Journal of Educational Measurement Mahasarakham University

Artificial Neural Network Applications in Selecting the Next Item for Multidimensional Computer-Adaptive Testing

Chaimongkol Pinasa — 2024-06-29

Selection for the next item using an artificial neural network has five important steps: 1. Dataset; 2. Data Preparation, comprising 2 methods—2.1 Formatting, followed by Data Transformation, and 2.2 Data Cleaning; 3. Modeling with Algorithms—Neural Net, Data mining by dividing the dataset into two parts: Train set—80%; Test set—20%. Adjust the parameter of the Neural Network to obtain the highest efficiency in the Training Cycle, Learning rate, and Momentum; 4 Test the performance of the model with the 10-fold cross-validation method and evaluate the performance of the model (Evaluation) with accuracy, precision, recall, and f-measure; and 5. Deployment Stage by developing the model in the form of a web application MCAT, using PHP language, scripting various commands, using the Prediction API (Application Programming Interface) to connect the algorithms Neural Net with the MCAT application in the item selection algorithm stage, and using a MySQL database for storing each attribute on cloud hosting.

The Guidelines for Utilizing the Assessment of Executive Function Skill in Research

Warabud Suppalarkbunlue — 2024-06-29

This article aims to review previous research related to the assessment of executive function (EF). EF skill is controlling emotions, thoughts, and behaviors so they are expressed appropriately in the situation one is facing, to achieve the goals (goal-directed behavior). EF continuously changes throughout one’s life and it is important for human development. Therefore, in assessing the EF skill, it is necessary to choose methods that are suitable for research objectives, elements of EF to be assessed, and the age of the sample, so that the assessment results can indicate the level of EF skill as close to reality as possible. This article describes 3 types of assessment (i.e., behavioral rating scale, performance-based assessment, and brain imaging). Each type reveals information from different contexts i.e., normal situations, challenging situations under standard conditions, and related areas of brain function, respectively. In addition, this article also presents the analysis of the strengths and limitations of each type, guidelines for selecting assessment methods, and examples of research that assessed EF skills in Thailand, so that researchers can choose an appropriate method for their research design.

The Road to Success in Microcredit Online Course for International Students at the Faculty of Physical Therapy, Mahidol University

Areerat Ngamkhum — 2024-06-29

Currently, there has been a notable transformation in pedagogical approaches, transitioning from conventional classroom settings to online modalities. This paradigm shift is a result of the COVID-19 pandemic situation. The recalibration of teaching and learning methodologies has been widely adopted across nearly all faculties within Mahidol University, aligning with the institution's contemporaneous operational directives. Faculty of Physical Therapy at Mahidol University has pioneered the development of Microcredit Online Courses tailored to accommodate the needs of international students. These courses are endorsed by the Faculty of Graduate Studies, Mahidol University. The deployment of these credit-accruing online modules has yielded positive responses and has been sustained by continuous funding support from the university's Department of International Relations over two years. Through surveys, it was found that only a few faculties had undertaken teaching initiatives to accumulate credits for international students and concretely succeeded as in the Faculty of Physical Therapy. Therefore, the faculty members involved have diligently formulated practical guidelines, procedural frameworks, and determinants conducive to the efficacious management of credit-accruing online courses designed for international physical therapy students at Mahidol University. These roadmaps serve as invaluable tools for the systematic analysis of benefits, the comprehensive assessment of outcomes, the evaluation of cost-effectiveness in operational endeavors, and the appraisal of risks inherent in teaching management, and are poised to inform graduate-level educational practices across all faculties in subsequent academic endeavors.

Research Synthesis on Learning Management with Contemplative Education: A Meta-Analysis

Kankanit Tanatpornpong — 2024-06-29

The research objectives were 1) to analyze the characteristics of research that used contemplative education in learning management and 2) to analyze the differences in the effect size of research, classified by characteristic variables of research that used contemplative education in learning management. The population consisted of researches on the learning process based on contemplative education, published from 2015 to 2022, totaling 304 titles. The sample consisted of 33 titles, purposively sampled. The tools used in the research included a research data extraction form and a research quality assessment form. The data analysis employed percentage, mean, standard deviation, effect size, a statistical test for the difference between two group means, and a statistical test for the difference between the means of over two groups.

