Original Article

Strategic Planning for Inclusive Education: Exploring and Validating the Measurement Model for Augmented Reality Applications Among Special Needs Children in Malaysia

Please cite this article as follows: Abu Bakar N, Suhaimi Ramli M. Strategic planning for inclusive education: exploring and validating the measurement model for augmented reality applications among special needs children in Malaysia. J Educ Community Health. 2024; 11(2):156-162. doi:10.34172/jech.2873

Introduction

Children with special needs have unique requirements that must be addressed in a particular manner, necessitating specific attention and different demands compared to typical children (1,2). These children also require a specialized education curriculum to meet their individual learning needs (3,4). Implementing an inclusion strategy by integrating them into mainstream education within regular learning settings is one approach to ensure they can interact with their peers, feel included in society, and subsequently, develop their self-confidence and daily living skills (5-7). In Malaysia, approximately 105 785 special education students are registered in accredited special education schools and institutes, ranging from preschool to high school levels across both rural and urban areas (8). Early diagnosis and intervention are crucial to begin the immediate process of specialized education programs and strategies for these children (9).

Early intervention programs are specifically designed to assist and support children with special needs such as those with autism spectrum disorder (ASD) by providing them with therapy, knowledge, support, and encouragement (10,11). The Treatment and Education of Autistic and Related Communication Handicapped Children (TEACCH) approach is an educational intervention that helps ASD children to overcome their communication difficulties by considering a variety of elements such as different items, environments, and weather conditions, as well as employing activities to promote social behavior, cognitive skills, and psychological capabilities (12).

The TEACCH approach also assigns specialized workstations for children with special needs, such as those with ASD, in designated areas where they can complete tasks and activities assigned to them by their therapists, thereby creating an organized teaching and learning environment (12,13). Meanwhile, the Picture Exchange Communication System (PECS), an evidence-based, low-tech, portable, and easy-to-use tool, was developed for children with special needs who were either non-verbal or had delayed speech to help them develop communication skills, express their emotions, and convey their messages both at home or in the classroom (14). The PECS approach comprises six steps that assist children with ASD in initiating communication with their partners (15). Prior research has identified PECS as one of the most effective educational interventions for kids with ASD (16).

Numerous local studies have been conducted on topics such as instructional strategies, learning materials, information and communication technology (ICT), and teacher preparation (11,17). The use of augmented reality (AR) applications has been acknowledged as an effective tool for improving user experiences. Furthermore, most autistic students possess stronger visual memories than auditory ones, which enables them to focus and comprehend new information more effectively through visual learning or visual thinking. This also helps in better retention. Employing visual strategies in media can potentially contribute to the development of their communication skills (18).

In addition, children with special needs frequently face behavioral problems and struggle to learn fundamental life skills, leading to the development of a curriculum that emphasizes practical and real-life experiences (19). PECS and TEACCH programs are among the most successful intervention strategies for enhancing social communication, visual information processing, and inclusive educational support for children with special needs, particularly those with autism (16,20). Moreover, the PECS approach, widely used in special education centers and schools for children with special needs, is regarded as an efficient communication training tool, especially during the teaching and learning processes (21). TEACCH, on the other hand, is a structured intervention program designed especially for kids with autism. It addresses all characteristics of autism, as well as the unique challenges faced by each autistic child by using alternative communication techniques, environmental modifications, and systematic intervention (13).

To provide autistic children with better alternative interventions using the PECS technique, Shminan et al (16) developed a mobile-based learning program called AutiPECS for parents of children with ASD in Malaysia. This program helps parents of ASD children reduce their reliance on therapists and the need for costly treatments at autism centres. In contrast, Taryadi and Kurniawan (22) conducted research in Indonesia to design a new application specifically made for autistic children to investigate the potential of using the PECS approach within AR applications for learning, teaching, behavioral stimulation, and monitoring. Meanwhile, Amado et al (23) in Peru developed an AR mobile application aimed at enhancing cognitive skills in autistic children in both online and in-person classroom settings.

