reserved. There were 14 items with a loading value smaller than 0.7 in Figure 1 including SAB_EE1, SAB_LSA2, SAB_LSA3, SAB_LSA7, SPTS_CS1, SPTS_LS1, SSB_LAE2, SSB_LAE10, SSB_LBE8, SSB_LBE10, SS_SAS2, SS_SIS1, SS_SIS2 and SS_SLS1. However, these items were retained, considering that the CA, CR and AVE values for the constructs in which these items were located all exceeded the minimum threshold required. Table 2. Cross-loadings for constructs SAB, SPTS, SS and SSB. Item SAB SPTS SS SSB SAB_EE1 0.677 -0.359 0.344 -0.433 SAB_EE3 0.756 -0.509 0.381 -0.419 SAB_IB4 0.780 -0.409 0.265 -0.465 SAB_IB7 0.782 -0.436 0.371 -0.384 SAB_LSA1 0.779 -0.452 0.279 -0.436 SAB_LSA2 0.579 -0.286 0.153 -0.452 SAB_LSA3 0.671 -0.371 0.319 -0.467 SAB_LSA4 0.739 -0.538 0.254 -0.45 SAB_LSA5 0.829 -0.554 0.367 -0.536 SAB_LSA7 0.628 -0.331 0.196 -0.332 SPTS_CS1 -0.337 0.609 -0.075 0.27 SPTS_CS2 -0.447 0.756 -0.221 0.32 SPTS_CS3 -0.418 0.749 -0.22 0.407 SPTS_CS4 -0.569 0.801 -0.186 0.465 SPTS_CS5 -0.448 0.727 -0.216 0.384 SPTS_LS1 -0.339 0.506 -0.083 0.271 SPTS_LS2 -0.553 0.765 -0.267 0.460 SPTS_LS3 -0.381 0.725 -0.153 0.213 SPTS_LS4 -0.498 0.847 -0.148 0.400 SPTS_LS6 -0.367 0.726 0.027 0.228 SPTS_LS9 -0.395 0.792 -0.288 0.349 SPTS_LS12 -0.383 0.750 -0.178 0.217 SSB_LAE1 -0.448 0.378 -0.283 0.726 SSB_LAE2 -0.353 0.225 -0.181 0.675 SSB_LAE3 -0.361 0.321 -0.244 0.707 SSB_LAE4 -0.365 0.361 -0.223 0.741 SSB_LAE6 -0.317 0.231 -0.099 0.715 SSB_LAE8 -0.461 0.439 -0.256 0.851 SSB_LAE9 -0.474 0.178 -0.239 0.771 SSB_LAE10 -0.458 0.353 -0.206 0.613 SSB_LBE2 -0.366 0.240 -0.173 0.749 SSB_LBE8 -0.484 0.348 -0.357 0.699 SSB_LBE10 -0.511 0.474 -0.205 0.556 SS_SAS2 0.261 -0.287 0.478 0.018 SS_SAS4 0.226 -0.134 0.830 -0.158 SS_SAS5 0.181 -0.129 0.814 -0.146 SS_SAS6 0.209 -0.153 0.767 -0.224 SS_SAS7 0.334 -0.193 0.801 -0.097 SS_SFDS6 0.356 -0.231 0.827 -0.409 SS_SFDS11 0.278 0.014 0.746 -0.279 SS_SIS1 0.229 -0.19 0.648 -0.237 SS_SIS2 0.106 -0.117 0.682 -0.208 SS_SIS3 0.231 -0.118 0.720 -0.244 SS_SLS1 0.440 -0.294 0.639 -0.282 SS_SLS2 0.368 -0.231 0.762 -0.364 SS_SLS9 0.340 -0.050 0.803 -0.291 Humanities and Social Sciences Letters, 2025, 13(1): 56-68 63 © 2025 Conscientia Beam. All Rights Reserved. Figure 1. Outer-loadings for all the retained items. The model demonstrates more considerable discriminant validity according to the Fornell-Larcker criterion when the square root of the AVE value for a particular construct is greater than its maximum correlation with any other construct (Hair et al., 2017). Table 3 illustrates the Fornell-Larcker criterion for the four constructs: SAB, SPTS, SS, and SSB. The square root of the AVE value for construct SPTS was 0.735 greater than the SAB (-0.596). The square root of construct SS's AVE yielded a value of 0.738 greater than its relationship with SAB (0.411) and SPTS (-0.237). Meanwhile, the square root of the AVE value for construct SSB (0.713) exceeded its highest correlations with SAB (-0.605), SPTS (0.467) and SS (-0.326). Table 3. Fornell-Larcker criterion for the constructs SAB, SPTS, SS and SSB. Construct SAB SPTS SS SSB SAB 0.726 SPTS -0.596 0.735 SS 0.411 -0.237 0.738 SSB -0.605 0.467 -0.326 0.713 The values below 0.9 show stronger discriminant validity than those above 0.9 from the perspective of the criteria Heterotrait-monotrait ratio. The study's findings indicate that all values fell under the threshold of 0.9 and the four constructs SAB, SS, SPTS and SSB in the path model were more distinct. The HTMT for SPTS to SAB was 0.634, SS to SAB was 0.403, SS to SPTS was 0.264, SSB to SAB was 0.655, SSB to SPTS was 0.486 and SSB to SS was 0.344. According to the above criteria tested in the research, the discriminant validity of the instruments has been established. Humanities and Social Sciences Letters, 2025, 13(1): 56-68 64 © 2025 Conscientia Beam. All Rights Reserved. 