경영학 연구에서의 구조방정식 모형의 적용: 문헌 연구와 비판 (Applications of Structural Equation Modeling inManagement Studies: A Critical Review)

서지정보

· 발행기관 : 한국경영학회
· 수록지 정보 : 경영학연구 / 36권 / 4호 / 897 ~ 923페이지
· 저자명 : 김진호, 홍세희, 추병대

초록

최근 경영학 연구에서는 잠재변수간의 관계가 복잡한 모형을 측정오차를 통제하면서 통계적으로 모형을 평가할 수 있다는장점 때문에 구조방정식 모형(structural equation modeling: SEM)을 활용하는 논문이 증가추세에 있으며 이 통계기법은 앞으로도 상당 기간 동안 경영학의 많은 영역에서 다양한 문제들을 조사하고 해결하는 데에 유용하게 활용되리라 예상된다. 본 연구에서는 지금까지 <경영학 연구>에 게재된 총 97편의 SEM 논문을 조사하여 모형 개발, 모형 추정, 모형평가의 세 영역에서 중요한 방법론적 쟁점들이 어떻게 다루어지고 있는지를 분석하였다. 이러한 분석을 바탕으로 경영학실증연구에 나타난 SEM 활용의 전반적인 문제점과 개선방향을 검토함으로써 앞으로 SEM을 이용하고자 하는 연구자들에게 바람직한 SEM 활용을 위한 가이드라인을 제시하였다.

영어초록

Since the 1970s, the development of structural equation modeling(SEM) methods and software has proceeded rapidly. Currently SEM has become a very popular research tool in management studies and the number of SEM papers published in the Korean Management Review clearly shows the increasing popularity of SEM. Only two SEM papers were published in the Korean Management Review up to 1995. However, since 2000, 80 SEM papers were published. In other words, SEM papers take about 18% of all the articles published in the Korean Management Review during that period. In fact, SEM methodology has much to offer to the empirical researchers, but there are many pitfalls that can make SEM a dangerous tool in the hands of inexperienced users. Applications of methodologies, particularly if they are as complex as SEM, are prone to misuse and thus an in-depth analysis of a critical mass of empirical studies is needed to assess negative aspects of the applied SEM papers and suggest avenues for improvement. The purpose of this paper is to critically evaluate previous empirical applications of SEM in the Korean Management Review and to provide guidance to future users on how to employ the methodology more appropriately. Specifically we have documented the number of applications of SEM between 1993 and present and classified these applications in terms of relevant criteria such as the purpose for which SEM is used. Also, we have tried to evaluate the quality of applications of SEM by assessing their conformance with formal statistical assumptions required for the valid use of SEM methodology, their adherence to guidelines derived empirically from simulation studies. To our knowledge, no comprehensive survey of SEM applications has been reported in any Korean journals. Building off Baumgartner & Homburg(1996), MacCallum & Austin(2000), and Holbert & Stephenson(2002), we document and characterize the number of application published in the Korean Management Review. Then, we discuss important methodological issues related to SEM and assess the quality of previous applications in terms of three aspects: issues related to model specification, issues related to model estimation, and issues related to model evaluation. On the basis of our finding, we identify problem areas, provide general guidelines for the use of SEM, and make recommendations concerning data screening, non-recursive modeling, cross-sectional data, model identification, degree of freedom, reporting parameter estimates, model fit indices, model modification and comparison, cross-validation, and decomposition of effects.

참고자료

· 없음