The TEACCH program has provided structured teaching methods as a recommendation that has been proven effective for children with autism, particularly when learning is visual and interactive, whether through conventional or digital methods (24,25). Researchers from various fields are now investigating the integration of PECS and TEACCH approaches into digital-based mediums such as computers and touchscreen mobile technology (26). Thus, the application of AR effectively contributes to the acquisition of these skills by providing a platform for individuals with special needs to increase their motivation and understanding of certain AR-related information (19).

Given the potential of PECS and TEACCH approaches in planning inclusive education, this study aimed to explore and validate a measurement model of AR applications integrated with PECS and TEACCH methods for children with special needs in Malaysia as part of strategic planning for inclusive education.

Materials and Methods

Study Design and Setting

This cross-sectional study was conducted from December 2021 to December 2022 on 300 respondents, including teachers, parents, and caretakers of special needs children, from four regions of Malaysia: the northern, east coast, southern, and central regions. The selected respondents represented a true population of 37 084 special needs children in Malaysia, aged between six and twelve years, enrolled in registered special needs schools and centers in 2021. To ensure the confidentiality of respondents’ personal information, all the respondents provided written consent prior to the questionnaire distribution. Before the field study, a pilot study involving 100 respondents was conducted, and the data were analyzed using exploratory factor analysis (EFA) as suggested by previous research (27).

Sampling

A total of 300 respondents were selected from among parents and teachers of special needs children based on the recommendations of Hair et al (28) to ensure the robustness and reliability of the statistical analysis using structural equation modeling, as well as to enhance the representativeness of the overall population (29). A sample size of 300 is also sufficient when the model contains seven or fewer constructs, with fewer than three measuring items for each construct (30). To ensure the true population representativeness, respondents were proportionately selected from four designated regions of Malaysia using a proportionate stratified random sampling procedure.

Instrument for Data Collection

In this quantitative research, a 5-point Likert scale survey questionnaire was adapted and modified from previous studies (12,31,32) to align with the research objectives. The first part of the survey gathered demographic information about the respondents. The second part contained item statements related to the construct of AR. The third part contained statements related to the construct of perceived ease of use, while the fourth part focused on the construct of perceived usefulness. Additionally, the fifth part of the survey contained questions regarding the intention to use the technology. The sixth and the seventh parts contained the items related to perceived efficacy and training, respectively. On this scale, a rating of ‘1’ indicates “strongly disagree”, and a rating of ‘5’ indicates “strongly agree”. To achieve criterion validity, face validity, and content validity for the constructs and item statements of the measurement model, an expert review process was conducted as part of the pretesting phase before the pilot study (33). During the pilot study, the questionnaire was distributed to 100 selected respondents, including teachers and parents of special needs children, to collect pilot data (34). Afterward, EFA was conducted on the pilot study data to explore and measure the dimensionality of the items for each construct and to remove those with low factor loadings from the measurement model of the AR application integrated with PECS and TEACCH methods for special needs children. The EFA was performed using IBM-SPSS version 26.0 to explore the measurement model. Out of 34 items, 33 exhibited high factor loadings during the EFA procedure, and these items were retained in the survey and arranged according to their respective constructs. In the field study, a series of self-administered questionnaires were distributed to the selected 300 respondents who were given sufficient time to complete the survey without interruption and without disclosing their identities as stated in the consent form. The pooled confirmatory factor analysis (CFA) was then conducted on the field study data using IBM-SPSS Amos version 24.0 to validate the measurement model of AR applications combined with PECS and TEACCH methods for special needs children in Malaysia as a strategic planning framework for inclusive education.

Results

Table 1 represents the results of Bartlett’s test of sphericity and Kaiser-Meyer-Olkin (KMO) Measure of Sampling Adequacy for all six constructs. The KMO and Bartlett’s test of sphericity were conducted to measure the adequacy and the suitability of data for factor analysis. A KMO value above 0.60 demonstrates that the sample is sufficient for the factor analysis (35). Meanwhile, Bartlett’s test of sphericity with a P-value of less than 0.001 indicates significant correlations among at least some of the variables, which is essential for conducting factor analysis (36).

