Moses Oluoke Oluoke Omopekunola

Mr. Moses Oluoke Oluoke Omopekunola

orcid

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 October 2024 in REID (Research and Evaluation in Education)
Verified
Journal Article
Open access
PDF file
MOMoses Oluoke Oluoke Omopekunola
Moses Oluoke Oluoke Omopekunola
EKElena Yu Kardanova
Elena Yu Kardanova
High-quality items are essential for producing reliable and valid assessments, offering valuable insights for decision-making processes. As the demand for items with strong psychometric properties increases for both summative and formative assessments, automatic item generation (AIG) has gained prominence. Research highlights the potential of large language models (LLMs) in the AIG process, noting the positive impact of generative AI tools like ChatGPT on educational assessments, recognized for their ability to generate various item types across different languages and subjects. This study fills a research gap by exploring how AI-generated items in secondary/high school physics aligned with educational taxonomy. It utilizes Bloom's taxonomy, a well-known framework for designing and categorizing assessment items across various cognitive levels, from low to high. It focuses on a preliminary assessment of LLMs ability to generate physics items that match the Bloom’s taxonomy application level. Two leading LLMs, ChatGPT (GPT-4) and Gemini, were chosen for their strong performance in creating high-quality educational content. The research utilized various prompts to generate items at different cognitive levels based on Bloom's taxonomy. These items were assessed using multiple criteria: clarity, accuracy, absence of misleading content, appropriate complexity, correct language use, alignment with the intended level of Bloom's taxonomy, solvability, and assurance of a single correct answer. The findings indicated that both ChatGPT and Gemini were skilled at generating physics assessment items, though their effectiveness varied based on the prompting methods used. Instructional prompts, particularly, resulted in excellent outputs from both models, producing items that were clear, precise, and consistently aligned with the Application level of Bloom's taxonomy.

19 June 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
MOMoses Oluoke Oluoke Omopekunola
Moses Oluoke Oluoke Omopekunola
BNBaizhanov Nurseit
Baizhanov Nurseit
EKElena Yu. Kardanova
Elena Yu. Kardanova
Assessment item development requires careful control of construct alignment, response structure, difficulty, language, and review evidence. Large language models can accelerate item drafting, but ad hoc prompting is difficult to audit or reproduce. AIGRA is an R package for structured, agentic item generation, review, and analysis. It supports tabular item-bank templates, Python-backed generation workflows, direct API-key handling, diagram-aware item generation, quality summaries, and HTML reporting. The package is designed for template-based item clone generation, where new items preserve the source item’s assessed skill, reasoning pattern, response format, language requirements, and diagram dependency. This article describes the software architecture, workflow, examples, and research impact of AIGRA.

27 June 2023 in Education Mind
Verified
Journal Article
Open access
PDF file
MOMoses Oluoke Oluoke Omopekunola
Moses Oluoke Oluoke Omopekunola
AFAyoola Waliu Fagbenro
Ayoola Waliu Fagbenro
AOAyomide Emmanuel Ojo
Ayomide Emmanuel Ojo
OAOlusoji Adeayo Akinbo
Olusoji Adeayo Akinbo
The study examined how "Just-in-Time-Teaching" (JiTT) affected physics students' academic performance in Ondo State, Nigeria. Two hypotheses were examined while the study was driven by two research questions. The study used pretest - posttest control group research design. 100 pupils were adopted for this study. Two secondary schools from the local government region were chosen at random, with one serving as the experimental group and the other as the control group. The study's target population was SS I students taking Physics in the identified schools. JiTT was used employed to for the experimental group while the control group was given the same information in a normal classroom situation. The Just-in-Time Physics Achievement Test (JiTTPAT), which comprises 20 multiple-choice questions was developed by the researchers on the topic of measurement of physical quantities, was used to gather data with reliability index of 0.76. Mean, standard deviation, and ANCOVA were used to analyze the data with the aid of SPSS. The study came to several conclusion, including the following: JiTT improves student achievement grades in Physics, and gender had no significant effect on students' performance in physics. According to the results, it was suggested, among other things, that JiTT be implemented in secondary schools to enhance and improve students' achievement grades in Physics.

