유통과학분야에서 탐색적 연구를 위한 요인분석

임명성 한국유통과학회 2015 유통과학연구 Vol.13 No.9

Purpose – This paper aims to provide a step-by-step approach to factor analytic procedures, such as principal component analysis (PCA) and exploratory factor analysis (EFA), and to offer a guideline for factor analysis. Authors have argued that the results of PCA and EFA are substantially similar. Additionally, they assert that PCA is a more appropriate technique for factor analysis because PCA produces easily interpreted results that are likely to be the basis of better decisions. For these reasons, many researchers have used PCA as a technique instead of EFA. However, these techniques are clearly different. PCA should be used for data reduction. On the other hand, EFA has been tailored to identify any underlying factor structure, a set of measured variables that cause the manifest variables to covary. Thus, it is needed for a guideline and for procedures to use in factor analysis. To date, however, these two techniques have been indiscriminately misused. Research design, data, and methodology – This research conducted a literature review. For this, we summarized the meaningful and consistent arguments and drew up guidelines and suggested procedures for rigorous EFA. Results – PCA can be used instead of common factor analysis when all measured variables have high communality. However, common factor analysis is recommended for EFA. First, researchers should evaluate the sample size and check for sampling adequacy before conducting factor analysis. If these conditions are not satisfied, then the next steps cannot be followed. Sample size must be at least 100 with communality above 0.5 and a minimum subject to item ratio of at least 5:1, with a minimum of five items in EFA. Next, Bartlett's sphericity test and the Kaiser-Mayer-Olkin (KMO) measure should be assessed for sampling adequacy. The chi-square value for Bartlett's test should be significant. In addition, a KMO of more than 0.8 is recommended. The next step is to conduct a factor analysis. The analysis is composed of three stages. The first stage determines a rotation technique. Generally, ML or PAFwill suggest to researchers the best results. Selection of one of the two techniques heavily hinges on data normality. ML requires normally distributed data; on the other hand, PAF does not. The second step is associated with determining the number of factors to retain in the EFA. The best way to determine the number of factors to retain is to apply three methods including eigenvalues greater than 1.0, the scree plot test, and the variance extracted. The last step is to select one of two rotation methods: orthogonal or oblique. If the research suggests some variables that are correlated to each other, then the oblique method should be selected for factor rotation because the method assumes all factors are correlated in the research. If not, the orthogonal method is possible for factor rotation. Conclusions – Recommendations are offered for the best factor analytic practice for empirical research.