Based on Table 1, the KMO values for all six constructs were greater than 0.6, indicating that the sample size was adequate to proceed with the data reduction procedure and further analysis (30).

Based on the EFA results in Table 2, 9 items of AR with PECS and TEACCH constructs were retained for having high factor loading above 0.60. One item was deleted due to the factor loading below 0.60. Meanwhile, all other constructs (i.e., perceived efficacy, training, perceived ease of use, perceived usefulness, and intention to use) were retained and not deleted as their items displayed high factor loading under a single component. The construct of AR with PECS and TEACCH approaches was divided into two new dimensions, while the other constructs remained as single components. After EFA, only 33 reliable items out of 34 were retained. Table 3 displays the internal reliability for AR with PECS and TEACCH approaches, perceived usefulness, perceived ease of use, intention to use constructs, perceived efficacy, and training constructs.

As observed in Table 3, Cronbach’s alpha values for all constructs were above 0.70, indicating high internal consistency and internal reliability of the constructs. The result of the EFA procedure demonstrated that the original 34 items of the AR with PECS and TEACCH methods across six constructs (perceived efficacy, training, perceived ease of use, perceived usefulness, and intention to use) were reduced to 33 high-loading items. The constructs were reliable since Cronbach’s Alpha values exceeded 0.7 for all constructs. Additionally, Bartlett’s test of sphericity indicated highly significant P values for all constructs (P< 0.001), and the KMO readings were greater than 0.7, exceeding the minimum threshold of 0.60. By eliminating non-reliable and low-factor loading items during the EFA procedure, the instrument’s validity was enhanced. The remaining 33 items with high factor loadings (> 0.60) for the six constructs were analyzed using CFA. To optimize efficiency, the pooled CFA approach was used, and the results are displayed in Figure 1.

Close

Figure 1.

Result of Pooled Confirmatory Factor Analysis

Figure 1.

Result of Pooled Confirmatory Factor Analysis

As presented in Figure 1, the fitness index values for probability value, root mean square error of approximation (RMSEA), comparative fit index (CFI), Tucker-Lewis index (TLI), and chi-square/degrees of freedom (df) have met the required thresholds, with respective values of 0.000, 0.070, 0.905, 0.894, and 2.864. Each item’s factor loading reached the necessary threshold, which is greater than 0.6. The items within each construct also demonstrated unidimensionality, with factor loadings surpassing the minimum threshold of 0.6, indicating a high level of validity.

Based on Table 4, the assessment of construct validity for all the latent constructs in the measurement model has met the requirement for construct validity.

As seen in Table 5, the average variance extraction (AVE) values for all constructs are above 0.45, and the composite reliability (CR) values exceed the minimum threshold of 0.60. Convergent validity for each construct was established through AVE calculations. Notably, all constructs within the measurement model exhibited convergence validity, surpassing the standard values of 0.5 for AVE and 0.45 for complex models. Before conducting the CFA, items with low factor loadings were excluded to avoid lowering the AVE (37). Therefore, it can be concluded that each construct in the measurement model achieved CR and convergent validity (30).

Discussion

In line with the objective of this study to explore and validate a measurement model of AR applications integrated with PECS and TEACCH methods among special needs children in Malaysia, the measurement model was developed based on the Technology Acceptance Model (TAM) by Davis et al (38). The development of the measurement model involved six constructs: AR integrated with PECS and TEACCH methods, perceived efficacy, training, perceived ease of use, perceived usefulness, and intention to use. These constructs were adapted from previous research (12,31,32). To ensure criterion validity, face validity, and content validity of the measurement model, an expert review process was conducted as part of the pretesting process prior to the pilot study (33).

The results of EFA on the pilot study data indicated that all 33 items across six constructs were reliable. Each construct had a Cronbach’s Alpha value above 0.7, Bartlett’s test of sphericity yielded highly significant P values (P< 0.001), and KMO values exceeded 0.7, surpassing the minimum threshold of 0.60. These findings are in line with previous research conducted by Georgiou et al (39). The results of EFA in the present study also confirm the validity and robustness of the item statements and constructs within the measurement model (34). Additionally, the results indicate that the extracted components for each construct, along with their corresponding items, are reliable (40). Meanwhile, the removal of unreliable items and those with low factor loadings during the EFA further enhanced the instrument’s validity (41).