8 January 2026 in Journal of Science Education and Technology
Verified
Journal Article
Closed access
PDF file
MOMoses Oluoke Oluoke Omopekunola
Moses Oluoke Oluoke Omopekunola
EKElena Yu. Kardanova
Elena Yu. Kardanova

1 October 2026 in Sage Open
Verified
Journal Article
Open access
LSLyazzat Shinetova
Lyazzat Shinetova
MOMoses Oluoke Oluoke Omopekunola
Moses Oluoke Oluoke Omopekunola
EKElena Yu. Kardanova
Elena Yu. Kardanova
BABolatbek Abdrasilov
Bolatbek Abdrasilov
NBNurseit Baizhanov
Nurseit Baizhanov
AAAnar Auezkhanova
Anar Auezkhanova
High-stakes educational assessment requires large numbers of high-quality items, yet item writing and review remain labor-intensive, especially when items include visual stimuli. This study presents a descriptive feasibility evaluation of an agentic large language model (LLM) framework implemented as a role-based, multi-stage LLM orchestration workflow for generating and reviewing multimodal physics multiple-choice items for Kazakhstan’s Unified National Testing (UNT). The framework combines retrieval-augmented generation, structured item generation, answer verification, automated critique, automated revision, and human subject-matter expert (SME) review. Using Gemini 3 Pro as the backbone model, the system generated 80 clone items aligned with the UNT blueprint, including 13 items with diagrams. Two SMEs evaluated the items on answer-key correctness, stem clarity, distractor quality, difficulty alignment, and diagram quality. Because the evaluation involved only two raters, all results are reported as descriptive observations from two individual raters, together with raw inter-rater agreement and chance-corrected agreement coefficients, rather than as generalizable reliability evidence. Most generated items were rated acceptable on stem clarity and distractor quality, while difficulty alignment and diagram fidelity required closer human review. The study’s methodological contribution is a practical human-AI workflow that separates generation, verification, critique, and revision rather than treating item generation as a single-pass task. The findings suggest that agentic LLM pipelines may support educational assessment development by reducing drafting burden and prioritizing expert time for higher-value judgments, although operational use still requires human oversight and further validation with examinee response data.

16 September 2023 in ScienceOpen Preprints
Verified
Preprint
Open access
MOMoses Oluoke Oluoke Omopekunola
Moses Oluoke Oluoke Omopekunola
JAJacob Abel
Jacob Abel
AAAhmed Abubakar
Ahmed Abubakar
AAAugustina Achimugu
Augustina Achimugu
This study assessed the trends and variability of rainfall in Wukari, Nigeria, from 1981 to 2021, using the ECMWF ERA5 reanalysis data sets. Rainfall trends and variations over the study period were analyzed using Mann Kendal trend test and the Theil Sen slope estimator. The assessment of monthly rainfall variation for the rainy months (May-October) showed an increasing trend for August, September, and October, with August having the highest increasing trend of magnitude 0.051 mmmonth -1. The result also showed an encroachment of the dry spell towards the rainy season and vice versa. This will create a variation in onset of rainfall and cessation in the coming decades, which will affect the farming season in Wukari in terms of the time of planting and harvesting of crops. A decline in annual rainfall of magnitude -0.005mm/ year was observed within the period 1981 to 2021. The rainfall pattern revealed a periodic trend on a decadal basis with an increasing trend been followed be decreasing trend in the next decade. The highest increasing trend of magnitude 0.73 mmdecade -1 was observed in the third decade, and the highest decreasing trend in the fourth decade, with magnitude -0.93 mmdecade -1. Based on the trend pattern, an increasing trend in rainfall amount is expected in next decade (2021-2030), with a higher increasing trend magnitude greater than that of the third decade. Therefore, it is recommended that Government Agencies and stakeholders in the agriculture sector should be proactive in educating/enlightening farmers on the likelihood of a change in the farming season and make adequate preparation to mitigate the effect of flooding in the area.