5. DISCUSSION The main objective of this study was to examine the internal consistency and reliability, convergent validity and discriminant validity of the four Likert scales by applying the PLS-SEM approach. The average undergraduate or junior college student who participated may complete the survey in five minutes. The number of items was reduced from 101 to 46 compared to the original four scales. There are metrics for multiple constructs in the validation process such as composite reliability, AVE value, Fornell-Larcker criterion and Heterotrait-monotrait ratio as well as measurements for specific items such as outer loading and cross-loading (Hair et al., 2017). The accuracy and precision of measurement findings may be increased by employing a variety of complex measuring standards. SSBS tested the levels of college students' self-efficacy beliefs in two dimensions: LAE and LBE. The results showed that CA and CR values for construct SSB were 0.898 and 0.907 respectively indicating strong internal consistency reliability for SSBS. Convergent validity-wise, the stated AVE value for SSB is 0.527 which is higher than the necessary threshold of 0.5. The square root of its AVE value during the discriminant validity analysis was 0.726. Finally, eleven objects were kept and the HTMT value ranged between 0.85 and 0.85. Therefore, the reliability and validity of SSBS were good. Liang and Hao (2005) calculated a CA value of 0.960 for the CCSPSS. The reflective measurement model SS's CA and CA values were 0.940 and 0.930, respectively as compared to this result, suggesting that the structure has a high level of internal consistency and reliability. The AVE value obtained during the validation of convergent validity was 0.545 exceeding the 0.5 threshold criterion. The square root of its AVE on the corresponding construct was revealed to be 0.738. 13 items (SS_SAS2, SS_SAS4, SS_SAS5, SS_SAS6, SS_SAS7, SS_SFDS11, SS_SFDS6, SS_SIS1, SS_SIS2, SS_SIS3, SS_SLS1, SS_SLS2 and SS_SLS9) in this structure were preserved because the loadings in the corresponding structures far exceeded the cross-loadings in other structures and the HTMT value of the structure was below 0.85. The student perceived teachers’ support behavior questionnaire was created by OuYang (2005) to examine three aspects of students' perceptions of teachers' support: LS, CS and ES. She reported an overall reliability of 0.870 and an internal consistency reliability of 0.860. Two dimensions (LS and CS) were retained in the current research. The findings further demonstrated the scale's solid internal consistency reliability, convergent validity and discriminant validity. The internal consistency reliability of the scale was determined by the reported values of CA and CR for construct SPTS which were 0.921 and 0.930, respectively. The square root of the AVE on the associated construct was 0.735 which was more significant than that on the other constructs. The AVE value for SPTS was 0.540. From the perspective of loadings for the items and HTMT ratio (0.634), 12 items of the scale were preserved, namely SPTS_CS1, SPTS_CS2, SPTS_CS3, SPTS_CS4, SPTS_CS5, SPTS_LS1, SPTS_LS12, SPTS_LS2, SPTS_LS3, SPTS_LS4, SPTS_LS6 and SPTS_LS9. It was significant to observe that the student perceived teachers' support scale which was used in the actual research, only had two dimensions: learning support and capacity support. The emotional support sub-dimension was removed from the instrument since it did not match the requirements. The undergraduate learning burnout scale underwent an internal consistency test by Lian et al. (2005). The outcome