|
KMID : 1038320170140010032
|
|
보건의료교육평가
2017 Volume.14 No. 1 p.32 ~ p.32
|
|
|
Usefulness of DETECT program to assess internal structure of dimensionality in simulated data and results of Korean Nursing Licensing Examination
|
|
Seo Dong-Gi
Choi Youn-Young
Huh Sun
|
|
|
Abstract
|
|
|
Purpose: The dimensionality of examinations provides empirical evidence to investigate the internal test structure underlying the responses to a set of items. The internal structure is one of the important evidence to demonstrate examination validity sources. Thus, aim of the study is to examine the internal structure of Korean nursing examinations and investigate performances of DETECT program.
Method: Non-parametric methods such as DETECT program have proposed to overcome the limitation of traditional parametric methods as a dimensional testing method. The non-parametric method (DETECT program) was investigated under several conditions in the simulation data and applied to Korean nursing examination.
Results: The DETECT program was performed well in terms of determining the number of underlying dimensions under several different conditions in simulated data. Further, DETECT program properly revealed the internal structure of Korean nursing examination, meaning that it detected the number of dimensions and clustered the items of its respective dimension.
Conclusion: DETECT program performed well to detect the number of dimensions and to assign items for each dimension. This result implied that DETECT method can be useful method to examine an internal structure of an assessment such as licensing examinations possessing relatively many domains and contents.
|
|
|
KEYWORD
|
|
|
Dimensionality test, DETECT, Korean nursing licensing examination, Internal structure
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
  
|