쌍요인(Bifactor) 모형을 이용한 심리척도의 측정적 속성 연구방법 개관 = Bifactor Modeling Approach to Investigate Studying of Psychometric Properties of Psychological Measures

다국어 초록 (Multilingual Abstract)

Bifactor modeling approach is increasingly being applied to the study of psychometric properties of psychological measures. A bifactor structure consists of a single general factor that is purported to explain co-variances of all the items and a set of group factors that are purported to explain residual co-variances of some items that cannot be accounted for by a general factor. The model assumes that the general and group factors are uncorrelated. Bifactor modeling approach enables researchers to test whether a given psychological scale that is originally designed to measure a theoretically unidimensional construct appears to be multidimensional due to nuisance factors such as method factors. In this article, we gave an overview of statistical indices such as omega coefficients and explained common variance(ECV) that can be effectively employed to investigate dimensionality of a given scale. We illustrated how to compute various types of omega coefficients and explained common variance(ECV) and interpret them using Rosenberg Self-esteem scale(RSES) and Emotional Approach Coping Scale(EAC).

국문 초록 (Abstract)

최근 심리척도의 차원 구조를 연구하기 위한 측정적 속성 탐색에 있어 쌍요인(Bifactor) 모형 활용에 대한 논의가 증가하고 있다. 쌍요인 모형은 척도의 모든 문항의 공통 분산을 공유하는 하나의 일반요인(general factor)과 일반요인이 설명하지 못하는 문항의 분산을 공유하는 집단요인(group factor)로 이루어져있으며 일반요인과 집단요인은 서로 상관이 없다(orthogonal)고 가정하는 모형이다. 쌍요인 모형을 활용하면, 이론적으로 하나의 일반요인을 측정하고자 제작된 척도가 문항의 표현 효과(wording effect) 등과 같은 방법 요인(method factor)으로 인해 다차원으로 구성된 척도인 것처럼 보이는 것이 아닌지 테스트할 수 있다. 본 논문에서는 척도의 차원성을 연구할 때 매우 유용하게 사용될 수 있고 쌍요인 모형 적합 결과부터 도출될 수 있는 모형 기반 신뢰도인 오메가 계수(ω coefficients)와 일차원 검정을 분명하게 할 수 있는 일차원 검정 지수인 ECV(Explained Common Variance)를 중심으로 쌍요인 모형의 유용성을 개관하고자 한다. 이들 지수들을 어떻게 계산하고 해석하는지 예시하기 위하여 본 개관 논문에서는 Rosenberg 자아존중감 척도와 정서접근적 대처 척도(Emotional Approach Coping Scale)를 사용하였다. 확인적 요인분석 결과 Rosenberg 자아존중감 척도의 경우 쌍요인 모형의 적합도가 가장 우수하며 오메가 계수와 일차원 지수(ECV) 값을 모두 고려할 때 방법요인으로부터 설명되는 개인차는 매우 적고 일반요인의 영향력이 가장 커 단일 차원으로 보는 것이 타당한 것으로 밝혀졌다. 정서접근적 대처 척도의 경우 모형 적합도는 쌍요인 모형이 가장 우수하였으나 오메가 계수와 일차원 지수(ECV) 값을 고려할 때 집단요인에 비해 일반요인의 크기가 상대적으로 지배적이지 않아 2요인 구조로 척도를 이해하는 것이 이론적․해석적 측면에서 용이한 것으로 나타났다. 이를 통해 쌍요인 모형 적용 시 적합도 뿐 아니라 도출할 수 있는 여러 지수를 통해 보다 명확한 척도의 차원성을 밝힐 수 있음을 제시하였다.

참고문헌 (Reference)

1 이순묵, "한국형 일상우울의 예비검사 개발" 107-, 2016

2 송승훈, "한국판외상 후 성장 척도(K-PTGI)의 신뢰도와타당도" 252-253, 2006

3 이미숙, "한국판 CES-D 척도(the Center for Epidemiologic Studies-Depression Scale)의 요인구조분석에 대한 재검토" 한국보건사회학회 12 : 43-62, 2002

4 허만세, "한국어판 축약형 CES-D 척도의 측정불변성 검증" 한국정신보건사회복지학회 43 (43): 313-339, 2015

5 이훈진, "편집증 척도의 신뢰도, 타당도 연구" 14 (14): 83-94, 1995

6 전겸구, "통합적 한국판CES-D 개발" 6 (6): 59-76, 2001

7 이미리, "청소년기 자아존중감과 가족, 친구, 학업, 여가, 직업 변인들의 관계: 긍정적 자아평가와 부정적 자아평가를 중심으로" 한국청소년정책연구원 16 (16): 263-293, 2005