revealed a total alpha coefficient of 0.865. Cronbach's alpha for construct SAB was 0.898 in this study whereas the reported values for composite reliability and AVE were 0.907 and 0.527 respectively. As a result, it can be said that SABS has demonstrated high internal consistency, reliability and convergent validity. Ten items (SAB_EE1, SAB_EE3, SAB_IB4, SAB_IB7, SAB_LSA1, SAB_LSA2, SAB_LSA3, SAB_LSA4, SAB_LSA5 and SAB_LSA7) were retained after the discriminant validity for SABS was confirmed by the Fornell-Larcker criterion, cross-loading and HTMT ratio. Humanities and Social Sciences Letters, 2025, 13(1): 56-68 65 © 2025 Conscientia Beam. All Rights Reserved. 6. LIMITATIONS AND SUGGESTIONS The purpose of this research is to determine the device's accuracy and dependability. Certain limitations still persist even with the researchers' best attempts to optimize the study's accuracy and dependability. Indicators pertaining to the four variables—students' self-efficacy beliefs, stress, perceived teacher support and academic burnout were included study's actual investigation. Nonetheless, the participants' demographic data was excluded. Actually, college students' levels of self-efficacy beliefs, stress, perceived teachers' support and academic burnout might change regarding gender, grade, categories of colleges and universities and so forth (Abdullah et al., 2022; Mantooth, Usher, & Love, 2020; Omari, Moubtassime, & Ridouani, 2020; Satpathy et al., 2021; Wang, Guan, Li, Xing, & Rui, 2019). Therefore, multi-group analysis can be conducted to test whether the models are invariant by gender, age, etc. Another potential constraint could be the presence of data bias. The only source of data for this study was participant self-reports. Although the researchers used different methods to ensure the reliability of the collected data, data bias was still inevitable. It is important to broaden the methods of data collecting and enhance the types of data that are collected in order to better accomplish the goal of the study. Interviews, observations, tests and even using secondary data in educational research (Cohen, Manion, & Morrison, 2018) are all good methods. Participants provide more real, comprehensive and in-depth ideas and information when collecting data through interviews and observation as opposed to formal questionnaires. Therefore, using multiple methods during the data collection helps analyze the interrelationships between individual latent constructs in greater depth. 7. CONCLUSION One may conclude from the discussion that the four scales (SSBS, SSS, SPTSS and SABS) employed in the study showed strong discriminant validity, convergent validity and internal consistency reliability. The scales are also valid and reliable. The final calculation findings compensated for the limitations of the original scale in the reliability and validity validation process by offering numerical support for the new scale's validity and reliability. Hence, the scale can be used to measure the mediating effect of academic burnout in the relationship between college students' self-efficacy beliefs, stress and perceived teachers' support on academic performance. Meanwhile, the new scale can be adopted for teacher training, measuring students' related behaviors and preventing students' adverse psychological conditions. It meets the requirements for quantitative measurement of students' self-efficacy beliefs, stress, perceived teachers' support and academic burnout and contributes to higher education research. Funding: This study received no specific financial support. Institutional Review Board Statement: The Ethical Committee of the Shandong Institute of Commerce and Technology, China has