Exploring the potential of artificial intelligence tools in educational measurement and assessment

Abstract

Artificial intelligence (AI) is transforming various industries, and education is no exception. Rapid advancements in AI technology have become essential for educators and educational assessment professionals to enhance teaching and learning experiences. AI-powered educational assessment tools provide numerous benefits, including improving the accuracy and efficiency of assessments, generating personalized feedback for students, and enabling teachers to adapt their teaching strategies to meet the unique needs of each student. Therefore, AI has the potential to revolutionize the way education is delivered and assessed, ultimately leading to better educational outcomes for students. This paper explores the various applications of AI tools in educational measurement and assessment. Specifically, it discusses the integration of large language AI models in classroom assessment, in specific areas such as test purpose determination and specification, developing, test blueprint, test item generation/development, preparation of test instructions, item assembly/selection, test administration, test scoring, interpretation of test results, test analysis/appraisal, and reporting. It analyses the role of teachers in AI-based assessment and the challenges of using AI-powered tools in educational assessment. Finally, the paper presents strategies to address these challenges and enhance the effectiveness of AI in educational assessment. In conclusion, using AI in educational assessment has benefits and limitations. As such, educators, policymakers, and stakeholders must work together to develop strategies that maximize the benefits of AI in educational assessment while mitigating the associated risks. The application of AI in educational assessment can ultimately transform education, improve learning outcomes, and equip students with the skills needed to succeed in the 21st century.

Revolutionizing Moroccan Education with AI: A Path to Customized Learning

“…These enhancements offer a more streamlined approach to resource management, allowing educators to devote their time to more meaningful educational activities. Using AI-driven insights, educational institutions can personalize educational materials to successfully address each student's unique needs, creating an environment that promotes academic and personal development [85]. Automating regular tasks such as grading, attendance monitoring, and scheduling increases operational efficiency and improves data management accuracy and reliability.…”

Using chatbots to develop multiple-choice questions. We got evidence, but we ain't there yet!

“…As illustrated in Table 1, the median and interquartile range (IQR) per item for quality are presented. The MCQs exhibited a high median quality score of 24.5 (IQR: [22][23][24][25]. A comparison revealed that chatbotgenerated items had a median score of 25 (IQR: 21.25-25), slightly higher than the human-generated items which scored a median of 24 (22.25-25), however, this cannot be interpreted as superiority.…”

“…From its integration into the teaching and learning environment, it uses as a collaboration coach, personal tutor, and study buddy to an ally for educators, to aiding in the design of educational material, and in our case -the development of assessment. In the eld of educational measurement and assessment, several authors (Khilnani, 2023;Owan et al, 2023) have proposed its use to ameliorate the workload of educators. However, using these types of tools requires evidences of validity before applying them in the educational context.…”

MxML (Exploring the Relationship between Measurement and Machine Learning): Current State of the Field

“…Given the typical lag between technological advancement and publishing research, very few generative AI studies have yet been published in measurement journals (although some exceptions include Zheng et al, 2023, which examined a Chinese composition tutoring system in the Chinese/English Journal of Educational Measurement and Evaluation; Zhang et al, 2022, a preprint that summarized current ChatGPT studies, including those relevant to testing; and Owan et al, 2023, which explained how ChatGPT and other artificial intelligence tools can be used in personalized teaching and automated grading). Most papers that currently exist are either preprints, white papers, or publications in journals that we did not consider for inclusion in this study (such as ML journals or educational technology journals).…”