Journal of Educational Measurement Mahasarakham University

Social Desirability Responding: Conceptual Dimensions, Impacts and Challenges in Measurement

Vinita Kaewkua — 2025-12-31

In research that measures affective domains using self-reports as the primary data collection inventory, self-reports are often found to be either under- or over-reported especially in some matters that are considered sensitive, such as taboo matters, revealing illegal behavior, or expressing antisocial attitudes, etc. The main problem with this inaccurate self-reporting is that social desirability has a serious impact on the validity and reliability of the data. The key components of social desirability responding include impression management and self-deception. Although several social desirability scales exist, such as the Marlowe-Crowne (MC) scale and the Balanced Inventory of Desirable Responding (BIDR), recent meta-research questions the reliability and validity of these scales. It was found that the social desirability score does not explicitly measure bias or specific characteristics, but rather combines both dimensions. Therefore, it is imperative that researchers be aware of this social desirability and employ appropriate techniques to minimize response errors and bias. This article aims to provide readers with knowledge and understanding about social desirability responding in the following topics: 1) Concepts about social desirability responding; 2) The impact of social desirability on data validity and analysis; 3) Empirical evidence of data distortion in various fields; 4) Tools and strategies for measuring and controlling social desirability; 5) The conceptual debate on form versus substance; 6) Researcher limitations in dealing with social desirability; 7) Recommendations for the use of social desirability scales and 8) Challenges in measuring social desirability.

An Analysis of Factors and Indicators of Research Competencies of Basic Education Teachers

Russanan Kaewta — 2025-12-31

This study aimed to analyze the factors and indicators of research competencies among basic education teachers. The sample consisted of 652 teachers from schools under the Office of the Basic Education Commission during the first semester of the 2024 academic year, selected through multi-stage random sampling. The research instrument was a 60-item questionnaire assessing teachers’ research competencies using a five-point Likert scale. Data were analyzed using descriptive statistics and confirmatory factor analysis. The results revealed that research competencies comprised six factors with 60 indicators: (1) fundamental knowledge and skills in research with 12 indicators, (2) knowledge of research methodology with 12 indicators, (3) research skills with 11 indicators, (4) research report writing skills with 8 indicators, (5) dissemination and utilization of research findings with 9 indicators, and (6) researcher attributes and ethics with 8 indicators. The proposed model demonstrated good fit with the empirical data and satisfactory construct validity, as indicated by = 3218.062, df = 1627, p < .001, RMSEA = 0.035, and CFI = 0.975.

Development of Human Capital Indicators for Pre-Service Teachers

Natpapha Nuchapat — 2025-12-31

This study aimed to develop and validate indicators of human capital for pre-service teachers by applying Nadler's concept of human capital, which emphasizes the holistic development of essential attributes for the teaching profession. The research employed exploratory sequential mixed-methods design, collecting qualitative data from 20 key informants and quantitative data from a sample of 400 pre-service teachers. The qualitative findings indicated that the human capital indicators for pre-service teachers comprised three main components: (1) Training, which included designing training curricula focused on essential competencies, integrating ethics, morality, and teacher spirit, and monitoring and evaluating the application of knowledge; (2) Education, which emphasized promoting self-development, integrating theoretical and practical knowledge, and fostering reflection and experience-sharing; and (3) Development, which involved comprehensive teacher competency development, nurturing teacher ethics and spirit, and designing participatory and sustainable development processes. The evaluation of the measurement model indicated a high degree of alignment with empirical data ( = 27.82, df = 24, p = 0.26, CFI = 1.00, GFI = 0.98, AGFI = 0.97, SRMR = 0.02, RMR = 0.02, RMSEA = 0.02, NFI = 0.99). The second-order confirmatory factor analysis found that all indicators were statistically significant at the .01 level, with factor loadings ranging from 0.67 to 0.74. This research highlights the importance of comprehensive and sustainable human capital development in the teaching profession.