요인 개수 결정을 위한 평행분석의 정확성 평가

임상돈,장승민 한국심리학회 2017 한국심리학회지 일반 Vol.36 No.4

Parallel analysis is a method of estimating the number of factors by comparing the eigenvalues ​​of sample data with the eigenvalues ​​of random data. This method is considered to be theoretically more valid and empirically more accurate in estimating the number of factors than other methods, such as Kaiser method and scree test, that estimate the number of factors based on the eigenvalues. However, several criticisms have been raised about the validity of the rationale for parallel analysis and various modifications have been proposed. There have also been concerns about the conditions under which parallel analysis shows relatively low accuracy. The current study examined the rationale and limitations of the use of eigenvalues ​​and parallel analysis to estimate the number of factors, and based on this, we specified the conditions under which the accuracy of parallel analysis may be low. We also examined, through a simulation, the effects of various factors that may affect the accuracy of parallel analysis and confirmed the conditions where cautions are needed when applying parallel analysis. The results of the simulation show that the accuracy of estimating the number of factors in the parallel analysis is greatly influenced by the size of factor correlations, the magnitude of factor loadings, the number of factors, and the number of variables per factor. In addition, we confirmed that the accuracy of the parallel analysis is significantly lower when a factor model includes a weak factor with low factor loadings. Overall, the accuracy of the parallel analysis for the reduced correlation matrix (PA-PAF) was higher than the parallel analysis for the correlation matrix (PA-PCA), which in particular, PA-PAF showed high accuracy when factor correlations were high, and PA-PCA showed high accuracy when factor correlations were low. Based on the results of the simulation analyses, we proposed sample sizes required for parallel analysis to provide accuracy of 90% or higher under conditions with different levels of factor correlation, factor loading, and the number of factors. 평행분석은 표본 자료의 고윳값과 무선 자료의 고윳값을 비교하여 요인의 개수를 추정하는 방법이다. 이 방법은 고윳값에 근거하여 요인의 개수를 추정하는 다른 절차들(카이저 방법, 스크리 검사)보다 이론적으로 더 타당한 근거를 갖고 있고 경험적으로도 요인의 개수를 더 정확히 추정하는 것으로 평가 받는다. 그러나 평행분석의 이론적 근거의 타당성에 대해서도 여러 비판이 제기되어 왔고 이에 따른 다양한 수정 절차들도 제안되었다. 또한 평행분석이 상대적으로 낮은 정확성을 나타내는 조건들에 대한 우려도 있어 왔다. 본 연구는 고윳값과 평행분석이 요인의 개수를 추정하는 데 사용될 수 있는 이론적 근거와 한계를 검토하고 이를 바탕으로 평행분석의 정확성이 낮게 나타날 수 있는 조건들을 구체화하였다. 또한 모의실험을 통해 평행분석의 정확성에 영향을 줄 수 있는 다양한 요인들의 효과를 검토하고 평행분석의 사용에 주의를 요하는 조건들을 확인하였다. 모의실험의 결과는 평행분석의 요인수 추정 정확률이 요인상관의 크기, 요인부하량의 크기, 요인의 개수, 요인당 변수의 개수에 따라 크게 영향을 받으며 정확률이 낮은 조건에서 표본크기의 영향이 매우 크다는 것을 보였다. 또한 요인부하량이 낮은 변수들로 구성된 약한 요인이 포함된 경우 요인수 추정의 정확률이 크게 낮아짐을 확인하였다. 전반적으로 상관행렬에 대한 평행분석(PA-PCA)보다 축소상관행렬에 대한 평행분석(PA-PAF)의 정확률이 높았으며 특히 요인상관이 높은 경우에는 PA-PAF가, 요인상관이 낮은 경우에는 PA-PCA가 높은 정확률을 보이는 경향이 확인되었다. 마지막으로 모의실험의 결과를 기초로 요인상관의 크기, 요인부하량의 크기, 요인의 개수의 조합으로 구성되는 다양한 조건에서 평행분석이 90% 이상의 정확률을 제공하기 위해 요구되는 표본크기를 제안하였다.

체육학에서 요인분석의 통계적 타당화: 탐색적, 확인적 요인분석을 중심으로

정연택,최연재 한국코칭능력개발원 2023 코칭능력개발지 Vol.25 No.5

요인분석은 체육학에서 널리 이용되는 통계기법이며 사용 빈도가 높다. 요인분석은 변수를 축소하거나 제거하여 측정하는 항목의 타당성을 검증한다. 특히, 요인분석은 측정된 자료를 이용하여 변수 생성 등의 목적을 가진다. 따라서 이 연구는 체육학에서 널리 사용하고 있는 분석의 개념과 탐색적, 확인적 요인분석에 대하여 이해하고자 하였다. 1장에서는 요인분석의 특성과 중요성에 관하여 서술하였으며, 2장은 요인분석의 개념과 기본과정에 관한 서술이 이루어졌다. 3장에서는 분석 절차 중 탐색적, 확인적 요인분석의 중요도를 설명하였다. 4장에서는 앞서 설명된 요인분석을 바탕으로 논의를 하였으며, 5장은 통계분석의 합리적 방안을 제시하였다. Factor analysis is a statistical technique widely used in physical education and is frequently used. Factor analysis verifies the validity of items to be measured by reducing or removing variables. In particular, factor analysis has the purpose of creating variables using measured data. Therefore, this study tried to understand the concept of factor analysis widely used in physical education and exploratory and confirmatory factor analysis. Chapter 1. describes the characteristics and importance of factor analysis, and Chapter 2. describes the concept and basic process of factor analysis. Chapter 3. explained the importance of exploratory factor analysis and confirmatory factor analysis among the procedures of factor analysis. In Chapter 4. discussions were made based on the factor analysis described above, and in Chapter 5. rational methods for statistical analysis were presented.

