IIAI Letters on Institutional Research

A Model for Understanding Student Status Using Attendance Data in the First Semester of University

Naruhiko Shiratori — 2025-03-03

This study developed a Hidden Markov Model (HMM) to analyze attendance behaviors of first-year university students during their spring semester, aiming to identify distinct behavioral patterns and examine their impacts. Weekly attendance data was used to estimate latent states, and clustering revealed four representative attendance patterns, including stable attendance and increased absenteeism. The results highlight the potential impact of specific behaviors on academic outcomes, underscoring the importance of preventive interventions in student support and its applicability to future academic guidance.

Analysis of Learning Behaviors Using a System for Visualizing and Sharing Learning Assessments

Nobuhiko Kondo — 2025-09-30

This paper presents a four-year classroom practice (2021–2024) that aims to promote autonomous learning behaviors by visualizing and sharing learning assessments. The study developed and implemented a point-based evaluation system using Google Sheets to ensure that each student’s progress is visible in real time. The system enabled students to monitor their learning status and compare it with peers anonymously. Data from each year exhibited a gradual shift toward higher cumulative point rates with a number of students displaying significant improvement toward the end of the semester. In addition, the responses indicated that students with more positive attitudes toward the system and frequent use tended to achieve high levels of performance. Although the study was unable to establish causal relationships, the results implied that active engagement with the system could positively impact learning. The study also discussed the effects of evaluation criteria and point allocation on learning behavior and highlighted the need for further analyses and intervention strategies to foster engagement.

A Study of the Relationship between Learning in a Professional Seminar and its Workplace Application for Graduates of a Social Sciences Laboratory

Tatsuya Tsumagari — 2025-09-30

This study examined the relationship between learning obtained in seminar activities, such as professional education, and applying that learning at work by conducting a correlation analysis of 148 graduates of a social science undergraduate laboratory at a regional public university in Japan. The results showed that “working skill,” “collaborative problem-solving skill,” and “making arrangements and stepping forward skill” were not significantly correlated with work situations. The following skills were found to be significantly related to work situations. “Communication skill” were positively correlated with work situations, such as “coordinating with other workers” and “cooperation with others,” suggesting that these skills are useful in jobs where interpersonal relationships are important. “Team management skill” was positively correlated with “managing complex tasks,” indicating that it is an important skill for engaging in work, but negatively correlated with “coordinating with other workers,” suggesting that it may not be suitable for external coordination work. Finally, “interviewing and video editing skills” were negatively correlated with work situations such as “efficient work execution,” suggesting that creative work has aspects that conflict with work efficiency.

A Preliminary Analysis Concerning Knowledge Creation and Performance of R&D-oriented SMEs under Uncertainty

Katsuhiro Suzuki — 2025-09-30

Both R&D (Research & Development) and patent applications are important activities that form the foundation of innovation creation in companies. This study analyzed the direct impact of patent application activities, which are considered outcomes of knowledge creation, on the performance of R&D-oriented high-tech small and medium-sized enterprises (SMEs) that support Japan's core industries.

The results showed that for SMEs that filed at least one patent in the 2010s, as the number of patent applications increases, (i) the firm's sales per employee increases and also (ii) the rate of change of sales in the year affected by COVID-19 becomes higher. In addition, the age of the firm was found to significantly contribute to the improvement in sales per employee, suggesting that continuous knowledge creation activity tends to improve firm performance.

Applying Text Generation AI to Assist Categorical Database Construction from Institutional Regulations

Tsunenori Inakura — 2025-09-30

We demonstrated that a text generation AI (ChatGPT) can support the construction of categorical databases from institutional documents. By extracting concepts, formulating functional relationships, and verifying the semantic validity of composite morphisms consistency through natural language, this approach supports the creation of database schemas that accurately reflect the meaning of the original documents. It also makes it easier for non-experts to take part in schema building, providing a clear way to transform written regulations into structured data. This semi-automated method not only reduces manual workload but also improves clarity and maintainability of database structures. Our findings highlight the potential of language models to bridge formal data modeling and natural language logic in educational and administrative domains.