The Development of a Multidimensional Forced-Choice Situational Judgment Test for Assessing Undergraduate Students’ Expectations Toward Blended Learning in the 21st Century

Nurseeta Phoesalae — 2025-12-31

This study aimed to (1) develop a multidimensional forced-choice situational judgment test to assess undergraduate students’ expectations toward 21st-century blended learning and (2) examine the quality of the instrument in terms of response balance, validity, and reliability. Validity evidence was gathered through content validity assessment, criterion-related validation, and construct validity using a multitrait-multimethod (MTMM) matrix. Reliability was assessed using internal consistency and test-retest methods. The instrument development involved four groups: three purposively selected samples, consisting of experts in educational measurement and evaluation, higher education instructors, and undergraduate students, along with one stratified random sample of 120 undergraduate students, with academic faculties serving as strata. The final version of the instrument comprised 25 items incorporating visual stimuli, representing five distinct scenarios, each measuring five dimensions of blended learning expectations. The response balance analysis revealed no initial signs of response bias. Correlations between scores obtained from the developed instrument and external measures provided supportive evidence of criterion-related validity. Construct validity, assessed using the MTMM matrix, demonstrated higher correlations between different traits measured by the same method, thereby supporting convergent validity. Internal consistency was found to be at an acceptable level, and test-retest reliability indicated a high average correlation coefficient across the full instrument.

Development and Validation of an English Error Identification Test for Assessing English Grammar Competence of Grade 11 students: An Application of Computerized Dynamic Assessment

Meta Zherdzhaii — 2025-12-31

This research aimed to develop and validate an English error identification test for assessing grammar competence of Grade 11 students by application of computerized dynamic assessment. The research sample consisted of 100 Grade 11 students in academic year 2024 from schools under the Secondary Educational Service Office Lopburi. The research instrument was an English error identification test created using the iSpring Suite MAX program. The test was two-tier multiple-choice questions. Level 1 comprised an error identification test with four options, while Level 2 consisted of an error correction test with three options. The psychometric properties were analyzed based on content validity, item difficulty, discrimination index, and reliability. Additionally, user satisfaction of the human-computer interface was assessed using descriptive statistics, including the mean and standard deviation.
The test content covered eight grammatical components, totaling 20 items, i.e., (1) Part of Speech, (2) Tense, (3) Article, (4) Word Order, (5) Comparison, (6) If-clause,
(7) Parallelism, and (8) Subject-Verb Agreement. The index of item-objective congruence (IOC) for Level 1 questions ranged from 0.80 to 1.00, for Level 2 questions from 0.60 to 1.00, and for the mediation from 0.60 to 1.00. The difficulty index ranged from 0.41 to 0.65. The discrimination index ranged from 0.53 to 0.85. The internal consistency reliability of the test was 0.91. User satisfaction with the human-computer interface revealed that the highest level of satisfaction was in the system capabilities aspect, particularly the system speed, with the highest score (M = 4.35, SD = 0.83).

Applying Polytomous Item Response Theory to the Development of an English Reading Literacy Item Bank

Pisit Pinitsakul — 2025-12-31

This study aimed to develop an English reading literacy item bank by applying Polytomous Item Response Theory (IRT) using the Graded Response Model (GRM). The sample consisted of 4,000 upper secondary school students in northeastern Thailand, selected through multi-stage random sampling. The instrument used was a situational English reading test comprising 438 items. All items demonstrated acceptable content validity based on expert ratings (4.00–5.00 on a five-point scale).The items were divided into 18 test forms, each containing 24 items, with 6 anchor items shared across forms. The scoring was based on an 8-level rubric. The tests were administered to the sample, and data were analyzed using the mirt package in R. A total of 339 items met the quality criteria. The overall item parameter estimates showed that the average discrimination () across the three levels was 1.295 with a standard deviation of 0.466. The average difficulty () across all thresholds was –0.643 with a standard deviation of 1.181. The item bank covers a wide range of ability levels, from low to high. The test items can assess learners from CEFR levels A1 to B1, with some items capable of measuring abilities higher than average ( > 2), reflecting the effectiveness of the item bank. The item bank was stored in an online database developed using PhpMyAdmin. This item bank will be used for further development into a Computerized Adaptive Testing (CAT) system.