The results of pooled CFA for the measurement model in this study revealed that the factor loadings for items AR7, AR10, and ITU3 fell below the minimum threshold of 0.6. Therefore, these three insignificant items were removed as they affected the measurement model and caused the constructs to have a poor fit. The removal of items with low factor loadings was consistent with the methodology in previous research by Georgiou et al (39). After removing the low factor loading items, the fitness index values for Probability Value, RMSEA, CFI, TLI, and Chi-square/df were 0.000, 0.071, 0.906, 0.895, and 2.926, respectively. These values achieved the required level as suggested by Awang et al (30). Additionally, the factor loading for each remaining item surpassed the required value, which is more than 0.6 (42).

Based on the results of CFA, the AVE and CR were also met for all constructs in the measurement model. The required threshold for AVE exceeded 0.50 (43), while the required value for CR exceeded 0.60 (44). Thus, it can be concluded that convergent validity and CR for the constructs of the measurement model concerning AR applications with PECS and TEACCH methods for children with special needs in Malaysia were satisfactorily accomplished. However, since this study was conducted across four regions of Peninsular Malaysia, its findings cannot be generalized to the entire population of Malaysian special needs children. Since the respondents of this cross-sectional study were teachers, parents, and caregivers of special needs children, the results might not accurately reflect the true experience of special needs children due to the subjective interpretations and expectations of these individuals.

Conclusion

Regarding construct validity, the measurement model for all latent constructs met the literature-recommended acceptance thresholds. The model satisfied the criteria for absolute fit, incremental fit, and parsimonious fit, encompassing the three categories essential for construct validity. Discriminant validity was also confirmed as the measurement model remained free from redundant items, with modification indices remaining below the threshold of 15.0. Furthermore, the instrument demonstrated satisfactory CR. Following the CFA procedure, the measurement model for latent constructs associated with AR applications, incorporating PECS and TEACCH methods for special needs children in Malaysia, retained 30 items across six constructs. Therefore, the measurement model was regarded as reliable and validated to be used in future strategic planning in inclusive education research for special needs children in Malaysia.

Acknowledgments

The authors acknowledge the financial and technical support for this Fundamental Research Grant Scheme project provided by the Ministry of Higher Education Malaysia (MOHE) under the grant number FRGS/1/2021/SS0/UNISZA/02/4.

Authors’ Contribution

Conceptualization: Norsuhaily Abu Bakar, Mohamad Suhaimi Ramli.

Data curation: Norsuhaily Abu Bakar, Mohamad Suhaimi Ramli.

Formal analysis: Mohamad Suhaimi Ramli.

Funding acquisition: Norsuhaily Abu Bakar.

Investigation: Norsuhaily Abu Bakar, Mohamad Suhaimi Ramli.

Methodology: Norsuhaily Abu Bakar, Mohamad Suhaimi Ramli.

Project administration: Norsuhaily Abu Bakar.

Resources: Norsuhaily Abu Bakar.

Software: Mohamad Suhaimi Ramli.

Supervision: Norsuhaily Abu Bakar.

Validation: Mohamad Suhaimi Ramli.

Visualization: Norsuhaily Abu Bakar.

Writing–original draft: Norsuhaily Abu Bakar, Mohamad Suhaimi Ramli.

Writing–review & editing: Mohamad Suhaimi Ramli.

Competing Interests

The authors declare that there is no conflict of interests.

Ethical Approval

This study was approved by the Human Ethics Research Committee (UHREC) of Universiti Sultan Zainal Abidin, Malaysia (Code: UniSZA/UHREC/2021/422). Informed consent was distributed to all participants, ensuring that they understood the purpose of the study, the procedures involved, and their right to withdraw at any time without penalty.

Funding

The authors acknowledge the financial and technical support for this Fundamental Research Grant Scheme project provided by Ministry of Higher Education Malaysia (MOHE) under the grant (FRGS/1/2021/SS0/UNISZA/02/4).