The research results found that most of the researches that were published during 2015-2016 were experimental research. The results of the research quality assessment were very good. The results of the analysis of variance on the effect size revealed that research characteristics that made the effect size different at the statistical significance level of .05 were 1) basic characteristics of the research: year-of-publication variable and research-type variable; 2) the content of the research: theoretical-concept variable, and research-objective variable. 3) the research methodology: hypothesis-type variable, independent-variable-type variable, dependent-variable-type variable, sample-type variable, sample-selection-process variable, research-instrument variable, checking-the-quality-of-research-tools variable, research-design variable, basic-statistic variable, and statistic-for-hypothesis-testing variable.

The Model of Human Resource Management for Coping with Teacher Burnout in the Schools under Mukdahan Primary Educational Service Area Office

Jenjira Huaisai — 2024-06-29

This research is mixed methods research with a multi-phase design. The purposes of this research were to study the components and indicators of human resource management for coping with teacher burnout, to verify the congruence between the model for measuring human resource management for coping with teacher burnout components in the schools under Mukdahan Primary Educational Service Area Office with the empirical data, and to propose the model of human resource management for coping with teacher burnout components in the schools under Mukdahan Primary Educational Service Area Office. The sample consisted of 280 school administrators and teachers from the schools under Mukdahan Primary Educational Service Area Office, obtained through multi-stage sampling. The tool used for data collection was a 5-level rating scale questionnaire with content validity between 0.80 and 1.00; and a reliability coefficient of 0.98. Descriptive and inferential statistics were utilized with the SPSS program. Confirmatory Factor Analysis (CFA) was utilized with the M-plus program. Then, the feasibility, propriety, and utility of the model were evaluated.
The research findings were as follows:
1. Human resource management for coping with teacher burnout consisted of 4 components and 10 indicators, namely 1) competency-based recruitment and selection, comprising 2 indicators, 2) collaborative professional development, comprising 3 indicators, 3) participatory performance appraisal, comprising 2 components, and 4) teacher retention for well-being, comprising 3 indicators.
2. The model for measuring the indicators of human resource management for coping with teacher burnout in the schools under Mukdahan Primary Educational Service Area Office was congruent with the empirical data, according to the statistical values of = 40.340, df = 28, / df = 1.441, P-Value = 0.0616, RMSEA = 0.040, SRMR = 0.021, CFI = 0.994, TLI = 0.990.
3. The model of human resource management for coping with teacher burnout in the schools under Mukdahan Primary Educational Service Area Office consisted of the title, principles and concepts, objectives, achievement goals, and mechanisms for success. The results of the model evaluation revealed that the feasibility, propriety, and utility of the model were at the highest level. ( = 4.67, SD = 0.44, = 4.53, SD = 0.49, = 4.74, SD = 0.39, respectively).

Effect of distributed leadership on teaching quality with teacher autonomy and professional collaboration as the mediators: Evidence from Singapore’s 2018 TALIS data