8 강성록, "정서접근적 대처 척도의 타당화 예비연구: 사관학교 생도 표본에서" 한국임상심리학회 34 (34): 455-475, 2015

9 고려대학교 부설 행동과학연구소, "심리척도핸드북" 학지사 1999

10 최수미, "부정문항이 포함된 척도의 요인구조 및 방법효과 검증과 남녀 간의 차이 비교: Rosenberg 자기존중감 척도를 중심으로" 한국심리학회 32 (32): 571-589, 2013

11 홍세희, "부정문항이 포함된 검사의 요인구조: 자아존중감 검사의 예" 한국교육평가학회 24 (24): 713-732, 2011

12 정병삼, "부모-자녀애착과 부모지도감독이 청소년의 자아존중감의 변화에 미치는 종단적 영향" 한국청소년정책연구원 21 (21): 5-30, 2010

13 Horan, P. M., "Wording effects in self-esteem scales : Methodological artifact or response style?" 10 (10): 435-455, 2003

14 Rhemtulla, M., "When can categorical variables be treated as continuous? A comparison of robust continuous and categorical SEM estimation methods under suboptimal conditions" 17 (17): 354-373, 2012

15 Cortina, J. M., "What is coefficient alpha? An examination of theory and applications" 78 (78): 98-104, 1993

16 Stucky, B. D., "Using logistic approximations of marginal trace lines to develop short assessments" 37 (37): 41-57, 2013

17 Schmitt, N., "Uses and abuses of coefficient alpha" 8 (8): 350-, 1996

18 Reise, S. P., "The role of the bifactor model in resolving dimensionality issues in health outcomes measures" 16 : 19-31, 2007

19 Reise, S. P., "The rediscovery of bifactor measurement models" 47 : 667-696, 2012

20 Tedeschi, R. G., "The posttraumatic growth inventory : Measuring the positive legacy of trauma" 9 (9): 455-471, 1996

21 Fox, K. R., "The physical self-perception profile : Development and preliminary validation" 11 (11): 408-430, 1989

22 Murray, A. L., "The limitations of model fit in comparing the bi-factor versus higher-order models of human cognitive ability structure" 41 (41): 407-422, 2013

23 Ten Berge, J. M., "The greatest lower bound to the reliability of a test and the hypothesis of unidimensionality" 69 (69): 613-625, 2004

24 Corwyn, R. F., "The factor structure of global self-esteem among adolescents and adults" 34 : 357-379, 2000

25 Holzinger, K. J., "The bi-factor method" 2 (2): 41-54, 1937

26 McKay, M. T., "The Rosenberg Self-Esteem Scale: A bifactor answer to a two-factor question?" 96 : 654-660, 2014

27 Gomez. R., "The Center for Epidemiologic Studies Depression Scale : Support for a bifactor model with a dominant general factor and a specific factor for positive affect" 22 (22): 351-360, 2015

28 Muthén, B. O., "Testing structural equation models" Sage 205-234, 1993

29 McDonald, R. P., "Test theory: A unified treatment" Erbaum 1999

30 Bollen, K., "Structural Equations with Latent Variables" Wiley & Sons 1989

31 Rosenberg, M., "Society and adolescent self-image" Princeton University Press 1965

32 전병재, "Self-esteem: A test of its measurability" 11 : 109-129, 1974

33 Reise, S. P., "Scoring and modeling psychological measures in the presence of multidimensionality" 95 : 129-140, 2013

34 이자영, "Rosenberg의 자아존중감 척도: 문항수준 타당도분석" 한국상담심리학회 21 (21): 173-189, 2009

35 Tomas, J. M., "Rosenberg's self‐esteem scale : Two factors or method effects" 6 (6): 84-98, 1999

36 Raykov, T., "Reliability of scales with general structure : Point and interval estimation using a structural equation modeling approach" 9 (9): 195-212, 2002

37 Dunbar, M., "Question wording effects in the assessment of global self-esteem" 16 : 13-19, 2000

38 Marsh, H. W., "Positive and negative global self-esteem: A substantively meaningful distinction or artifactors?" 70 : 810-819, 1996

39 Sijtsma, K., "On the use, the misuse, and the very limited usefulness of Cronbach’s alpha" 74 (74): 107-120, 2009

40 Muthén, L. K., "Mplus User’s Guide" Muthén &Muthén 2010

41 March, J. S., "Manual for the Multidimensional Anxiety Scale for Children (MASC)" Multi-Health Systems 1998

42 Tavakol, M., "Making sense of Cronbach's alpha" 2 : 53-55, 2011

43 Marsh, H. W., "Longitudinal tests of competing factor structures for the Rosenberg Self-Esteem Scale : Traits, ephemeral artifacts, and stable response styles" 22 : 366-381, 2010