Clinical Reasoning and Decision Making through Knowledge Networks and Abduction

Yasuo Nakata — 2025-02-22

This study examines the integration of clinical decision making processes with knowledge networks and abductive reasoning in nursing practice, proposing a sustainable framework based on eduinformatics. While clinical reasoning traditionally relies on deductive and inductive approaches, the complexity of modern healthcare demands more sophisticated decision-making methodologies. Through analysis of clinical cases and reasoning patterns, we demonstrate how abductive reasoning complements traditional approaches, particularly in situations where complete information is unavailable. The knowledge network theory provides a structured framework for understanding how clinical knowledge is created, shared, and applied. By integrating these elements through eduinformatics, we develop a comprehensive approach that enhances clinical reasoning capabilities in nursing education and practice. This framework offers a systematic way to improve clinical decision-making while maintaining sustainability in increasingly complex healthcare environments.

Obtaining Factors of What Was Acquired through Questionnaire Aimed at Graduates of an Educational Institute

Yuya Yokoyama — 2025-09-30

Radical alternations of social structures have brought about the situation where the importance of relearning new knowledge and skills for adults is becoming increasingly prioritized. However, various hindrances have put Japan behind other countries when it comes to adult learners. There-fore, our study aims to analyze the fundamental factors of study motivation of adult learners and to grasp what drives their learning. Based on the prior works using a questionnaire aimed at gen-eral working adults, in this paper we create a questionnaire targeting the graduate students of our affiliations. After aggregation of the questionnaires, factor analysis was applied to the 29 ques-tions about what was acquired during studying to obtain five factors.

Developing a Tutorial for SCOT Student Training with Automated Feedback Support

Kenshin Murakami — 2025-09-30

This study proposes a tutorial designed to support SCOT (Students Consulting On Teaching), a student-led class evaluation activity, by enhancing students’ understanding of evaluation criteria key factors influencing the quality of evaluations. The tutorial incorporates a self-assessment rubric that allows students to monitor their own comprehension as they progress through the learning process. Feedback tailored to their rubric responses is generated by a system. An experiment was conducted to investigate the impact of system-generated feedback on evaluation activities within the tutorial. Additionally, this paper proposes a method for a future experiment to verify the effectiveness of the tutorial in actual SCOT implementations.

A Trial for Developing Moodle-based E-portfolio for Post Graduate Medical Education in Japan

Yoshikazu Asada — 2025-09-30

This study aimed to develop and trial an e-portfolio system for postgraduate medical education in Japan using Moodle. Building on the national clinical training e-portfolio framework, the system was designed to flexibly manage clinical experiences by integrating multiple Moodle modules: (1) Database module for case recording, (2) Questionnaire module for self-assessment, (3) Quiz module for skill tracking, and (4) Configurable Reports plugin for data aggregation. The system demonstrated strengths in supporting both formative and summative assessments, although operational challenges such as the need for technical expertise and UI improvements were noted. These findings suggest that Moodle offers a scalable and adaptable foundation for e-portfolio development in postgraduate medical education.

Investigation of Latent Effects and Changes of Adult Learners at Colleges or Graduate Schools

Yuya Yokoyama — 2025-03-03

In contemporary society, where radical changes of social structures are being taking place, it is getting more important for adult learners to study to obtain new knowledge and skills. From these circumstances, what triggers adult learners to study and what prevents them from learning are greatly emphasized. There are several forms of study motivation: getting a certification, seeking new career paths, or simply academic interest, among others. However, Japan is behind other countries when it comes to adult relearning due to various hurdles along the way. Meanwhile, recurrent education is attracting attention along with the widespread use of various education methods to handle multiple demands. Therefore, this paper aims to analyze the potential factor of study motivation of adult learners and to grasp what motivates or prevents their relearning. As a first phase, we formulate four hypotheses regarding the relationships between study motivation and achievement. These hypotheses are then validated using questionnaire targeting adult learners. As a result of analysis, the relationships between before and after studying at college or graduate school can be observed. It could also be implied that the questionnaire targeting adult learners who completed college or graduate school would be effective in examining effects and changes.