References

1. Abu Bakar N, Raihan NZ, Sulaiman IM, Murtini LA, Munir R, Farouk S. An analysis of teachers’ knowledge and training: implication of teaching autistic children. Ann Rom Soc Cell Biol 2021; 25(6):5314-25. [ Google Scholar]
2. Sulaiman IM, Malik M, Abu Bakar N, Hassan H, Mamat M, Hashim ZI. Modelling autism spectrum disorders: an optimization approach with application. J Hunan Univ Nat Sci 2021; 48(8):279-91. [ Google Scholar]
3. Bryant DP, Bryant BR, Smith DD. Teaching Students with Special Needs in Inclusive Classrooms. 2nd ed. SAGE Publications; 2019. p. 504.
4. Abdullah MF, Shuhaimi AA, Osman MM, Rabe NS. Factors influencing parents in selecting school for children with special education needs. Plan Malays 2018; 16(2):207-16. doi: 10.21837/pm.v16i6.475 [Crossref] [ Google Scholar]
5. Ab Wahab L, Ismail K, Lop NS, Hanid M. Accessibility and inclusivity of playgrounds for children with disabilities in Malaysia Plan Malays. 2022 Se p 29; 20(3):172-83. doi: 10.21837/pm.v20i22.1136 [Crossref] [ Google Scholar]
6. Dogan A, Bengisoy A. The Opinions of Teachers Working at Special Education Centers on Inclusive/Integration Education. Cypriot J Educ Sci 2017; 12(3):121-32. [ Google Scholar]
7. Uzunboylu H, Özcan D. Teaching methods used in special education: a content analysis study. Int J Cogn Res Sci Eng Educ 2019; 7(2):99-107. [ Google Scholar]
8. Ministry of Education. Data Pendidikan Khas 2022. Data Pendidik Khas 2022. Available from: https://www.moe.gov.my/en/muat-turun/pendidikankhas/buku-data-pendidikan-khas/5570-buku-data-pendidikan-khas-tahun-2022/file.
9. Abu Bakar N, Al-Smadi MS, Mohd Pauzi H. Stress and issues of mental health for parent with special needs children. Int J Adv Sci Technol 2020; 29(7s):471-8. [ Google Scholar]
10. Abu Bakar N, Ali Z, Saputra J, Jusoh Z, Al-Smadi MS, Solahudin MH. Improvement of teaching effectiveness for autism children. Int J Adv Soc Sci Humanit 2022; 1(4):179-86. doi: 10.56225/ijassh.v1i4.94 [Crossref] [ Google Scholar]
11. Ali Z, Abu Bakar N, Ahmad W, Saputra J. Evaluating the use of web-based games on students’ vocabulary retention. Int J Data Netw Sci 2022; 6(3):711-20. doi: 10.5267/j.ijdns.2022.4.001 [Crossref] [ Google Scholar]
12. Oliveira MS, Pereira CP, de Santana KC, Rossinholli KO. Autisdata: software to help the development of people with ASD based on TEACCH and PECS methodologies. In: Proceedings of the 11th International Conference on Computer Supported Education (CSEDU 2019). SciTePress - Science and Technology Publications; 2019. p. 331-8. doi: 10.5220/0007716303310338.
13. Panerai S, Ferrante L, Zingale M. Benefits of the Treatment and Education of Autistic and Communication Handicapped Children (TEACCH) programme as compared with a non‐specific approach. J Intellect Disabil Res 2002; 46(4):318-27. doi: 10.1046/j.1365-2788.2002.00388.x [Crossref] [ Google Scholar]
14. Bondy A, Frost L. A Picture’s Worth: PECS and Other Visual Communication Strategies in Autism. 2nd ed. Woodbine House; 2011. p. 155.
15. Ivy S, Robbins A, Kerr MG. Adapted Picture Exchange Communication System using tangible symbols for young learners with significant multiple disabilities. Augment Altern Commun 2020; 36(3):166-78. doi: 10.1080/07434618.2020.1826051 [Crossref] [ Google Scholar]