Chalermwut Khrongsakun — 2024-06-29

This research was a study on the data in the Teaching and Learning International Survey 2018 (TALIS), collected by the Organization for Economic Co-operation and Development (OECD). The research had 3 main objectives. Firstly, to validate a proposed causal relationship model of distributed leadership. Secondly, to analyze the direct and indirect effects of distributed leadership on teaching quality. Finally, to test the mediating effects of teacher autonomy and professional collaboration on the relationship between distributed leadership and teaching quality. The sample consisted of 3,049 teachers from 169 schools in Singapore, which were drawn through stratified sampling. The research instrument was a 15-item Likert scale questionnaire, demonstrating a reliability range of .54 to .85. The research employed structural equation modeling (SEM) and bootstrapping techniques to test mediation. The result showed that the causal relationship model fit the empirical data. Distributed leadership had both direct ( = .11) and indirect ( = .13) effects on teaching quality, with statistical significance at the .05 level. The variables in the model together explained 15 percent of the variance in teaching quality. The mediation testing with bootstrapping (95% CI) showed that professional collaboration and teacher autonomy mediated the relationship between distributed leadership and teaching quality with an effect size of .08 and .05, respectively. The findings provide insight into the shared goal and shared decision-making processes in teacher development through distributed leadership in Singapore that may apply to the Thai education system.

A Learning Management Model to Enhance Reasoning Ability and Desirable Characteristics for a Course in Labor Law for Vocational Students

Boonlert Phokham — 2024-06-29

This research and development aimed to 1) develop a learning management model to enhance reasoning and desirable characteristics in learning labor law of vocational students and 2) to evaluate the developed learning model on using reasoning and desirable characteristics in studying labor law by applying the concepts Self-knowledge creation, reasoning, problem-based learning, and social learning theory. This research was action in the Faculty of Industrial Technology, Nakhon Phanom University and research process included phase 1 was developed model by interviewing 3 teachers, 9 students, also observing 3 labor law classes to assess problems and needs to develop a learning management model and drafting a learning management model according to the prior concepts and information, also quality assessed by 5 experts. Phase 2: Experimenting a model to assess reasoning abilities and desirable qualities in studying labor law subjects. A sample was 13 people enrolled by purposively selected. Research tools included the model outlines, learning plans, and assessments of reasoning and desirable characteristics forms. Data analyzing statistics were used including percentage, mean, and standard deviation, comparing increasingly mean and percentage of learning outcome scores and content analysis.

Research results: The learning management model components consisted of 1. concepts and principle basic theories 2. Objectives 3. Learning management processes 4. Social factors and responses 5. Supporting factors and learning resources 6. Quality assessment and the draft format appropriateness is a high level, and the development this learning management models was applied concepts of self-knowledge creation, reasoning, problem-based learning, and social learning theory, also average evaluated score for 4.3. The draft learning management model trial on enhancing reasoning ability and desirable characteristics in labor law for vocational students was averages scores for 29.6 and 54.4, and representing accounting for 65.9 and 67.9 percent, respectively. The criteria of rubrics scoring in the reasoning ability assessment have been clarified for clearly assessment and more appropriate. This learning model can be applied to students in relevant situations to be more successful and efficient.

Development of Climate Change Awareness Measuring Instruments of Upper Secondary School Students

Piyatida Supa — 2024-06-29

Climate change is a major threat to the future of today’s youths. It is necessary to increase individual climate change awareness of young people to cope with climate change. Thus, effective measuring instruments are required to understand young people's awareness of climate change. This research aimed to create a measuring instrument to assess youths’ awareness of climate change and validate it by assessing three components: understanding, attitudes, and behavior. The sample was 300 upper secondary school students, obtained by multi-stage random sampling. The data were collected by an understanding of climate change test, which was a true-false item test, and a 5-point rating scale questionnaire inquiring about climate change attitudes and behavior. The reliability of the understanding test was 0.70 and the reliability of the attitudes and behavior questionnaire was 0.97, respectively. As a result of the study, the confirmatory factor analysis of climate change measuring instruments consisted of 3 components and 15 observed variables. The measuring instruments revealed construct validity and consistency between the model and empirical data. (Chi-square = 91.187, df = 72, Relative Chi-square = 1.266, p-value = 0.063 RMSEA = 0.030, RMR = 0.039, GFI = 0.961, NFI = 0.969, TLI = 0.990, CFI = 0.993).