granted approval for this study on 25 October 2022 (Ref. No. SICT/2022-RB10- 06). Transparency: The authors state that the manuscript is honest, truthful, and transparent, that no key aspects of the investigation have been omitted, and that any differences from the study as planned have been clarified. This study followed all writing ethics. Competing Interests: The authors declare that they have no competing interests. Authors’ Contributions: All authors contributed equally to the conception and design of the study. All authors have read and agreed to the published version of the manuscript. REFERENCES Abdullah, N. A., Shamsi, N. A., Jenatabadi, H. S., Ng, B.-K., & Mentri, K. A. C. (2022). Factors affecting undergraduates’ academic performance during COVID-19: Fear, stress and teacher-parents’ support. Sustainability, 14(13), 7694. https://doi.org/10.3390/su14137694 Acharya, L., Jin, L., & Collins, W. (2018). College life is stressful today–Emerging stressors and depressive symptoms in college students. Journal of American College Health, 66(7), 655-664. https://doi.org/10.1080/07448481.2018.1451869 https://doi.org/10.3390/su14137694 https://doi.org/10.1080/07448481.2018.1451869 Humanities and Social Sciences Letters, 2025, 13(1): 56-68 66 © 2025 Conscientia Beam. All Rights Reserved. Aftab, N., Shah, A. A., & Mehmood, R. (2012). Relationship of self efficacy and burnout among physicians. Academic Research International, 2(2), 539-548. Alduais, F., Samara, A. I., Al-Jalabneh, H. M., Alduais, A., Alfadda, H., & Alaudan, R. (2022). Examining perceived stress and coping strategies of university students during COVID-19: A cross-sectional study in Jordan. International Journal of Environmental Research and Public Health, 19(15), 9154. https://doi.org/10.3390/ijerph19159154 An, R., & Pei, Y. Y. (2017). The mediating role of college students' mental toughness between emotional intelligence and academic stress. Chinese Journal of School Health, 38(07), 1092-1095. https://doi.org/10.16835/j.cnki.1000- 9817.2017.07.040 Asikainen, H., Salmela-Aro, K., Parpala, A., & Katajavuori, N. (2020). Learning profiles and their relation to study-related burnout and academic achievement among university students. Learning and Individual Differences, 78, 101781. https://doi.org/10.1016/j.lindif.2019.101781 Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215. https://doi.org/10.1037/0033-295X.84.2.191 Basith, A., Syahputra, A., & Ichwanto, M. A. (2020). Academic self-efficacy as predictor of academic achievement. Indonesian Education Journal, 9(1), 163-170. https://doi.org/10.23887/jpi-undiksha.v9i1.24403 Brandisauskiene, A., Buksnyte-Marmiene, L., Cesnaviciene, J., Daugirdiene, A., Kemeryte-Ivanauskiene, E., & Nedzinskaite- Maciuniene, R. (2021). Connection between teacher support and student’s achievement: could growth mindset Be the moderator? Sustainability, 13(24), 13632. https://doi.org/10.3390/su132413632 Chahid, H., Ahami, P. A. O. T., Chigr, P. F., & Najimi, P. M. (2018). Burnout and school performance: A study among students in the region of Béni Mellal (Morocco). World Journal of Research and Review, 6(6), 262646. Chen, J.-Y., & Tu, C.