16. Shminan AS, Adzani RA, Sharif S, Lee NK. AutiPECS: mobile based learning of picture exchange communication intervention for caregivers of autistic children. In: 2017 International Conference on Computer and Drone Applications (IConDA). Kuching, Malaysia: IEEE; 2017. p. 49-54. doi: 10.1109/iconda.2017.8270398.
17. Abdul Nasir MN, Erman Efendi AN. Special education for children with disabilities in Malaysia: progress and obstacles. Geografia Malays J Soc Space 2016; 12(10):78-87. [ Google Scholar]
18. Taryadi Taryadi, Kurniawan I. The improvement of autism spectrum disorders on children communication ability with PECS method Multimedia Augmented Reality-Based. J Phys Conf Ser 2018; 947(1):012009. doi: 10.1088/1742-6596/947/1/012009 [Crossref] [ Google Scholar]
19. Cakir R, Korkmaz O. The effectiveness of augmented reality environments on individuals with special education needs. Educ Inf Technol 2019; 24(2):1631-59. doi: 10.1007/s10639-018-9848-6 [Crossref] [ Google Scholar]
20. Shminan AS, Choi LJ, Sharif S. AutiTEACCH: mobile-based learning in a structured teaching approach for autistic children caregivers. In: 2020 International Conference on Informatics, Multimedia, Cyber and Information System (ICIMCIS). Jakarta, Indonesia: IEEE; 2020. p. 259-64. doi: 10.1109/icimcis51567.2020.9354288.
21. Flippin M, Reszka S, Watson LR. Effectiveness of the Picture Exchange Communication System (PECS) on communication and speech for children with autism spectrum disorders: a meta-analysis. Am J Speech Lang Pathol 2010; 19(2):178-95. doi: 10.1044/1058-0360(2010/09-0022) [Crossref] [ Google Scholar]
22. Taryadi T. Improved communication skills of children with autism spectrum disorder using augmented reality based on PECS (Picture Exchange Communication System). JAICT (Journal of Applied Information and Communication Technologies) 2018; 3(2):1-8. [ Google Scholar]
23. Amado ML, Ruiz LC, Andrade-Arenas L. Prototype of an augmented reality application for cognitive improvement in children with autism using the DesingScrum methodology. Adv Sci Technol Eng Syst 2021; 6(1):587-96. doi: 10.25046/aj060163 [Crossref] [ Google Scholar]
24. Abu Bakar N, Baijuri IZ. Understanding special education teachers’ perspectives on professionalism in educating autistic students. Int J Adv Sci Technol 2020; 29(1):34-41. [ Google Scholar]
25. Rao SM, Gagie B. Learning through seeing and doing: visual supports for children with autism. Teach Except Child 2006; 38(6):26-33. doi: 10.1177/004005990603800604 [Crossref] [ Google Scholar]
26. Kamaruzaman MF, Rani NM, Nor HM, Azahari MH. Developing user interface design application for children with autism. Procedia Soc Behav Sci 2016; 217:887-94. doi: 10.1016/j.sbspro.2016.02.022 [Crossref] [ Google Scholar]
27. Alkhawaja MI, Sobihah M, Awang Z. Exploring and developing an instrument for measuring system quality construct in the context of e-learning. Int J Acad Res Bus Soc Sci 2020; 10(11):403-13. doi: 10.6007/IJARBSS/v10-i11/7953 [Crossref] [ Google Scholar]
28. Hair JF, Hult GT, Ringle CM, Sarstedt M. A Primer on Partial Least Squares Structural Equation Modeling. SAGE Publications; 2014.
29. Hair JF, Anderson RE, Babin BJ, Black WC. Multivariate Data Analysis: A Global Perspective. Vol 7. Upper Saddle River, NJ: Pearson; 2010.
30. Awang Z, Afthanorhan WM, Lim SH, Zainudin NF. SEM Made Simple 2.0 A Gentle Approach of Structural Equation Modelling. Gong Badak: Penerbit UniSZA; 2023. p. 185.