Designing Behavioral Indicators for Assessing Learning Outcomes Based on Diploma Policy

Sayaka Matsumoto — 2025-09-30

In higher education, assessing learning outcomes based on diploma policy (DP) is essential for ensuring educational quality. However, in Japan, such assessments often rely on subjective self-evaluation, which is prone to bias, contextual effects, and gaps between perception and actual behavior. This study proposes a method to visualize and deconstruct the abstract language of Diploma Policy (DP) statements using the principles of ontology engineering, followed by the development of concrete behavioral indicators through expert discussion. The target DP, defined by the former Tokyo Institute of Technology, spans the bachelor’s, master’s, and doctoral programs. A total of 24 behavioral indicators were developed across six domains: specialist skills, liberal arts (1 and 2), communication skills, and applied skills (inquiry/problem setting and practice/problem solving). These indicators visualize student competencies based on the frequency of observed behaviors, as an alternative or complement to self-perception. The framework is applicable not only at graduation but also during a program, allowing timely feedback for educational improvement and student support. This study presented a novel approach for evaluating DP achievement in a more objective and multidimensional manner. It also holds potential as a complementary framework for designing behavioral indicators and surveys in educational systems that already use outcome-based assessments.

Do Involuntary Enrollees in Teacher Training Courses Aspire to Teach?

Ryosuke Sakurai — 2025-09-30

This study examines whether students who entered a teacher training university in Japan as a second or subsequent choice still aspire to become teachers. Specifically, the study conducted a questionnaire survey that targeted first-year students in the teacher training course at a national university with multiple campuses. The study analyzed the association between admission categories, university or campus preferences, and career aspirations using a log-linear model. The analysis of 675 responses showed significant associations between admission categories and university preferences, and between university preferences and career aspirations. Parameter estimation showed that students who were admitted through recommendation selection had a lower rate of involuntary enrollment, whereas those who were admitted in the second round in the general selection had a higher rate of involuntary enrollment. More importantly, involuntary university-level enrollees showed a lower rate of teacher aspirations than voluntary enrollees, whereas involuntary campus-level enrollees showed a rate of teacher aspirations that was similar to those of voluntary enrollees. These results suggest that by designing an entrance examination that discourages involuntary enrollment from other universities, it is possible to select a higher rate of students who aspire to become teachers.

Humor, Mankind’s Greatest Blessing: The Relationship between Sense of Humor and Psychological Wellbeing among High School Students in Taiwan

CHIH-AN LIN — 2025-09-30

The aim of the present study was to investigate the relationships between sense of humor and self-esteem, depression, emotional blockage, and poor coping. In addition, the statistical differences between the variables in different groups of participants were examined. Chinese questionnaires with multiple scales were used to collect the data. For all participants, there was a positive correlation between sense of humor and self-esteem and a negative correlation with poor coping. As the result suggests, sense of humor is related to self-esteem and not to depression. People with higher sense of humor scores do not always have lower depression scores. In addition, no correlation was found between sense of humor, self-esteem, depression, emotional blockage, and poor coping in the male students. Finally, the students with higher scores for sense of humor and depression had lower scores for self-esteem.

The Practical Application of Cluster Analysis of Aca-demic Fields in Bibliometric Information to Enhance Research Performance Evaluation

Satoshi Ozeki — 2025-03-03

In recent years, the research capabilities of Japanese universities have declined compared to other countries, highlighting the need for effective evaluation of research performance. Due to the diversity of academic disciplines and the rise of interdisciplinary research, identifying comparable researchers presents a challenge. Therefore, this paper presents a method for identifying comparable researchers by classifying research fields through the application of bibliometrics. Specifically, using the Scopus database, this study conducted cluster analysis on the research topic profiles of researchers from multiple universities within the same field to group them. Additionally, this paper demonstrates how the results of cluster analysis can be applied to enhance the evaluation of research performance.

Evaluation of an Asynchronous Lecture Model based on Motivation Scales

Shinnosuke Yamaguchi — 2025-09-30

In this paper, we proposed an asynchronous lecture model to encourage the self-directed learning of students, and we conducted the lecture in practice. We evaluate this asynchronous lecture model using students’ motivation scales and learning activity records obtained from the learning management system (LMS). On the basis of our investigation of student activities categorized by motivation level, we observed a high assignment completion rate among highly motivated students, as well as voluntary learning activities among some students whose motivation had increased.