The Rasisalai School Supervision Model to Promote the Competency of Student-Centered Learning Management

Roongsawat Maneewong — 2024-06-29

The purposes of this study were 1) to study baseline data of needs for a model of internal supervision to promote competency of student-centered learning management in Rasisalai School, 2) to develop an internal supervision model, 3) to try out the internal supervision model, and 4) to study the impact of using the Rasisalai School supervision model to promote student-centered learning management competency. The target group was 96 teachers in the academic year 2022. The instruments used were document analysis, interviews, an achievement test, a questionnaire, observations, and evaluations. Data analysis employed percentage (%), arithmetic means (), standard deviation (S.D), t-test for dependent samples, and content analysis.

The results were as follows: 1) the baseline data of needs were law, policy, concepts, and theories in education concerning student-centered learning management; urgent problems were low academic achievement, teachers lacking cognizance of student-centered learning and not promoted to seriously organize student-centered learning; 2) The Rasisalai School internal supervision model to promote student-centered Learning competencies was called the SUPER Model. It had 5 main steps as components. Step 1 was Set about; S—Preparation; Step 2 was Upgrade; U—Enhancement of experience; Step 3 was Process; P— Internal Supervision Process; Step 4 was Evaluation; E—Evaluation, and Step 5 was Reflection; R—Reflection. 3) The results of the tryout revealed that the teachers had the highest level of overall student-centered learning competency, and 4) The impacts of using the model were: that the supervisors and the supervisees were, on the whole, very satisfied with the Rasisalai School’s internal supervision model. The O-NET results and learning achievement in the academic year 2022 were higher than in 2021. This was because all elements of the model were appropriate and consistent with each other. It was a process that was continuous and inter-related. The supervisors and the supervisees understood their roles and had a sincere commitment to learning management development, under the guidance of experts and the support of the administrators.

The Effectiveness of Learning Management Model Enhancing Consciousness and Self-regulation Based on Metacognition Approach, Collaborative Learning and Social Learning Theory for Sixth Grade Students

Vorrapapha Tesprasit Tesprasit — 2024-06-29

The purposes of this research were 1) to compare students’ consciousness and self-regulation before studying the instructional model with those after studying the instructional model based on the metacognition approach, collaborative learning, and social learning theory, 2) to compare consciousness and self-regulation of students in the group that were taught with the learning management model and the those in the control group that were taught with normal practice. This research followed the Research and Development process composed of three phases: 1) studying the state of problems, needs, and the body of knowledge related to problem-solving, 2) developing the instructional model and trying it out, and 3) studying the effectiveness of the instructional model. In phase 3, the sample was 37 grade 6 students from Group 1 and 36 grade 6 students from Group 2 of Tessaban 4 School (Rattanakosin 200 Years), Muang Nakhon Phanom Municipality, Nakhon Phanom Province in the academic year 2022, selected by cluster random sampling, using the student group as the sampling unit. The research instruments used for data collection were 1) the consciousness measurement form, composed of 25 items, and 2) the self-regulation assessment form, composed of 20 items. The statistics used for data analysis were the mean, standard deviation, t-test for dependent samples, and One-way MANOVA.

The results were as follows:

1. The students in the experimental group, assigned to learn with the learning management model, had consciousness and self-regulation after learning higher than before learning, with a statistical significance level of .05.

2. The students in the experimental group, assigned to learn with the learning management model, had consciousness and self-regulation after learning higher than the students in the control group, with a statistical significance level of .05.

The Development of Learning Management to Promote Reading Literacy and Learning Achievement Using PISA Competency-based Approach for Grade 9 Students

