Computerized adaptive testing (CAT) studies in Turkey: A systematic review

This study aims to systematically compile and analyze the current state of research on Computerized Adaptive Testing (CAT) in Turkey. The study examines the years in which CAT research in Turkey has concentrated, the techniques used, the types of publications, sample diversity, and other methodological characteristics. Within the scope of the research, CAT studies in the literature were selected and analyzed based on specific criteria. The most commonly used types of CAT, item selection and ability estimation methods, test initiation and termination rules, and other technical details were examined. Additionally, the advantages and limitations of CAT were comprehensively evaluated. The findings reveal that there has been an increase in CAT research in Turkey in recent years, with the majority of studies concentrated in the field of education. The most commonly used item selection method is Maximum Fisher Information (MFI), the ability estimation method is Expected A Posteriori (EAP), the test initiation rule is the neutral theta starting point, and the test termination rule is the standard error threshold. Furthermore, a significant portion of the studies lacked information on item calibration methods, and content balancing and item usage frequency control were addressed in very few studies. In conclusion, this review summarizes the current state of CAT research in Turkey, deepening the knowledge base in this field. The data obtained provide valuable insights for future research and contribute to the more effective use of CAT methods.