-C. (2019). A study on teacher support and learning adaptation among six-year normal university Freshmen: The moderating effect of the hardiness. EURASIA Journal of Mathematics, Science and Technology Education, 15(12). https://doi.org/10.29333/ejmste/115847 Chen, J., & Zhou, L. H. (2021). A study on the relationship between academic self-efficacy, academic stress and learning burnout of Xinjiang minority college students in Mainland China. Journal of Higher Education, 7(30), 74-77. Chen, Q., & Ma, Y. (2022). The influence of teacher support on vocational college students’ information literacy: The mediating role of network perceived usefulness and information and communication technology self-efficacy. Frontiers in Psychology, 13, 1032791. https://doi.org/10.3389/fpsyg.2022.1032791 Cohen, L., Manion, L., & Morrison, K. (2018). Research methods in education (8th ed.). Routledge: New York. Cullen, F. T. (1994). Social support as an organizing concept for criminology: Presidential address to the academy of criminal justice sciences. Justice Quarterly, 11(4), 527-559. https://doi.org/10.1080/07418829400092421 Dash, G., & Paul, J. (2021). CB-SEM vs PLS-SEM methods for research in social sciences and technology forecasting. Technological Forecasting and Social Change, 173, 121092. https://doi.org/10.1016/j.techfore.2021.121092 Demirel, M. V., Türkel, A., & Aydin, I. S. (2020). Speaking self-efficacy beliefs of Turkish university students. Cypriot Journal of Educational Sciences, 15(3), 399-411. Deng, Y., Cherian, J., Khan, N. U. N., Kumari, K., Sial, M. S., Comite, U., . . . Popp, J. (2022). Family and academic stress and their impact on students' depression level and academic performance. Frontiers in Psychiatry, 13, 869337. https://doi.org/10.3389/fpsyt.2022.869337 Frazier, P., Gabriel, A., Merians, A., & Lust, K. (2019). Understanding stress as an impediment to academic performance. Journal of American College Health, 67(6), 562-570. https://doi.org/10.1080/07448481.2018.1499649 Ghasemy, M., Teeroovengadum, V., Becker, J.-M., & Ringle, C. M. (2020). This fast car can move faster: A review of PLS-SEM application in higher education research. Higher Education, 80(6), 1121-1152. https://doi.org/10.1007/s10734-020- 00534-1 Guo, W. B., & Su, M. (2021). Research on the influence of teachers' support on college students' learning engagement in e- learning space. Theory and Practice of Education, 41(30), 50-54. https://doi.org/10.3390/ijerph19159154 https://doi.org/10.16835/j.cnki.1000-9817.2017.07.040 https://doi.org/10.16835/j.cnki.1000-9817.2017.07.040 https://doi.org/10.1016/j.lindif.2019.101781 https://doi.org/10.1037/0033-295X.84.2.191 https://doi.org/10.23887/jpi-undiksha.v9i1.24403 https://doi.org/10.3390/su132413632 https://doi.org/10.29333/ejmste/115847 https://doi.org/10.3389/fpsyg.2022.1032791 https://doi.org/10.1080/07418829400092421 https://doi.org/10.1016/j.techfore.2021.121092 https://doi.org/10.3389/fpsyt.2022.869337 https://doi.org/10.1080/07448481.2018.1499649 https://doi.org/10.1007/s10734-020-00534-1 https://doi.org/10.1007/s10734-020-00534-1 Humanities and Social Sciences Letters, 2025, 13(1): 56-68 67 © 2025 Conscientia Beam. All Rights Reserved. Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2017). A primer on partial least squares structural equation modelling (PLS- SEM) (2nd ed.). United States of America: Sage Publication. Hair, J. F., Ringle, C. M., & Sarstedt, M. (2011). PLS-SEM: Indeed a silver bullet. Journal of Marketing Theory and Practice, 19(2), 139-152. https://doi.org/10.2753/MTP1069-6679190202 Hair, J. F., Risher, J. J., Sarstedt, M., & Ringle, C. M. (2019). When to use and how to report the results of PLS-SEM. European Business Review, 31(1), 2-24. https://doi.org/10.1108/ebr-11-2018-0203 Hamann, K., Pilotti, M. A., & Wilson, B. M. (2020). Students’ self-efficacy, causal attribution habits and test grades. Education Sciences, 10(9), 231. https://doi.org/10.3390/educsci10090231 Hao, L., Cui, Y., & Chiu, M. M. (2018). The relationship between teacher support and students' academic emotions: A meta- analysis. Frontiers in Psychology, 8, 2288. https://doi.org/10.3389/fpsyg.2017.02288 Henseler, J., Hubona, G., & Ray, P. A. (2016). Using PLS path modeling in new technology research: updated