Development and Operation of an All-in-One IR System

Tetsuya Oishi — 2025-09-30

In recent years, Institutional Research (IR) has become increasingly important for decisionmaking support in Japanese higher education institutions. This paper reports on the development and operation of an IR system at Kyushu Institute of Technology, which has been under construction since fiscal year 2022. The system integrates a data lake server with an ETL (Extract, Transform, Load) server to provide various functions including grade distribution visualization, automated infection report generation, and integration with academic affairs systems. The key feature of this system lies in its utilization of programmatically operable tools such as a data lake, an ETL Tool, a statistical software, and PDF creation to automate data collection, restructuring, analysis, and reporting processes within the information support cycle. This approach has improved operational efficiency and reduced costs at the university while establishing a foundation for data-driven decision-making. Future plans include system expansion with the introduction of data warehouse and BI tools.

Practice and Evaluation of Data Science Education for First-Year Students in Foreign Language Faculties

Shuntaro Iseri — 2025-09-30

This study presents the design, implementation, and evaluation of an introductory data science (hereinafter, DS) course tailored specifically for first-year non-STEM students at Kanda University of International Studies (KUIS), focusing on the 2024 academic year cohort. Recognizing common barriers faced by non-STEM students, such as limited mathematical literacy, inadequate ICT skills, and low motivation, the course employed accessible methods to foster engagement and effective DS education. Key features included an emphasis on numerical reasoning over complex mathematics, utilizing student-centered activities using authentic datasets, and extensive practical exercises utilizing Tableau, a no-code business intelligence (BI) tool. Delivered in an on-demand format to approximately 900 students, the course achieved a 90.3% pass rate and led to significant improvement in students' self-assessed DS competencies. It effectively bridged knowledge and skill gaps between students with and without prior DS experience. However, motivational gains were modest, indicating areas for pedagogical improvement. Future research should address potential biases from voluntary survey participation, deepen motivational analyses, and explore strategies that explicitly link DS education with broader career relevance.

Secular Changes in Career Preferences Among Medical Students in Japan

Koji Tsunekawa — 2025-09-30

Introduction: Career-related research is one of the important tasks in institutional research at medical schools. This study investigated changes in the career preferences of medical students in Japan per specialty and year to identify solutions to the maldistribution of medical specialties. Methods: The study recruited 873 medical students who were enrolled during academic years 2018–2023. The survey enabled multiple responses to their interests in 22 medical specialties. The survey covered seven periods from enrollment to graduation and investigated choice rates, changes (selection and shift), and trends in student choices. Results: Several specialties exhibited more than 5% difference in choice rate by period. Fifth-year students obtained a significantly higher change rate compared with those in other levels. Internal medicine, orthopedics, radiology, and plastic surgery displayed significantly higher rates for selection, while pediatrics produced higher rates for shift. Each student demonstrated a characteristic choice trend. Discussion: these results were consistent with previous studies on many comprehensive specialties except for surgery and also revealed that clinical clerkship is very important in their career choice. In conclusion, we found that establishing education designed based on an understanding of each student’s type as well as appropriate timing of education are crucial.

Assessing Social Applicability of Student Learning Out-comes Considering Faculty and Career Path Differences

Shinji Hosokawa — 2025-09-30

Although the social implications of student learning outcomes have been studied, the impact of diversity on educational objectives and career trajectories remains unclear. This study examined graduates’ self-assessments of student learning outcomes to evaluate variations across faculties and career paths. Using cluster analysis, four groups were formed, each demonstrating patterns of skill recognition—positive, indifferent, selective awareness, and balanced—which highlighted unique distribution patterns among faculty and career selections.

Mapping ISO 15408 Security Functions to SDLC Phases: Insights from a Questionnaire-Based Study

Samodi Fernando — 2025-09-30

The ISO/IEC 15408 standard provides a systematic approach to defining security functions that are essential for software systems, yet integrating these security functions into the Software De-velopment Life Cycle (SDLC) phases remains challenging. This research aims to establish con-nections between ISO 15408 security functions and SDLC phases by using survey data from software professionals. The survey collected 144 responses from software developers together with quality assurance engineers and security professionals. The analysis shows patterns of se-curity function implementation, which reveal both early-phase adoption gaps and inconsistent audit-related practices. The research evidence demonstrates how better integration of ISO 15408 functions with SDLC would enhance secure software development practices. This research pro-vides practical insights about uniting security standards with actual development workflows.

