Abstract

본 연구는 혼합분포 문항반응모형 적용 시 집단의 이질성에 따른 잠재집단 추정의 정확성을 확인하여, 모형 활용을 위해 요구되는 이질성 수준을 제안하는 데 목적이 있다. 특히 기존 혼합분포 문항반응모형 연구에서 다루지 않은 하위집단 간 거리로 집단의 이질성을 측정하였다. 2-모수 혼합분포 문항반응모형을 사용하였으며, 모의실험을 통해 다양한 검사 조건(잠재집단의 수, 피험자 수, 잠재집단의 비율, 모수분포 평균의 차이)에서 잠재집단 간 마할라노비스 거리에 따른 잠재집단 수 결정의 정확성과 피험자 분류 정확성을 확인하였다. 연구결과, 잠재집단 간 마할라노비스 거리가 3 이상인 조건에서 잠재집단 수 결정 정확도와 피험자 분류 정확도가 눈에 띄게 높아졌다. 본 연구는 추후 혼합분포 문항반응모형을 활용하고자 하는 연구자들에게 잠재집단 추정의 정확성을 예측하기 위한 정보와 이를 바탕으로 모형 활용을 위해 요구되는 집단의 이질성 수준에 대한 정보를 제공하는 데 의의가 있다.The purpose of this study is to suggest a degree of heterogeneity between classes in order to utilize a model, after confirming the accuracy of latent class estimation, based on distance, when utilizing mixture IRT. The heterogeneity between the classes was measured by the distance between latent classes, which is not a factor being dealt within the existing mixture IRT research. In order to confirm this, a simulation study was used that allowed for the assumptions of the following varying conditions; numbers of classes (2class, 3class), sample sizes (1,000, 2,500), mixing proportions (equal, unequal) the difference of mean parameter distributions (0, 0.5, 1, 1.5), and the distance between the latent classes (2, 3, 4, 5). The accuracy of correct model selections based on the heterogeneity of latent classes and the accuracy of examinee classification, were confirmed by the simulation study. In conclusion The results show that there is a consistent tendency that if the Mahalanobis distance between latent classes is 3 or more, then the higher the accuracy of examinee classifications and correct model selections. This study provided information to researchers who intend to utilize mixture IRT; the information to predict the accuracy of latent class estimation and based on this, information about the degree of heterogeneity between classes required to utilize a model.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.