축구 경기력 분석 평가를 위한 개념구조 탐색

이용수(Lee, Young-Soo),황보관(Hwang, Bo-Kwan),김용래(Kim, Yong-Rae) 한국체육과학회 2013 한국체육과학회지 Vol.22 No.5

The purpose of this study was to explore the soccer performance analysis assessment factors which coaches actually want information on and professional soccer players in need to develop the accompanying scale for them. For the purpose of exploring the soccer performance analysis assessment factors and develop the scale for them, the open-ended questionnaire was established, and the pilot question items were developed for the soccer performance analysis assessment. Then item analysis, exploratory factor analysis and reliability analysis were conducted for 123 coaches and professional soccer players at the state of questionnaire research. As a result, total of 24 items were drawn in regard to tour factors such as the physical strength factor, the technical factor, the tactical factor and the behavioral (psychological) factor. Cronbach’ a was found to be .810~.888(total .906). And confirmatory factor analysis was conducted to investigate the goodness of fit of the soccer performance analysis assessment model concerning the final question item developed through the item analysis and exploratory factor analysis. As a result, four factors(physical strength factor, technical factor, tactical factor and behavioral factor) were structured, and the goodness of fit indexes(RMR, GFI, RMSEA, RLI, GFI) satisfied the criterion for the optimal value, and the model was tested to be valid. The result could assist providing further information for the soccer performance analysis assessment factor wanted by coaches and professional soccer players.

한국형 여가만족척도의 다집단(Multi-Group)분석과 잠재평균

김영재(YoungJaeKim) 한국체육학회 2009 한국체육학회지 Vol.48 No.5

본 연구의 목적은 일반적으로 여가활동을 통해 느끼는 만족을 측정할 수 있는 검사지를 개발하는데 있다. 본 연구에서는 여가만족의 과정을 심층적으로 탐색한 김영재(2004)의 선행연구를 토대로 그에 따른 하위 요인(5요인)으로 구성된 모형을 측정모형 분석과 함께 다집단 분석(multi-group analysis)을 통해 형태 동일성, 측정 동일성, 절편 동일성을 검증하여 한국형 여가만족 척도가 남녀 대학생 집단에게 공통적으로 사용될 수 있는가를 제시하고, 잠재평균 분석(Latent Means Analysis;LMA)을 통하여 여가만족 요인에 대한 성별집단의 잠재평균 차이를 비교 검증고자 했다. 본 연구의 대상자는 서울, 경기지역에 소재하고 있는 대학생 441명(남=261명, 여=180명)을 대상으로 조사하였다. 본 연구결과 여가만족척도의 다집단 분석(multi-group analysis)결과 형태 동일성, 측정 동일성, 절편 동일성 검증되어 여가만족 척도가 남여 일반대학생 집단에게 공통적으로 사용될 수 있음을 확인하였다. 또한 5가지의 여가만족요인을 성별에 따라 잠재평균 분석을 실시하여 그 차이점을 살펴본 결과 여자가 남자에 비해 사회적 만족과 신체적 만족요인이 더 높은 것으로 나타났다. The purpose of this study was to investigate leisure satisfaction scale construct validity and to verify change of factor pattern with gender using multi-group factor invariance analysis. To carry out this research, used 441 university students(male261, female 180). The data from this study were analyzed using SPSS 15.0 for common factor analysis of exploratory factor analysis; using AMOS 7.0 for confirmatory factor analysis and chi-square test of multi-group analysis. The results of this study were as following; first, the configural, metric, partial scalar invariance were satisfied for multi-group confirmatory factor analysis. Second, according to the latent mean analysis, social satisfaction factor latent mean was .22, physical satisfaction factor was .73, emotion satisfaction factor was .01, surrounding satisfaction factor was -.13, and education satisfaction factor was ―.09. Third, when it comes to effect size; social satisfaction factor .27, physical satisfaction factor .67, emotion satisfaction factor .01, surrounding satisfaction factor .19 and education satisfaction factor was .18.