Development of Electronic Portfolio for Assessing English Communicative Competences of Upper Secondary Students

Wannida Tongthai — 2025-12-31

The objectives of this research were: 1) To develop and examine the quality of the tasks and criteria for assessing the English communicative competences, and 2) To develop and examine the quality of electronic portfolio for assessing the English communicative competences. The participants were 30 eleventh-grade students from Wangthong phitthayakhom school in Phitsanulok province, selected through simple random sampling. The instruments included 1) English communicative competences assessment tasks in reading skill, listening skill, writing skill, and speaking skill, 2) criteria for the reading skill, listening skill, writing skill, and speaking skill tasks; 3) an electronic portfolio assessment form; and 4) a satisfaction questionnaire regarding the assessment of English communicative competences using the electronic portfolio. The data were analyzed using mean, standard deviation, the CVI index, and the ICC index. This research found that 1) the tasks had high content validity (CVI = 0.80–1.00), 2) the assessment rubrics had the qualities of content validity (CVI = 0.60–1.00) and inter-rater reliability (ICC = 0.810–0.961), and 3) The electronic portfolio developed for the assessment of English communicative competences was of good quality, with a high level of upper secondary students’ satisfaction.

A Learning Activity Model to Promote Mathematical Problem-Solving Skills with Active Learning Principles among Grade 9 Students

Thikhumporn Krittakom — 2025-12-31

This study employed a research and development (R&D) approach with the objectives to: (1) investigate problems and needs in developing learning activities to promote mathematical problem-solving skills among Grade 9 students; (2) develop an active learning–based learning activity model; (3) implement and examine the effectiveness of the developed model; and (4) assess students’ satisfaction with the model. The participants were Grade 9 students from a lower secondary school, selected through purposive sampling. Research instruments included a problems-and-needs questionnaire, lesson plans and an implementation manual, a mathematical problem-solving skills test, and a student satisfaction questionnaire. Data were analyzed using the 80/80 efficiency criterion, mean, standard deviation, and dependent-samples t-test. The results indicated that: (1) Students reported the highest level of learning problems and the highest level of need for the development of learning activities; (2) the TKP active learning model consisted of three stages: Trigger, Knowledge Pursuit & Problem Solving, and Precision, emphasizing questioning, hands-on practice, and inquiry-based learning; (3) the model achieved an effectiveness of 88.77/82.70, exceeding the established criterion; and (4) students reported the highest level of satisfaction with the learning activities.

A Multiple Intelligences–Based Administrative Model for Promoting Vocational Skills among Primary School Students

Sunya Srisuk — 2025-12-31

This study aimed to: (1) examine the context, problems, and needs related to learning management; (2) develop multiple intelligences–based administrative model integrating learning management to promote students’ vocational skills; and (3) implement and evaluate the developed model. A mixed methods approach was employed under a research and development (R&D) framework consisting of four phases: contextual analysis, model development and validation, implementation, and evaluation and refinement. The participants included 30 Grade 6 students, 10 teachers, and one school administrator. Research instruments comprised interviews, questionnaires, model evaluation forms, and achievement tests. Data were analyzed using content analysis, descriptive statistics, and t-tests. The findings indicated that the school possessed strong local resources and community support but lacked an instructional management model aligned with students’ multiple intelligences. The developed administrative model consisted of five components—background, principles, objectives, core components, and a six-step operational process. Following implementation, students’ learning achievement significantly improved at the .01 level, and their satisfaction was rated high. Teachers and administrators also perceived the model as feasible and practical, highlighting its potential for systematic and sustainable application in rural schools.