guidelines. Industrial Management & Data Systems, 116(1), 2-20. https://doi.org/10.1108/IMDS-09-2015-0382 Hwang, S. (2021). The mediating effects of self-efficacy and classroom stress on professional development and student-centered instruction. International Journal of Instruction, 14(1), 1-16. https://doi.org/10.29333/iji.2021.1411a Hyytinen, H., Tuononen, T., Nevgi, A., & Toom, A. (2022). The first-year students' motives for attending university studies and study-related burnout in relation to academic achievement. Learning and Individual differences, 97, 102165. https://doi.org/10.1016/j.lindif.2022.102165 Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer. Li, P., Yang, J., Zhou, Z., Zhao, Z., & Liu, T. (2022). The influence of college students' academic stressors on mental health during COVID-19: The mediating effect of social support, social well-being, and self-identity. Frontiers in Public Health, 10, 917581. https://doi.org/10.3389/fpubh.2022.917581 Lian, R., Yang, L. X., & Wu, L. H. (2005). Relationship between professional commitment and learning burnout of undergraduates and scales developing. Acta Psychologica Sinica, 37(05), 632-636. Liang, B. Y., & Hao, Z. H. (2005). Development of Chinese college student psychological stress scale. Studies of Psychology and Behavior, 3(02), 81-87. Liang, Y. S. (2000). Study on achievement goals, attribution styles and academic self-efficacy of collage students. Master, Central China Normal University. Liu, L., Zhou, H. H., Li, S. Y., Wang, Z., & Teng, B. Y. (2018). On relationship amoing colleges students' learing stress, psychological resiliency, and learning burnout. Journal of Nanjing University of Aeronaautics and Astronautics (Social Sciences), 20(01), 96-100. https://doi.org/10.16297/j.nuaass.201801019 Liu, Q., Du, X. J., & Lu, H. Y. (2022). Teacher support and learning engagement of EFL learners: The mediating role of self- efficacy and achievement goal orientation. Current Psychology, 42(4), 2619-2635. https://doi.org/10.1007/s12144-022- 04043-5 Ma, A. M., Qu, Z. W., Yan, J., & Fu, J. (2017). Mediating effect of coping efficacy on relationship between psychological stress and adjustment in college freshmen. Chinese Mental Health Journal, 31(12), 994-998. Madigan, D. J., & Curran, T. (2020). Does burnout affect academic achievement? A meta-analysis of over 100,000 students. Educational Psychology Review, 33(2), 387-405. https://doi.org/10.1007/s10648-020-09533-1 Mantooth, R., Usher, E. L., & Love, A. M. A. (2020). Changing classrooms bring new questions: environmental influences, self- efficacy, and academic achievement. Learning Environments Research, 24(3), 519-535. https://doi.org/10.1007/s10984- 020-09341-y Omari, O., Moubtassime, M., & Ridouani, D. (2020). Factors affecting students’ self-efficacy beliefs in moroccan higher education. Journal of Language and Education, 6(3), 108-124. https://doi.org/10.17323/jle.2020.9911 Othman, N., Ahmad, F., El Morr, C., & Ritvo, P. (2019). Perceived impact of contextual determinants on depression, anxiety and stress: A survey with university students. International Journal of Mental Health Systems, 13(1), 1-9. https://doi.org/10.1186/s13033-019-0275-x https://doi.org/10.2753/MTP1069-6679190202 https://doi.org/10.1108/ebr-11-2018-0203 https://doi.org/10.3390/educsci10090231 https://doi.org/10.3389/fpsyg.2017.02288 https://doi.org/10.1108/IMDS-09-2015-0382 https://doi.org/10.29333/iji.2021.1411a https://doi.org/10.1016/j.lindif.2022.102165 https://doi.org/10.3389/fpubh.2022.917581 https://doi.org/10.16297/j.nuaass.201801019 https://doi.org/10.1007/s12144-022-04043-5 https://doi.org/10.1007/s12144-022-04043-5 https://doi.org/10.1007/s10648-020-09533-1 