근골격계질환 예방을 위한 유해요인조사 결과보고 제도의 실효성 방안에 대한 연구

이신우(Shinwoo Lee),김유창(Yuchang Kim) 대한인간공학회 2020 大韓人間工學會誌 Vol.39 No.4

Objective: The purpose of this study is to understand the appropriateness and disadvantages of the result reporting system about risk factors analysis and to find out how to improve the effectiveness of the result reporting system. Background: Although the risk factors analysis for preventing musculoskeletal disorders was conducted since 2003, the implementation rate was low in Korea. Recently, as the legal effectiveness of the risk factors analysis has been lowed due to the low implementation rate, the issue of legal obligation the reporting system about risk factors analysis has been raised. In order to protect all workers from musculoskeletal disorders, it was necessary to find out how to increase the implementation rate by legally mandating the result reporting system about risk factors analysis. Method: This study was conducted a questionnaire of 126 safety/health managers, workers, and management supervisors, safety/health agency workers, and business owners in various industries. The main contents of the questionnaire are appropriateness, advantages, disadvantages, and effectiveness of the result reporting system about risk factors analysis. Results: The results showed that the reliability of risk factors analysis would be increased by the implementation of the result reporting system. The cost support system about risk factors analysis would be needed because of the cost problem on the corporation. Conclusion: It was judged that plan for enforcing the legal obligation of the reporting system would be needed to enhance the effectiveness of risk factor analysis. In order to increase the reliability of the risk factors analysis, it was necessary to report to the Minister of Employment and Labor after confirming by worker representative. Also, it was considered that a cost support system would be necessary to activate the result reporting system about risk factor analysis. Application: These results can contribute to the improvement of the risk factor analysis and the prevention of musculoskeletal disorders. It can be used as the basic data for research on the result reporting system about risk factors analysis.

체육수업 재미거리 질문지(PCESQM) 타당성 및 신뢰성 검증

박준성(Jun Sung Park) 한국사회체육학회 2009 한국사회체육학회지 Vol.0 No.37

This paper verified the validity of the structure and reliability whether the questionnaire on the source of enjoyment in physical education classes for middle school students developed by Chang-sub Lee and Sang-woo Nam(2003) can effectively measure the fun factors in the physical education classes for the male and female students(592) of middle schools in S city in 2008. As a result of the item analysis and the reliability analysis, since questions with a mean over 4.5 and questions with the skewness and kurtosis values over ±2.0 did not appear, the results of this paper showed to have an appropriate level of the mean, standard deviation, skewness, kurtosis, and reliability. Furthermore, as the results of the correlation matrix between questions, show that since more than 50% of the questions have a correlation over ±3.0 between each question, the question of this paper are judged to be adequate questions for the factor analysis. Moreover, in the exploratory factor analysis, the maintenance of physical and mental health factor was developed into questions number 2, 3, 5, 6, 7, sense of accomplishment factor to questions number 1, 4, 12, the easiness of the class factor to questions number 9, 10, 15, the freedom of the class factor to questions number 13, 14, 16, and the sociability cultivation factor to questions number 8, 11, 17 that are the sub-factor of the questionnaire on the source of fun in physical education classes for middle school students developed by Chang-sub Lee and Sang-woo Nam(2003). Also in the exploratory factor analysis of this paper, as the questions related to factor 1 showed to be questions number 6, 7, 5, 2, 3, factor 2(1, 4, 12), factor 3(10, 9, 15), factor 4(14, 13, 16), and factor 5(17, 8, 11), it indicates that is composed of the five sub-factors developed by Chang-sub Lee and Sang-woo Nam(2003). For the confirmatory factor analysis that assesses the fitness of the model, as the Q value was 2.004, GFI .910, AGFI .897, CFI .884, RMSEA .71, and RMR .51, it is judged that the model is appropriate. As a result of evaluating the fitness of the model through the item analysis, correlation analysis, exploratory factor analysis, and the confirmatory factor analysis as mentioned above, the questionnaire on the source of enjoyment in physical education classes for the middle school students developed by Chang-sub Lee and Sang-woo Nam(2003) showed to be an appropriate questionnaire in measuring the enjoyment factors for the male and female students of middle schools in S city.