Development of Mathematical Connection Ability through the 4MAT Learning Model Integrated with GeoGebra on Systems of Linear Equations in Two Variables among Grade 9 Students

Woranitipat Worakaensai — 2025-12-31

This research aimed to: (1) develop lesson plans using the 4MAT instructional model integrated with GeoGebra; (2) compare students’ learning achievement and mathematical connection ability on the topic of systems of linear equations in two variables with predetermined criteria; and (3) examine students’ satisfaction with the 4MAT–GeoGebra instructional approach. The sample consisted of 26 Grade 9 students from Khuanchangwittayakarn School, selected through cluster random sampling. Research instruments included 4MAT–GeoGebra lesson plans, a learning achievement test, a mathematical connection ability test, and a satisfaction questionnaire. Data were analyzed using the mean, standard deviation, the E1/E2 efficiency index, and hypothesis testing via Hotelling’s T² test. The results indicated that the developed lesson plans achieved an efficiency of 82.11/76.67, exceeding the established criterion. Students’ mathematical connection ability was significantly higher than the criterion at the .05 level, and students reported the highest level of satisfaction with the developed instructional approach.

Multilevel Construct Validity of the Personnel Administration Subscale of the Internal Integrity and Transparency Assessment

Anu Jarernvongrayab — 2025-12-31

The purpose of this study was to provide construct validity evidence for item Q26 in the personnel administration subscale of the Internal Integrity and Transparency Assessment (IIT). The data were obtained from 3,040 internal respondents across 105 local government organizations in Phitsanulok and Phichit provinces. Ten items of the personnel administration subscale were analyzed, yielding a Cronbach’s alpha coefficient of 0.794. Multilevel confirmatory factor analysis (MCFA) was employed to examine the factor structure at the individual and organizational levels. The results indicated that item Q26 exhibited negative and low factor loadings across all analytical models, including the single-level model and both within-level and between-level models. In contrast, the remaining nine items demonstrated positive and substantial factor loadings. Furthermore, factor loadings in the between-level model were consistently higher than those in the within-level and single-level models. These findings suggest that including item Q26 with the other items may lead to negative distortion in the interpretation of organizational performance outcomes, indicating that Q26 may not be an appropriate indicator of the personnel administration construct.

A Mediation Model of Creative Self-Efficacy on Creativity and Openness to Intellect among Students in Singapore: A Secondary Data Analysis from PISA 2022

Tewanart Sosongchan — 2025-12-31

This study is a secondary analysis of data from the Programme for International Student Assessment (PISA), which was collected by the Organization for Economic Co-operation and Development (OECD). The study has three main objectives: to examine the model fit of the mediation model of creative self-efficacy on creative and openness to intellect; (2) to analyze the direct and indirect effects of factors influencing Creativity and Openness to Intellect; and (3) to explore the mediating effect of Creative Self-Efficacy on Creative and Openness to Intellect. The research sample consists of 6,606 students from 163 schools in Singapore. The instrument used was a 32-item Likert scale questionnaire, with a reliability range between .82 and .92. This study employed path analysis and Sobel test techniques to examine mediation effects. The results indicated that the research model fit the empirical data. The creative school and class environment had a positive influence on creative and openness to intellect, both directly ( = .11) and indirectly ( = .18), with statistical significance at the .05 level. Similarly, the creative peer and family environment also had a positive influence on creative and openness to intellect, both directly ( = .13) and indirectly ( = .27), with statistical significance at the .05 level. The Sobel test confirmed that creative self-efficacy played a mediating role between both types of environments and students’ creative and openness to intellect. The explained variances for creativity and openness to intellect accounted for 36 percent. The findings from this study are beneficial for informing the design or refinement of educational policies and learning activities that support students’ creativity and openness to new ideas, with an emphasis on the collaborative involvement of all relevant stakeholders.