https://doi.org/10.1007/s10984-020-09341-y https://doi.org/10.1007/s10984-020-09341-y https://doi.org/10.17323/jle.2020.9911 https://doi.org/10.1186/s13033-019-0275-x Humanities and Social Sciences Letters, 2025, 13(1): 56-68 68 © 2025 Conscientia Beam. All Rights Reserved. OuYang, D. (2005). Research on the relation among teachers' expectation, self- conception of students' academic achievement, students' perception of teacher's behavioral supporting and the Study achievement. Master, Guangxi Normal University. Raisanen, M., Postareff, L., & Lindblom-Ylanne, S. (2021). Students' experiences of study-related exhaustion, regulation of learning, peer learning and peer support during university studies. European Journal of Psychology of Education, 36(4), 1135-1157. https://doi.org/10.1007/s10212-020-00512-2 Saleh, D., Camart, N., & Romo, L. (2017). Predictors of stress in college students. Frontiers in Psychology, 8, 19. https://doi.org/10.3389/fpsyg.2017.00019 Satpathy, P., Siddiqui, N., Parida, D., & Sutar, R. (2021). Prevalence of stress, stressors, and coping strategies among medical undergraduate students in a medical college of Mumbai. Journal of Education and Health Promotion, 10(1), 1–6. Viviers, H. A., De Villiers, R. R., & van der Merwe, N. (2023). The impact of self-efficacy beliefs on first-year accounting students’ performance: A South African perspective. Accounting Education, 32(6), 646-669. https://doi.org/10.1080/09639284.2022.2089047 Wang, M., Guan, H., Li, Y., Xing, C., & Rui, B. (2019). Academic burnout and professional self-concept of nursing students: A cross-sectional study. Nurse Education Today, 77, 27-31. https://doi.org/10.1016/j.nedt.2019.03.004 Wu, Z. Y., Yu, D. D., An, Y. J., & Li, J. (2021). Associations between effort-reward imbalance of subject competition and learing burnout: A moderated mediation model Journa of Nanyang Institute of Technology, 13(05), 90-97. Xu, P. P., Luo, S. Z., Yu, J., Tian, M., & Zhu, J. L. (2021). Study on the correlation between learning self-efficacy and learning engagement of undergraduate nursing students education. Health Vocational Education, 39(24), 43-45. Yu, X. Y., Yin, M. Y., Zhao, Y. F., & Xin, S. F. (2020). A cross－temporal meta－analysis of changes in Chinese college students' learning burnout. Psychology: Techniques and Applications, 8(2), 74-83. https://doi.org/10.16842/j.cnki.issn2095- 5588.2020.02.002 Appendix 1. Part content of the Likert scales. Code Statement Students' self-efficacy beliefs scale SSB_LAE1 I believe in my ability to get good grades in my studies. SSB_LBE8 When I review for the exam, I can integrate the knowledge I have learned to review. Student stress scale SS_SAS2 Failed to enter the ideal universities SS_SLS9 Be criticized SS_SIS3 No bosom friend SS_SFDS11 Too difficult to find employments Student perceived teachers' support scale SPTS_LS9 When I answer the question, the teacher will smile at me. SPTS_CS2 My teacher often recommends me to take part in various activities or competitions. Student academic burnout scale SAB_EE1 I find the knowledge I have learned useless. SAB_IB7 I often doze off when I study. SAB_LSA7 I am qualified for this stage course. Views and opinions expressed in this article are the views and opinions of the author(s), Humanities and Social Sciences Letters shall not be responsible or answerable for any loss, damage or liability etc. caused in relation to/arising out of the use of the content. https://doi.org/10.1007/s10212-020-00512-2 https://doi.org/10.3389/fpsyg.2017.00019 https://doi.org/10.1080/09639284.2022.2089047 https://doi.org/10.1016/j.nedt.2019.03.004 https://doi.org/10.16842/j.cnki.issn2095-5588.2020.02.002 https://doi.org/10.16842/j.cnki.issn2095-5588.2020.02.002