직무만족 요인분석에 관한 연구 - 한국고용정보원 고용패널 데이터 활용을 중심으로

김호원,오성욱,이재춘 한국취업진로학회 2015 취업진로연구 Vol.5 No.2

본 연구는 직무만족도 변수들의 요인을 분석하기 위하여 한국고용정보원 2005 대졸자 직업이동경로 조사 1차년도 부터 3차년도 자료, 2009 대졸자 직업이동경로조사 1차년도 그리고 고졸자 직업이동경로 조사와 청년패널 자료를 활용하여 분석하였다. 왜냐하면 직무만족을 구성하고 있는 다양한 요인들이 존재하고 있음에도 불구하고 GOMS(대졸자이 동경로조사), HSGOMS(고졸자이동경로조사) 그리고 YP(청년패널) 자료를 기초로 한 다수의 보고서에 서 직무만족도 요인분석 과정에서 합리적이지 못한 부분이 적용되고 있었기 때문이다. 선행 연구들에서 직무만족도는 최소한 내재적 요인과 외재적 요인으로 구분되어 있다는 사실을 확 인하였다. 그러나 본 연구에서는 한국고용정보원 패널 데이터 분석 결과 탐색적 요인 분석에서는 직무 만족도 변수를 내재적 요인과 외재적 요인으로 구분할 수 없었으며, 확인적 요인분석을 통하여 두 개의 요인을 구분할 수 있었다. 그리고 고용패널 선행연구들에서 많이 사용한 전체 변수들을 분석에 이용한 경우를 또 다른 측정 모델로 구성하여 모두 5개의 측정 모델을 구성하였으며, 확인적 요인분석을 통해 모형 적합도를 확인한 결과, 2005 GOMS I, II, III과 2009 GOMS I 모두 일관되게 “인간관계”항목을 내 재적 요인에 포함하여 “직무내용”, “하고 있는 일의 자율성과 권한”, “개인의 발전가능성”항목으로 내재 적 요인을 구성한 모델(2009 GOMS I_3, 2005 GOMS III_3, 2005 GOMS II_3, 2005 GOMS I_3)이 가 장 우수한 적합도 수치를 나타내었다. 따라서 한국고용정보원 패널 데이터로 직무만족도에 대한 구체적인 연구를 진행하고자 하는 경우 탐색적 요인분석을 통하여 직무만족도 요인을 분류하기 보다는 그동안 선행연구를 기초로 내재적 요인 과 외재적 요인으로 구분한 후 확인적 요인분석을 실시하여 내재적 만족도와 외재적 만족도를 구분하 여 분석하는 것이 필요하다. 특히 연구주제가 직무만족도에 초점을 맞추게 되는 경우 본 연구에서 제안 하는 방법을 고려해 볼 필요성이 존재한다고 할 수 있다. In order to analyzes the factors of job satisfaction variables, This study were analyzed by 2005 GOMS I∼III, 2009 GOMS I, HSGOMS and YP employment panel data. Because despite the fact that there are various factors that make up the job satisfaction, not rational part was being applied in job satisfaction factor analysis from a number of reports on the basis of the GOMS(graduates flyway research), HSGOMS(high school graduates flyway research) and YP (Youth Panel) data . At least, In previous studies confirmed that the intrinsic job satisfaction factors and the extrinsic job satisfaction factors were separated. In this study, however, In that case of Korea Employment Information Service panel data analysis, exploratory factor analysis could not be separated by the intrinsic job satisfaction factors and the extrinsic job satisfaction factors. Just through confirmatory factor analysis was able to distinguish between the two factors. For the relationship of job satisfaction variables prior research has included extrinsic factors, if the configuration of relationships, relationships with colleagues and superiors supervision. However, in the case of Korea Employment Information Institute panel data the relationships is ambiguity whether the relationship is job duties intrinsic relationships or job duties external relationships Therefore, for confirmatory factor analysis three model was constructed. The first model included variables intrinsicl relationships. The second model is included a human relations for extrinsicl variables. The third model did not include any elements in the relationship. In addition, the development potential of individual variables was composed of more than a model in the intrinsic factor. Because through previous research the development of individual potential variables could not be sure that the intrinsic or extrinsic factors. And finally, the fifth confirmatory factor analysis model is measured using the whole lot of variables used in previous studies to analyze employment panel. And through confirmatory factor analysis confirmed the model fitness. The result 2005 GOMS I, II, III and 2009 GOMS I consistently “human relationships”, “job description”, “autonomy and authority of the work, “individuals development potential factors” the configured model intrinsic to the items (2009 GOMS I_3, 2005 GOMS III_3, 2005 GOMS II_3, 2005 GOMS I_3) showed the best fit value.Thus, when you want to proceed to a detailed study on job satisfaction by the KEIS panel data, on the basis of previous studies and to distinguish between the intrinsic job satisfaction factors and the extrinsic job satisfaction factors, it is necessary to analyze a confirmatory factor analysis. Research topic, especially when focused on job satisfaction in this study to suggest that there is a need to consider how you can. The present study is significant that proposed a way for a more systematic analysis of job satisfaction variables in that has been used indiscriminately through the analysis of research data from existing panel data.