Japanese Tokenization using Simple Associative Arrays

Michiko Yasukawa — 2025-09-30

This study addresses a technical challenge in text analysis for the practice of Institutional Research (IR). In the research field of Natural Language Processing (NLP), development focuses on accuracy and speed. Therefore, suitable data for evaluation and efficient programming languages are used. On the other hand, tasks outside the scope of NLP are also important in IR. For example, not only textual data, but also numerical data must be analyzed, and in some cases, the analysis of data stored in web servers is required. Since NLP and IR involve different tasks, the tools required are also different. One serious problem for text analysis in IR tasks is the limited development of Japanese tokenization. In this study, we implement a Japanese tokenizer, focusing on its applicability in web servers. Since function libraries in scripting languages for web environments are limited, our main technical challenge is to achieve efficiency in tokenization by combining the available functions. To deal with this problem, our method uses associative arrays to reduce unnecessary steps in tokenization. Evaluation experiments were conducted using Japanese text data, and the results showed that our tokenizer is viable in web server environments.

Eduinformatics and the Universities’ Challenge for “Ri”

Kunihiko Takamatsu — 2025-03-03

This paper examines the transformation of higher education through the lens of Eduinformatics and the Japanese concept Shu-Ha-Ri. It analyzes the current state of universities, characterized by rigid academic divisions and external pressures, and proposes a path through interdisciplinary fusion and the introduction of absolute perspectives. The paper then envisions a future state 100 years from now that transcends constraints of time, place, and culture. It argues that a "Ri" part of universities creates new values through the fusion of Japanese cultural elements with global diversity by maximizing creativity and sensibility, and actively engages with society to solve real-world problems. The study concludes thatwhile the path to transformation is not unimpeded, it is essential for the future relevance and impact of higher education.

Exploring Differences in Time Spent Tracking

Yuji Kobayashi — 2025-09-30

This study compares “Time Spent” measurements recorded by two Moodle analytics plugins—IntelliBoard and Edwiser Reports Pro—in a graduate-level instructional design course. Data from various activities were analyzed using descriptive statistics and paired t-tests. IntelliBoard recorded a significantly longer total Time Spent than Edwiser Reports Pro. Since Edwiser Reports Pro was installed after the course, it calculated Time Spent using standard Moodle log data. Scatter plots were created to visualize the relationship between the two plugins. The findings suggest that differences in tracking methods and learner behavior can influence Time Spent data. These metrics should be interpreted with caution in educational evaluations. Future research should examine diverse learning contexts and incorporate learner self-assessments to validate Time Spent as an indicator of engagement.

Research on First Year Students’ Global Competence Development and its Impact on Study Abroad: An Analysis Through the Intercultural Collaborative Learning Course “MCW”

Noboru Miyoshi — 2025-09-30

In 2024, the Ministry of Education, Culture, Sports, Science and Technology in Japan (MEXT) launched the Support Program for Creating Social Impact through University Internationalization. This involves setting intercultural collaborative learning courses where domestic and international students study together, gaining global competence, and fostering interest in study abroad programs to actively send domestic students. Building on this background, the present study seeks to explore three main questions: (1) Did students in intercultural collaborative learning course develop global competences? (2) What factors positively affect global competence, and (3) Whether gaining global competence promotes study abroad. Within this framework, APU’s (Ritsumeikan Asia Pacific University, Japan) SI project plan¹^） references previous intercultural collaborative efforts, including a first year experience course called “Multicultural Collaborative Workshop (MCW) ²^）”, and residential experience in the international first-year dormitory AP House. The analysis yielded three key findings. First, students generally showed an increase in global competence from the beginning to the end of the MCW course. Second, notable improvements in global competence were observed among students who actively engaged in group work during MCW, and those with residential experience in AP House who frequently interacted with peers from the opposite language standard. Third, students who developed higher levels were more likely to be motivated to participate in study abroad programs.