Wanwisa pramual — 2024-06-30

The objectives of this research were to 1) assess the effectiveness of organizing learning activities, 2) assess reading literacy, and 3) study the achievement of grade 9 students, based on the PISA competency-based approach. The sample was 41 grade 9 students, from 1 group, of Mahasarakham University Demonstration School (Secondary) in the second semester of the academic year 2022, obtained by cluster random sampling, using the student group (7 groups of students having the same characteristics) as the sampling unit. The research tools were 1) a learning plan, 2) a reading skill exercise sheet, 3) a reading literacy assessment form following PISA competency-based, and 4) a learning achievement test. The analysis of data employed percentage, standard deviation, and t-tests for hypothesis testing.
The research findings were as follows:
1. The efficiency (E1/E2) of the learning management model based on the PISA competency-based approach for grade 9 students was 84.88/85.27, higher than the preset criterion (80/80).
2. The reading literacy of the grade 9 students taught by the PISA competency-based approach, following the PISA 2018 evaluation framework, was composed of 5 components: 1) accessing and retrieving information, 2) integration and inferences, 3) evaluating and reflecting, 4) using reading information, and 5) reading engagement. On the whole, most of the students had their reading literacy at level 3 (moderate), accounting for 73.17 percent, followed by level 4 (difficult and complicated level), accounting for 26.83 percent, and there were no students in level 1 (lower than the basic level), level 2 (basic level) and level 5 (high level).
3. The students had higher learning achievement after learning (an average score of 21.32), which was higher than before learning (an average score of 14.22), using a t-test for dependent samples (t = 21.03, P <0.0001) at the significant level of .05.

Presentation of the Guidelines for Improving Teaching and Learning for Students of Sukhothai Thammathirat Open University: Application of Information from Data Mining Analysis

Vinita Kaewkua — 2024-06-30

The objectives of this research were to 1) present guidelines for improving the teaching and learning of the courses in management science that were analyzed, and 2) develop a model for the characteristics of students in the dominant group and underprivileged group in management science. The sample was 233 Sukhothai Thammathirat Open University studying management science, political science, and law. The purposively sampled students had ID numbers ranging from 60 to 63. The research instruments were a questionnaire on factors related to education in the distance learning system and data on educational outcomes in subjects in management science that had the least number of students who passed the exam, and academic achievements in general education subjects. The research methodology was divided into 5 steps: 1) Data Collection: Information from the online questionnaire and students’ educational results; 2) Data Preparation: Selecting completed data to prepare for analysis; 3)Feature Selection:Specific attributes were selected for analysis; 4)Data Transformation: The data was converted into the CSV format; 5) Data Analysis: The data was analyzed using data mining techniques with the RapidMiner program, selecting the decision tree classification method.The research findings revealed that the decision tree data classification efficiency of the dominant group (Class S) of students gave an accuracy of 77.06% and the decision tree data classification efficiency of the underprivileged group (Class W) gave an accuracy of the model of 66.52%. From the research, 8 rules could be created as decision rules.

Evaluation of the Psychometric Properties of Sukhothai Thammathirat Open University’s English Proficiency Test

Sukolrat Ingchatcharoen — 2024-06-30

The objectives of this research were to 1)analyze the quality of Sukhothai Thammathirat Open University’s English proficiency test (STOU-EPT) using a computer system in three parameters: the discrimination parameter (a), the item difficulty parameter (b), and the guessing parameter (c) following Item Response Theory (IRT); and 2) test differential item functioning (DIF) in STOU-EPT. The sample was 1,272 persons who took STOU-EPT, using purposive sampling in compliance with the criteria of having secondary data, as well as data on gender, dwelling place, and experience in taking STOU-EPT. The research instrument was the 100-question STOU-EPT which consists of 25 listening skills questions, 35 grammar skills questions, and 40 reading skills questions. The data was examined using RStudio. The study's findings showed that:
1. The quality of STOU-EPT included: 1) for listening skills, the a-parameter ranged from - 9.62 to 4.5, with 22 items (88%) passing the criterion a > 0, the b-parameter ranged from -1.80 to 8.19, with 22 items (88%) passing the criterion -2 ≤ b ≤ +2, and the c-parameter ranged from 0 to 0.61; 2) regarding grammar skills, the a-parameter ranged from -3.40 to 7.65, with 31 items (88.57%) passing the criterion a > 0, the b-parameter ranged from -4.21 to 3.06, with 22 items (62.86%) passing the criterion -2 ≤ b ≤ +2, and the c-parameter ranged from 0 to 0.44; and 3) reading skills, the a-parameter ranged from -8.75. to 7.87, with 36 items (90%) passing the criterion a > 0, the b-parameter ranged from -2.02 to 23.75, with 35 items (87.50%) passing the criterion -2 ≤ b ≤ +2, and the c-parameter ranged from -0 to 0.63.
2. Testing the DIF in STOU-EPT using LR, SIBTEST, and MH, by comparing the number of items that found the DIF matched from two methods and over, revealed that 1) listening skills had the DIF between gender (female = focal), dwelling place (Bangkok and surrounding areas = focal), and experience in taking the exam (> 1 time = focal), with the same number of 3 items each (12%); 2) grammar skills had the DIF between gender, dwelling place, and experience in taking the exam, with 3, 5, and 1 items, respectively (8.57%, 14.29%, and 2.86%), and 3) reading skills had the DIF between gender, dwelling place, and experience in taking the exam, with 1, 3, and 0 items, respectively (2.50%, 7.50%, and 0%).