A Meta-Analysis of Factors Influencing Workplace Happiness

Supawadee Hemung — 2025-12-31

This study aimed to: (1) analyze differences in mean correlation coefficients classified by research characteristics, and (2) analyze the pooled correlation matrix of factors influencing workplace happiness. The studies synthesized were correlational research articles published between 2015 and 2023, totaling 147 studies. Research characteristics and study quality were recorded using a research characteristic coding form and a research quality assessment form. Effect sizes were calculated using the method proposed by Hedges and Olkin (1985), yielding a total of 696 effect size estimates. Research characteristics were classified into three domains: publication, content, and methodology, comprising 20 variables in total. One-way analysis of variance (ANOVA) and Pearson correlation analysis were employed for data analysis. The results indicated that: (1) statistically significant differences in research characteristics were found in (a) publication-related dimensions, including the field of study and type of research; (b) content-related dimensions, including research objectives; and (c) methodological dimensions, including sampling method, sample size, number of research instruments used, type of research instruments, and type of data analysis. (2) All factors influencing workplace happiness were positively correlated with workplace happiness at the .01 significance level, including quality of work life, work environment, workplace relationships, organizational culture, and job characteristics.

Factors Affecting Thai Students' Self-Directed Learning Efficacy

Nattapat Preechakul — 2025-12-31

This study investigated the causal factors influencing Thai students’ self-directed learning efficacy during the COVID-19 pandemic, using the international database from PISA 2022. The objectives were: 1) to validate the self-directed learning efficacy model among Thai students, 2) to analyze the direct and indirect effects of factors influencing self-directed learning efficacy, and 3) to test the mediating effect of feelings about learning at home on self-directed learning efficacy. The sample consisted of 6,430 Thai students aged 15 years from 278 schools, using student questionnaire data from PISA 2022. The reliability of the instrument ranged from .80 to .92. Data were analyzed using path analysis, and mediating effects were tested through the Bootstrap method with 1,000 replications.
The results revealed that the proposed model demonstrated a good fit to the empirical data ((1, N = 6,430) = 0.002, p > .05, CFI = 1.000, TLI = 1.000, RMSEA = 0.000, SRMR = 0.000). School support and family support had direct effects on self-directed learning efficacy, with effect sizes of .269 and .096, respectively. In addition, both indirect effects were observed through students’ feelings toward learning at home, with effect sizes of .073 and .108, respectively. The bootstrap analysis confirmed that both indirect effects were statistically significant at the 95% confidence interval. These results highlight the crucial role of school and family support in fostering students’ self-directed learning efficacy and provide practical implications for educational policy and the development of creative learning practices both in schools and at home to enhance the overall quality of education.

Development of External Assessor Standards for Basic Education

Wichuda Kijtorntham — 2025-12-31

The purposes of this study were: (1) to develop and examine the quality of external assessor standard indicators for basic education, and (2) to test the developed external assessor standard indicators. This study employed an exploratory sequential mixed-methods design and was conducted in three phases. Phase 1 involved the synthesis and development of a draft of external assessor standards for basic education based on relevant documents, followed by an examination of logical validity through a focus group discussion with nine experts. Phase 2 focused on developing and validating an external assessor standard assessment form based on findings from Phase 1, with content validity and reliability analyses conducted. Phase 3 involved testing the developed standard indicators, in which the discrimination power of the indicators was examined using the known-group technique.
The participants consisted of 42 external assessors. The research instrument was an external assessor standard assessment form for basic education. Data were analyzed using the Shapiro–Wilk W test for normality and the Mann–Whitney U test to examine the discrimination power of the indicators. The results revealed that: (1) the assessment form demonstrated good content validity, with item–objective congruence (IOC) values ranging from .80 to 1.00, and high reliability (α = .97); and (2) the external assessor standards for basic education comprised two standards, five components, and 39 indicators. Standard 1 (Qualifications) included two components: background (five indicators) and disqualifying characteristics (four indicators). Standard 2 (Competency) consisted of three components: knowledge and skills (ten indicators), professional conduct (five indicators), and work performance (fifteen indicators). The external assessor standards were classified into three levels: expert, senior professional, and advisory.