GGobi를 이용한 한중일 청소년의 가치관에 관한 요인분석

오정아,이은경 한국자료분석학회 2013 Journal of the Korean Data Analysis Society Vol.15 No.2

Factor analysis is a widely used multi-variate data analysis method in sociology and psychology area. However the result of exploratory factor analysis depends on the initial factor extract method, number of factors, and method of rotation. It is not easy to choose the right options for the factor analysis in specific data set and we need to be careful to choose these options. In this paper, we use dynamic graphical methods to explore the result from factor analysis for the right options of the factor analysis. We explore and compare the results from the factor analysis with various options. Also we analyze the results from different population. We apply 2008 survey on the value of teenagers in three countries to our methods, find the optimal factor analysis model, and compare the result from three countries using dynamic graphical method. In the importance in life part, the result of the exploratory factor analysis on Korea is a little bit different from the results of the other two countries. On the other hands, the results of Japan in the reliability in family, society, and nation are quite different from the results of Korea and China. This result is verified from the result of the confirmatory factor analysis. 요인분석은 사회학, 심리학 관련분야에서 활발하게 사용하고 있는 다변량 통계 분석 방법이다. 그러나 요인분석 중 탐색적 요인분석은 초기 요인추출방법, 공통요인의 수, 회전 방법 등을 분석자가 주관적으로 결정하는 경향이 있으나 이들의 선택에 따라 결과가 크게 달라질 수 있다. 그러므로 최종 요인모형을 선택하기 위해 요인분석의 결과를 면밀히 살펴보아야 할 필요가 있다. 본 연구에서는 요인분석에서 최적의 요인 수, 초기 추출방법, 회전 방법 등을 선택하기 위하여 각 조건 하에서의 요인분석결과를 동적그래프 방법을 이용하여 살펴보고 이를 비교, 분석하는 방안을 제시하였다. 또한 실제 자료인 한, 중, 일 청소년 가치관에 대한 설문조사자료 중 인생에서의 중요도와 가족, 사회, 그리고 국가에 대한 신뢰도에 관한 항목들을 이용하여 요인분석을 실시하고 본 논문에서 제안한 방법을 이용하여 최적의 요인수와 초기추출방법, 그리고 회전 방법을 찾고 요인분석을 실시한 후 각 나라마다의 결과를 비교, 분석하였다. 인생에서의 중요도의 경우 중국과 일본이 한국과는 다른 요인분석 결과를 나타내고 있으나 차이가 크지는 않았다. 반면 가족, 사회, 그리고 국가에 대한 신뢰도의 경우 일본이 한국과 중국의 결과와 차이를 보이고 있으며 이 차이는 다소 크게 나타나고 있다. 이 결과는 확증적 요인분석을 통하여서도 확인할 수 있었다.

투자기회집합을 어떻게 측정할 것인가?