The Construction of Mathematical Connection Ability Test for Grade Four Students

Amonrat Sridatham — 2024-06-30

This research had the objectives to (1) construct a mathematical connection ability test for grade 4 students, (2) analyze the psychometric properties of the mathematical connection ability test for grade 4 students, and (3) analyze the mathematical connection ability of grade 4 students. The sample consisted of 100 grade 4 students at Ban Khlong Bua School (Iam Saengrot), Bang Khen District Office, under Bangkok Metropolitan Administration (BMA), using the technique of two-stage cluster sampling. The research instrument was a fifteen-item response test. The psychometric properties were analyzed from (1) content validity, (2) reliability, (3) difficulty, and (4) discrimination. The mathematical connection ability was analyzed using descriptive statistics of mean, frequency, and percentage.
The findings were as follows:
1. The construction of the mathematical connection ability test for grade 4 students yielded a fifteen-item response test with polytomous scoring. All items measured the concept of Connectivism by the National Council of Teachers of Mathematics of the United States of America (NCTM, 2000) in three standards: (1) remembering and making connections for mathematical concepts—five items, (2) understanding the connections of mathematical knowledge to see how they connect and using the knowledge to construct new knowledge in consistency with existing knowledge—five items, (3) remembering and applying mathematics in other contexts beside mathematics—five items.
2. The analysis of psychometric properties of the mathematical connection ability test of the students under study revealed that the content validity of all items passed the criteria. The reliability was 0.764, rather a high level, and passed the criterion. All the fifteen test items exhibited difficulty between 0.444 and 0.840. The discrimination was from -0.037 to 0.867.
3. The analysis of the mathematical connection ability of the 100 students under investigation found the following: The students had an average score of 32.25 from 52. When considering the assessment of mathematical connection ability, it was found that 73 students exhibited the mathematical connection ability passing the criterion with a score of 26 or higher than 50 percent of the maximum score (73.00 percent), meanwhile, 27 students exhibited the mathematical connection ability lower than the criterion with the score lower than 26 or lower than 50 percent of the maximum score (27.00 percent).

Readiness of Corporations According to Cooperative and Work-Integrated Education in the Perspectives of University Supervisors and Students: A Case Study of Prince of Songkla University, Phuket Campus

Kamonwan Sangtong — 2024-06-30

The Ministry of Higher Education, Science, Research and Innovation has a policy to promote collaboration between educational institutions and corporations in implementing cooperative and work-integrated education (CWIE) programs in order to produce graduates with high potential and competence, meeting the expectations of the corporation sector. The research aimed to: 1) study the readiness of corporations according to CWIE, and 2) compare the readiness in the perspectives of university supervisors and students. The study collected the readiness assessment results from corporations affiliated with Prince of Songkla University, Phuket Campus. The quality of the research tool was verified by experts. The research population consisted of 64 university supervisors and 164 students. Secondary data were analyzed statistically, utilizing frequency distributions and means. The findings indicated that the overall readiness of corporations in the context of CWIE, in the perspectives of both university supervisors and students, was at the highest level. The comparison of corporation readiness in the perspectives of university supervisors and students showed a unanimous opinion that corporations were highly prepared, especially in terms of task assignment and professional mentorship.