Contradiction and Abduction

Kunihiko Takamatsu — 2025-09-30

The increasing use of data in educational research, framed within fields like Eduinformatics, necessitates effective strategies for hypothesis generation within the cyclical data-driven and hypothesis-driven approach. However, the crucial step of forming novel hypotheses often relies on Abduction, a form of inference known to be challenging. This paper addresses the difficulty of abductive hypothesis formation by exploring potential aids. Through conceptual analysis grounded in Eduinformatics and IR Philosophy, we investigate the role of contradiction in facilitating this process. We argue that constructively engaging with contradictions—whether they arise from data conflicting with existing theories or from differing normative viewpoints—serves as a powerful catalyst for the "surprise" that triggers abduction. The development of the Significant Other Groups (SOGs) concept is presented as an illustrative case where analyzing contradiction led to new theoretical insight. This study concludes that reframing contradiction not as an obstacle but as a vital prompt for inquiry offers a valuable perspective for enhancing creativity and knowledge generation, thereby aiding abduction within the research cycle in Eduinformatics. This perspective helps reconstruct the cycle linking hypotheses and data by leveraging the generative power inherent in contradiction.

Categorical Database for Ensuring Data Integrity in Institutional Research

Tsunenori Inakura — 2025-03-03

Institutional research deals with large and different datasets from various departments, and this can make it hard to keep the data accurate. In this paper, we present categorical databases, which is a method based on category theory, that helps to maintain data integrity. By organizing the database as a category, we can see how the data elements are connected. This makes it easier to do meaningful and precise data analysis. The connections between data can be shown as simple sentences that still make sense, even when the data is updated. This way ensures that data remain consistent in both databases and data warehouses through natural transformations, which means that references to the data stay trustworthy. Categorical databases provide a solid way to manage complex data structures, and they make sure that data integrity is kept.

The Impact of Introducing University Research Administrators (URAs) into Japanese National Universities

Masakuni Taniguchi — 2025-09-30

Japanese national universities have traditionally operated with a government-allocated management expense grant (hereinafter, the grant). However, due to several factors weakening their financial bases, including reductions in the total size of the grant and general price increases, national universities now face the need to improve financial health by diversifying income sources. Consequently, academic staff striving to obtain funding find themselves with less time for research. In this context, university research administrators (URAs) have attracted significant attention as a means of revitalizing research capabilities by relieving academic staff of many administrative tasks. However, research on URAs remains underdeveloped. This study aims at clarifying the status of URAs in Japanese national universities by conducting a correlation analysis using various evaluation indicators to determine how URAs, directly and indirectly, contribute to strengthening research capabilities and improving reputation management. The results show that the number of URAs in national universities positively impacts the amount of academic research funding acquired and the volume of research outputs. URAs also contribute to a university’s reputation. Therefore, the position of URAs in national universities is expected to become increasingly important.

A Study on 3D Printer in Japanese Schools: Focus on Courses of Study and Number of Articles

SHOKO USUI — 2025-09-30

The use of 3D printers in education is becoming increasingly common. However, there is a paucity of objective data regarding their implementation in Japanese schools. Furthermore, few resources in languages other than Japanese present this information, creating barriers to international comparative research. To bridge this knowledge gap, this paper provides empirical data on the integration of 3D printers into Japanese school education. Our analysis focuses on two key areas: (1) how 3D printers are addressed in the Commentary on the Courses of Study (a significant shift from their absence in the prior edition to inclusion in four current subject areas), and (2) the publication trends of 3D printer-related articles in Japanese academic literature. Our findings indicate that while such articles began appearing around 2014, their volume has remained modest with only a slight increase over time.

Examining Current State of Data Science Education in High Schools and Higher Education Institutions

ERIKO TANAKA — 2025-09-30

Data science education (DSE) has become a global trend, and in Japan, it is gaining increasing emphasis, as evidenced by the introduction of accredited programs for higher education institutions. However, despite the proactive establishment of institutional frameworks, there has been limited discussion regarding the educational effectiveness within individual institutions, such as high schools and universities, as well as the continuity between these educational stages. This study administered a data science comprehension assessment to 406 high school and 1,652 first-year university students in early April 2025, prior to receiving formal DSE at their respective institutions. While university students scored slightly higher than high school students, the overall comprehension levels were low, and differences across departments, particularly between STEM and non-STEM students, were negligible. These results suggest the need to reassess the respective roles of individual educational institutions, strengthen curricular continuity, and refine the specific content delivered at each stage.