홍철규 ( Cheol Kyu Hong ) 한국회계학회 2012 회계학연구 Vol.37 No.3

본 연구의 목적은 우리나라 자료를 이용하여 기존 문헌에 등장하는 주요 IOS 후보변수 및 변수군들의 적합성을 분석하여 연구자들이 투자기회집합(IOS) 대리변수를 선택하는데 도움을 주고자 하는 것이다. 기업의 중요한 회계재무정책과 계약비용, 투자기회는 서로 밀접히 연관되어 있다고 알려져 있어, IOS의 측정은 실증연구에서 매우 중요한 요소이다. 그러나, 기업이 직면한 투자기회는 성격상 직접적인 관찰이 불가능하고, 이를 어떻게 측정할 것인가에 대해서도 기존 문헌에서 통일된 의견을 발견할 수 없다. 이로 인해 연구자들을 임시방편적인 선택에 의존하게 하고 있다. 본 연구에서는 각 개별 IOS 후보변수들과 후보변수군들을 대상으로 한 요인분석을 통해 추출한 공통요인들이 IOS 대리변수로서 어느 정도 적합한지를 분석하여 바람직한 대리변수를 제시하고 한다. 연구결과, 제조기업 표본과 비제조기업 표본을 포함한 전체표본을 사용할 경우, 자산 시장-장부가치 비율, R&D지출-총자산 비율, 주가-주당이익 비율 변수군이 가장 적합한 것으로 나타났으며, 이 변수군을 사용하는 것이 단일 후보변수를 사용하는 것보다 더 바람직한 것으로 나타났다. 추가적으로 기업규모별로 4등분하여 실시한 분석에서는 가장 작은 규모의 기업그룹을 제외하고는 대체로 만족스러운 결과를 보이지 않았다. 본 연구는 IOS 측정에 관한 문제를 최초로 심도 있게 다룸으로써, IOS에 대한 이해의 폭을 넓혀주고 대리변수의 선택에 관해 많은 시사점을 제시해 주고 있어, 향후 관련 실증연구에 도움을 줄 것으로 기대된다. This study addresses the choice of an investment opportunity set(IOS) proxy variable from among IOS candidate variables and/or sets of variables, based on the Korean non-financial firms listed in KOSPI and KOSDAQ as of the end of 2009. It has been known that IOS is intricately related with various accounting and financial policies including choice of accounting procedure, capital structure, ownership structure and dividend policies, which are in turn largely involved with contracting costs. The proper measurement of IOS thus is essential for many empirical studies in the areas of accounting and finance. However, IOS faced by firms is unobservable in nature and it seems that no consensus has still emerged in the related areas concerning the choice of an IOS proxy variable or measurement of IOS. Some researchers choose to use a single variable as an IOS proxy variable, while others, recognizing that IOS is imperfectly measured by any single proxy variable, conduct sensitivity tests using various proxy variables or try to derive a factor from a set of observable IOS candidate variables using common factor analysis. However, the existing studies utilizing factor analysis often do not explicitly state why they have decided to choose the set of specific candidate variables for factor analysis, do not conduct a reliability test in terms of whether one meaningful factor can reliably be derived from their factor analysis, or do not provide a validity test in terms of the degree to which the derived factor predicts the latent investment opportunities. Furthermore, none of the existing studies conducts split sample tests, ignoring the fact that the performance of an IOS proxy variable may vary according to sub-sample groups. For these reasons, researchers often rely on the ad hoc choice of an IOS proxy variable, and many of the existing studies may involve measurement problems and some unknown biases in their results. To find an appropriate IOS proxy variable, this study works with seven IOS candidate variables used in Baber et al.(1996): investment intensity, geometric mean annual growth rate of market value of assets, market-to-book value of assets, R&D expenditure to total assets, market-to-book value of equity, earnings-to-price ratio, variance of return on market value. Factor analysis has been conducted for the sets which include more than two IOS candidate variables. A correlation analysis is also carried out between the factors derived from factor analysis/individual candidate variables and three IOS revelation variables to analyse the extent to which they predict firm`s investment opportunities. IOS revelation variables used are investment intensity, growth of book value of assets, revenue growth for the future five years, following the approach by Baber et al. Results of factor analysis are examined to judge whether it is possible to derive one meaningful factor from the chosen sets of variables. Analyses have first been carried out for the full sample, and then for split samples to ensure that the candidate proxy variable works not only for the full sample, but also for the split samples. For the full sample I find that the factors derived from the sets of variables of investment intensity, market-to-book value of assets, R&D expenditure to total assets, and earnings-to-price ratio are more appropriate than the factors derived from the sets of the variables used in existing literature in the light of correlation with IOS revelation variables. I also find that market-to-book value of assets and R&D expenditure to total assets are preferable if the use of single candidate variables is desired. Further, it is found that the use of the factor is preferable to any single candidate variables. For the split samples of manufacturing and non-manufacturing firms, I find that the use of a single variable suggests R&D expenditure to total assets for manufacturing firms and market-to-book value of assets and R&D expenditure to total assets for non-manufacturing firms. The factor from market-to-book value of assets, R&D expenditure to total assets, and earnings-to-price ratio and the factors from the most sets of variables of interest are desired for manufacturing firms and non- manufacturing firms, respectively. Combining these results, I find that for the full sample, R&D expenditure to total assets in case of a single variable, and the factor derived from market-to-book value of assets, R&D expenditure to total assets, and earnings-to-price ratio are desirable. The use of factor is also found to be preferable to that of a single candidate variable. Additionally, analyses based on the split samples according to firm size show that many of the IOS candidate variables and the factors derived from them are highly correlated with IOS revelation variables for firms in the fourth(smallest) quartile of size. However, disappointingly any meaningful relationship does not appear in other quartiles. Further research on the effect of size upon the choice of an IOS proxy variable seems to be required. By providing a comprehensive and in-depth analysis of the measurement of an IOS proxy variable, this study contributes to the enhanced understanding of IOS and helps researchers in the related areas.