The Development of Assessment Criteria for Science Communication via Infographics for Secondary School Students

Pattamaporn Kaewkhongkha — 2024-06-30

The objectives of this research were to: 1) develop assessment criteria for science communication via infographics; and 2) study the results of trying out the developed assessment criteria by assessing the infographics produced by 18 Grade 10 students who participated in the study voluntarily. The results revealed that the content validity of the assessment criteria was validated by 3 science education experts. The assessment criteria developed covered 2 components: 1) the content and information, and 2) the design. The score level for each component was divided into 3 levels: 2, 1, and 0. In data analysis, the researchers separately used the criteria to evaluate the student infographics, then took the results, discussed them, and drew conclusions until all the scores matched (inter-rater reliability = 1). The percentage of each component score level was calculated. The results of trying out the assessment criteria for science communication via infographics showed that, regarding the components of content and information, most of the workpieces could not identify complete scientific content. Some scientific misconceptions were found. Regarding the design and the components of the design, it was found that most workpieces used inappropriate layouts. The font styles and illustrations were also inconsistent with the content and information being presented in the infographics.

Factors Predicting Career Choices among Fourth-Year Students in the Faculty of Humanities and Social Sciences, Burapha University: A Multilevel Multiple Regression Analysis

Pattrawadee Makmee — 2024-06-30

The purposes of this research were to study multilevel predictive factors and to create multilevel predictive equations affecting factors predicting career choices of fourth-year students at the Faculty of Humanities and Social Sciences, Burapha University. The sample consisted of 420 fourth-year undergraduate students from 14 courses by simple random sampling. A 5-point rating scale questionnaire was utilized, with content validity ranging from .67 to 1.00, item discrimination from .20 to .65, and reliability of .91. The statistical method employed was multilevel multiple regression analysis using the Mplus 8.10 program for analysis.

The results indicated that, at the student level, factors predicting career choices, which accounted for 53.60% of the variance, included average monthly family income (.05) with a statistical significance level of .05, while career type (.16) and motivation (.72) were significant at the .01 level, respectively. At the course level, which could predict 89.20% of the variance, factors included the type of career (-.43) with a statistical significance level of .05, and motivation (1.00) significant at the .01 level, respectively.

Meta-Analysis of the Effective Learning Activities Model On Students' Problem-Solving Thinking: Propensity Score Matching

Supansa Surin — 2024-06-30

The research aimed to: 1) investigate the effect sizes and research characteristics that influence students' problem-solving thinking, and 2) compare the effect sizes of the learning activity model to develop students' problem-solving thinking after adjusting the propensity score matching. Thirty-five graduate-level research papers published from 2002 to 2022 were studied. Data were gathered, using a form for recording research characteristics and an assessment form for assessing research quality. Effect sizes were calculated, using Glass's method, and the data were analyzed, using random effects, fixed effects, meta-regression, and propensity score matching techniques. The research findings revealed that:

1) Researches focusing on learning management models significantly influenced students' problem-solving abilities at a high level ( = 1.395). The research characteristics with 4 variables—the university producing the research, the field, the total duration, and the quality—significantly influenced students' problem-solving thinking at a statistically significant level of .05; and 2) after adjusting the propensity score matching, it was found that inquiry-based learning had the greatest impact on students' problem-solving thinking. Designing learning activities that allow students to explore problems, create understanding, expand their thinking through collaborative discussion, and assess outcomes according to the appropriateness of each context will improve students' problem-solving thinking.