44 Gana, K., "Longitudinal factorial invariance of the Rosenberg Self-Esteem Scale : Determining the nature of method effects due to item wording" 47 : 406-416, 2013

45 Reise, S. P., "Is the bifactor model a better model or is it just better at modeling implausible responses? Application of iteratively reweighted least squares to the Rosenberg Self-Esteem Scale" 51 (51): 818-838, 2016

46 Stucky, B. D., "Handbook of item response theory modeling:Applications to typical performance assessment" Routledge/Taylor & Francis Group 183-206, 2014

47 DiStefano, C., "Further investigating method effects associated with negatively worded items on self-report surveys" 13 : 440-464, 2006

48 Dunn, T. J., "From alpha to omega : A practical solution to the pervasive problem of internal consistency estimation" 105 : 399-412, 2014

49 Supple, A. J., "Factor structure of the Rosenberg Self-Esteem Scale" 44 : 748-764, 2013

50 McDonald, R. P., "Factor analysis and related methods" Erlbaum 1985

51 Donnellan, M. B., "Extending structural analyses of the Rosenberg Self-Esteem Scale to consider criterion-related validity : Can composite self-esteem scores be good enough?" 98 (98): 169-177, 2016

52 Asparouhov, T., "Exploratory structural equation modeling" 16 (16): 397-438, 2009

53 Gu, H., "Examining and controlling for wording effect in a self-report measure : A Monte Carlo simulation study" 24 (24): 545-555, 2017

54 Rodriguez, A., "Evaluating bifactor models : Calculating and interpreting statistical indices" 21 (21): 137-150, 2016

55 Raykov, T., "Estimation of reliability for multiple-component measuring instruments in hierarchical designs" 12 (12): 536-550, 2005

56 Raykov, T., "Estimation of congeneric scale reliability using covariance structure analysis with nonlinear constraints" 54 (54): 315-323, 2001

57 Raykov, T., "Estimation of composite reliability for congeneric measures" 21 (21): 173-184, 1997

58 Zinbarg, R. E., "Estimating generalizability to a latent variable common to all of a scale's indicators : A comparison of estimators for ωh" 30 (30): 121-144, 2006

59 Hu, L., "Cutoff criteria for fit indexes in covariance structure analysis : Conventional criteria versus new alternatives" 6 (6): 1-55, 1999

60 Zinbarg, R. E., "Cronbach’s α, Revelle’s β, and McDonald’s ω H : Their relations with each other and two alternative conceptualizations of reliability" 70 (70): 123-133, 2005

61 Quilty, L. C., "Correlates of the Rosenberg Self-Esteem Scale method effects" 13 (13): 99-117, 2006

62 Stanton, A. L., "Coping through emotional approach : Scale construction and validation" 74 : 1078-1092, 2000

63 Bentler, P. M., "Comparative fit indexes in structural models" 107 : 238-246, 1990

64 Yang, Y., "Coefficient alpha: A reliability coefficient for the 21st century?" 29 (29): 377-392, 2011

65 Revelle, W., "Coefficient alpha, beta, omega, and the glb : Comments on Sijtsma" 74 : 145-154, 2009

66 배성우, "CES-D 척도(The Center for Epidemiologic Studies-Depression Scale)의 요인구조 분석: 확인적 요인분석 방법의 적용" 한국보건사회학회 18 : 134-155, 2005

67 신재은, "Bifactor 모형을 적용한 CES-D 척도의 차원성검증"

68 Chung, C., "Bifactor approach to modeling multidimensionality of physical self-perception profile" 20 (20): 1-15, 2016

69 Reise, S. P., "Bifactor Models and rotations : Exploring the extent to which multidimensional data yield univocal scale scores" 92 : 544-559, 2010

70 Bagozzi, R. P., "Assessing construct validity in personality research : Applications to measures of self-esteem" 27 : 49-87, 1993

71 Rodriguez, A., "Applying bifactor statistical indices in the evaluation of psychological measures" 98 (98): 223-237, 2016

72 Wu, C. H., "An examination of the wording effect in the Rosenberg Self-Esteem Scale among culturally Chinese people" 148 (148): 535-552, 2008

73 Reise, S. P., "APA handbook of testing and assessment in psychology: Vol. 1. Test theory and testing and assessment in industrial and organizational psychology" American Psychological Association 101-119, 2013

74 Akaike, H., "A new look at the statistical model identification" 19 : 716-723, 1974

75 Lance, C. E., "A critique of the correlated trait-correlated method and correlated uniqueness models for multitrait-multimethod data" 7 (7): 228-, 2002

76 Thege, B. K., "A bifactor model of the Posttraumatic Growth Inventory" 2 (2): 529-540, 2014