혼합 문항유형 검사의 가교문항 수에 따른 IRT 동등화 정확성 비교 = Comparison of the Accuracy of Mixed-Format Test Equating Results According to Anchor Item Number Using IRT Equating Methods

국문 초록 (Abstract)

  학교현장과 대단위 평가에서 많이 사용되는 혼합 문항유형(혼합형) 검사는 문항유형에 기인한 다차원성을 지니고 있기 때문에 동등화에 있어 선다형 문항만으로 구성된 검사의 동등화에서 적용되어온 기본 가정이나 경험적인 연구결과를 그대로 적용할 수 있는가에 대한 확인이 필요하다. 이 연구에서는 가교문항 수 대한 기존의 연구결과들이 혼합형 검사의 동등화에서도 적용될 수 있는가를 탐색하였다. 모의 자료에 기초하여 검사 차원성의 5조건과 가교문항의 구성의 7조건을 설정하여 전체 35개 조건에서 4개 문항반응이론 동등화 방법으로 혼합형 검사 동등화를 시행하였다. 연구결과, 혼합형 검사에서는 가교문항 수가 많을수록 동등화 결과의 정확성이 높아진다는 기존 연구결과는 적용되기 어려운것으로 나타났으며 가교문항 수보다는 가교문항 유형의 구성에 따라 동등화 오차가 감소함을 확인할 수 있었다. 또한 가교문항 유형이 전체 검사의 문항유형 구성에 대한 대표성을 가질 때 최소한 요구되는 가교문항 수로도 다른 가교문항 유형에 비해 오차가 적은 경향을 보였다.

다국어 초록 (Multilingual Abstract)

  A simulation study was conducted to investigate the accuracy of equating results according to the anchor item number under the 35 conditions of test dimensionality and anchor test type and under the every conditions 4 IRT equating methods were conducted.<BR>  As a results, considering the number of anchor items, three aspects were investigated. First, regardless of format effects, comparing the effect of the number of the anchor items, when the number of the anchor items of single-format was largest in this study, the error decreased under the unidimensional condition. but when the number of the anchor items of mixed-format was smallest, the error tended to decrease under the multidimensional condition. Second, controlling the constitution of item format to be similar, the error tended to be decreased as the number of anchor items increased. Third, controlling the number of the anchor items, the three anchor test types which had different item format constitution yielded different accuracy of the equating results under the multidimensional conditions. That is, even though the number of the anchor item was same, the equating error in mixedformat anchor tests which had different item format constitutions showed larger discrepancy as the multidimensionality became larger.<BR>  In conclusion, how to consist the different item format in anchor test seems to be more important than increasing the number of anchor items to increase the accuracy of mixed-format test equating results if each item format doesn"t measure same construct.

목차 (Table of Contents)

• Ⅰ. 서론
  Ⅱ. 문항반응이론에 의한 동등화
  Ⅲ. 연구방법
  Ⅳ. 연구결과
  Ⅴ. 결론 및 논의
  참고문헌
  저자소개
  〈ABSTRACT〉
  

참고문헌 (Reference)

1 "공통문항유형과 검사 차원성에 따른 혼합형 검사의 IRT 동등화" 18 (18): 157-182, 2005

2 "The equating of mixed-format forms when test performance is multidimensional" Montreal, Quebec 2005

3 "Test equating, Scaling, and Linking" New York: Springer-Verlag. 2004

4 "Test equating under the multiple-choice model" 26 : 255-227, 2002

5 "Specifying the characteristics of linking items used for item response theory with item calibration" Princeton 1987

6 "Robustness to format effects of IRT linking methods for mixed-format tests" 19 : 357381-, 2006

7 "RESGEN version 4.0: Item Response Generator" 2000

8 "Performance of a proposed method for the linking of mixed format tests with constructed response and multiple choice items" 37 : 329-346, 2000

9 "PARSCALE (version 4.1): IRT item analysis and test scoring for rating-scale data" Chicago, IL: Scientific Software, Inc. 2003

10 "Obtaining a common scale for item response theory item parameters using separate versus concurrent estimation in the common-item equating design" 26 : 3-24, 2002

11 "Multidimensionality and the Equating of a Mixed-format Math Examination" New Orleans, LA 2002

12 "Linking and equating" Westport, CT: Praeger 187-220, 2006

13 "Journal of Educational Measurement" 179-193, 1980

14 "Item characteristic curve solution to three intractable testing problems" 14 : 139-16, 1977

15 "Information functions of the generalized partial credit model" 351-363, 1993

16 "Extension of four IRT linking methods" 43 : 53-76, 2006

17 "Equating logistic ability scales by a weighted least squares method" 144-149, 1980

18 "Efficiency of linear equating as a function of the length of the anchor test Journal of Educational Measurement" 13-2, 1985

19 "Developing a common metric in item response theory" 207-12, 1983

20 "Data analysis for the reading assessment National Center for Educational Statistics" 21-, 1992

21 "Comparison of scale linking results in mixed-format test under the different conditions of common item and test dimensionality" Montreal, Quebec 2005

22 "Applications of item response theory to practical testing problems" 1980

23 "An empirical examination of the IRT information of polytomous Scored reading items under the generalized partial credit model" 31 : 295-311, 1994

24 "American Council on Education Reprinted as W" (rms and equivalent scores rms) : 508-600, 19711984

25 "A computer program for scale transformation under unidimensional item response theory models: version 1" -1, 2004

26 "A comparison of procedures for constructing large item pools University of Missouri Department of Educational Psychology" 1981

27 "A cautionary note on IRT -based linking of tests with polytomous items" 36 : 336-346, 1999