The Validation of the Construct Validity of the Core Competency Measurement Model for Third Grade Students

Treekom Prommaboon — 2024-06-30

The purpose of this research was to create and develop a core competency test for grade 3 students, and to validate the structural validity of the model for measuring the core competencies of grade 3 students. A sample of 370 grade 3 students in Surin Province was drawn from multi-stage random sampling. The research tools were four sets of core competency tests for grade 3 students developed by the researchers. Data were analyzed, using the mean and standard deviation to verify the prediction accuracy from the receiver operating characteristic (ROC) graph, and the area under the ROC curve was obtained. The validation of the structural validity of the model for measuring core competencies was done through second-order confirmatory factor analysis. The research results revealed that:
1. The results of the creation and development of the core competency test showed that there are 4 competencies: (1) the literacy competency test had 44 items; (2) the ability core competency had 29 items, (3) the well-being core competency had 26 items, and (4) the core competency for social concern had 26 items. The items had a discrimination from 0.20 to 0.69, and the test had a reliability from 0.85 to 0.91. In the results of the analysis of the receiver operating characteristic (ROC) and area under the curve to check the sensitivity, there was an AUC value between 0.817 and 0.999. It can be concluded that every test set was highly accurate.
2. The validation of the construct validity of the core competency measurement model for third-grade students revealed 4core competencies and it was found that: (1) The model for measuring the literacy competencies had construct validity, with a Chi-Square test = 19.13, df =13, p-value = 0.10, RMSEA = 0.036, CFI = 0.99, TLI = 0.99, and SRMR = 0.01; (2) the ability core competency measurement model had construct validity, with a Chi-Square test = 3.89, df = 6, p-value = 0.52, RMSEA = 0.00, CFI = 1.00, TLI = 1.00, and SRMR = 0.01; (3) the model for measuring well-being competencies had construct validity, with a Chi-Square test = 4.48, df = 3, p-value = 0.21, RMSEA = 0.03, CFI = 0.99, TLI = 0.99, SRMR = 0.01, and the model for measuring social concern competencies had construct validity, with a Chi-Square test = 1.05, df=2, p-value =.59, RMSEA = 0.01, CFI = 1.00, TLI = 1.00, and SRMR = 0.01. Therefore, it can be concluded that all models measuring the core competencies of students in Grade 3 had construct validity.

An Application of Latent Profile Analysis in Group Classification and Study of the Relationship Between Assessment Literacy and Background of Teachers under the Office of the Private Education Commission of Chiang Mai Province

Walaiporn Yaanan — 2024-06-30

This research aimed to classify groups and study the relationship between assessment literacy and the background of teachers under the Office of the Private Education Commission of Chiang Mai Province, by studying 686 teachers obtained through multi-stage sampling. The research instruments included a 5-point rating scale questionnaire on teachers' assessment literacy, consisting of 80 items with an Item-Objective Congruence Index (IOC) ranging from 0.80 to 1.00, index of discrimination (r_xy) ranging from 0.432 to 0.742, t-values ranging from 3.640 to 10.928, and a reliability () of 0.981. Data analysis employed latent profile analysis with the Mplus 7.4 program and Chi-square Analysis. The research findings were as follows:

The grouping of teachers based on their assessment literacy revealed that they could be classified into 3 groups. Group 1, the Accelerated Development Group in Assessment Literacy, had 58 teachers (8.455%). Group 2, the Basic Level Assessment Literacy Group, had 301 teachers (43.878%). Group 3, the Advanced Level Assessment Literacy Group, had 327 teachers (47.668%). (E_k=0.952, Likelihood=-2,356.576, AIC=3,493.660, BIC= 3,629.587, ABIC = 3,534.332, p-value=.005). Furthermore, the study revealed no significant relationship between gender, teaching experience, education level, and learning area groups with the differences in evaluation literacy groups. However, the level of education taught, the school setting, and the school size were significantly related at a significance level of .01.