31. Pasalidou C, Fachantidis N. Teachers’ perceptions towards the use of mobile augmented reality. In: Auer ME, Tsiatsos T, eds. Internet of Things, Infrastructures and Mobile Applications. Vol 1192. Cham: Springer; 2021. p. 1039-50. doi: 10.1007/978-3-030-49932-7_97.
32. Rakap S, Balikci S, Kalkan S. Teachers’ knowledge about autism spectrum disorder: the case of Turkey. Turk J Educ 2018; 7(4):169-85. doi: 10.19128/turje.388398 [Crossref] [ Google Scholar]
33. Ikart EM. Survey questionnaire survey pretesting method: an evaluation of survey questionnaire via expert reviews technique. Asian J Soc Sci Stud 2019; 4(2):1-17. doi: 10.20849/ajsss.v4i2.565 [Crossref] [ Google Scholar]
34. Baistaman J, Awang Z, Afthanorhan A, Zulkifli Abdul Rahim M. Developing and validating the measurement model for financial literacy construct using confirmatory factor analysis. Humanit Soc Sci Rev 2020; 8(2):413-22. doi: 10.18510/hssr.2020.8247 [Crossref] [ Google Scholar]
35. Hoque AS, Siddiqui BA, Awang ZB, Baharu SM. Exploratory factor analysis of Entrepreneurial orientation in the context of Bangladeshi Small and Medium Enterprises (SMEs). Eur J Manag Mark Stud 2018; 3(2):81-94. doi: 10.46827/ejmms.v0i0.384 [Crossref] [ Google Scholar]
36. Anuar N, Muhammad AM, Awang Z. An exploratory factor analysis of elicited students’ salient beliefs toward critical reading. Int J Mod Lang Appl Linguist 2020; 4(4):101-14. [ Google Scholar]
37. Awang Z, Afthanorhan A, Mohamad M, Asri MA. An evaluation of measurement model for medical tourism research: the confirmatory factor analysis approach. Int J Tour Policy 2015; 6(1):29-45. doi: 10.1504/ijtp.2015.075141 [Crossref] [ Google Scholar]
38. Davis FD, Bagozzi RP, Warshaw PR. User acceptance of computer technology: a comparison of two theoretical models. Manage Sci 1989; 35(8):982-1003. doi: 10.1287/mnsc.35.8.982 [Crossref] [ Google Scholar]
39. Georgiou Y, Kyza EA. The development and validation of the ARI questionnaire: an instrument for measuring immersion in location-based augmented reality settings. Int J Hum Comput Stud 2017; 98:24-37. doi: 10.1016/j.ijhcs.2016.09.014 [Crossref] [ Google Scholar]
40. Cheah JH, Sarstedt M, Ringle CM, Ramayah T, Ting H. Convergent validity assessment of formatively measured constructs in PLS-SEM. Int J Contemp Hosp Manag 2018; 30(11):3192-210. doi: 10.1108/ijchm-10-2017-0649 [Crossref] [ Google Scholar]
41. Acar Güvendir M, Özer Özkan Y. Item removal strategies conducted in exploratory factor analysis: a comparative study. Int J Assess Tool Educ 2022; 9(1):165-80. doi: 10.21449/ijate.827950 [Crossref] [ Google Scholar]
42. Haron Shafiee HI, Halim MS, Ismail M. Assessment model for determinant factor constructs in edu-tourism using confirmatory factor analysis (CFA). Int J Sustain Dev Plan 2023; 18(7):2037-43. doi: 10.18280/ijsdp.180705 [Crossref] [ Google Scholar]
43. Ahmad S, Zulkurnain N, Khairushalimi F. Assessing the validity and reliability of a measurement model in structural equation modeling (SEM). Br J Math Comput Sci 2016; 15(3):1-8. doi: 10.9734/bjmcs/2016/25183 [Crossref] [ Google Scholar]
44. Ab Hamid MR, Sami W, Mohmad Sidek MH. Discriminant validity assessment: use of Fornell & Larcker criterion versus HTMT criterion. J Phys Conf Ser 2017; 890(1):012163. doi: 10.1088/1742-6596/890/1/012163 [Crossref] [ Google Scholar]