http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Introduction to the LIVECAT web-based computerized adaptive testing platform
서동기,최정욱 한국보건의료인국가시험원 2020 보건의료교육평가 Vol.17 No.-
This study introduces LIVECAT, a web-based computerized adaptive testing platform. This platform provides many functions, including writing item content, managing an item bank, creating and administering a test, reporting test results, and providing informationabout a test and examinees. The LIVECAT provides examination administrators with an easy and flexible environment for composingand managing examinations. It is available at http://www.thecatkorea.com/. Several tools were used to program LIVECAT, as follows:operating system, Amazon Linux; web server, nginx 1.18; WAS, Apache Tomcat 8.5; database, Amazon RDMS—Maria DB; and languages, JAVA8, HTML5/CSS, Javascript, and jQuery. The LIVECAT platform can be used to implement several item response theory(IRT) models such as the Rasch and 1-, 2-, 3-parameter logistic models. The administrator can choose a specific model of test construction in LIVECAT. Multimedia data such as images, audio files, and movies can be uploaded to items in LIVECAT. Two scoring methods (maximum likelihood estimation and expected a posteriori) are available in LIVECAT and the maximum Fisher information itemselection method is applied to every IRT model in LIVECAT. The LIVECAT platform showed equal or better performance comparedwith a conventional test platform. The LIVECAT platform enables users without psychometric expertise to easily implement and perform computerized adaptive testing at their institutions. The most recent LIVECAT version only provides a dichotomous item response model and the basic components of CAT. Shortly, LIVECAT will include advanced functions, such as polytomous item response models, weighted likelihood estimation method, and content balancing method.
서동기,Jae Kum Kim 한국보건의료인국가시험원 2021 보건의료교육평가 Vol.18 No.-
Purpose: Diagnostic classification models (DCMs) were developed to identify the mastery or non-mastery of the attributes required for solving test items, but their application has been limited to very low-level attributes, and the accuracy and consistency of high-level attributes using DCMs have rarely been reported compared with classical test theory (CTT) and item response theory models. This paper compared the accuracy of high-level attribute mastery between deterministic inputs, noisy “and” gate (DINA) and Rasch models, along with sub-scores based on CTT. Methods: First, a simulation study explored the effects of attribute length (number of items per attribute) and the correlations among attributes with respect to the accuracy of mastery. Second, a real-data study examined model and item fit and investigated the consistency of mastery for each attribute among the 3 models using the 2017 Korean Medical Licensing Examination with 360 items. Results: Accuracy of mastery increased with a higher number of items measuring each attribute across all conditions. The DINA model was more accurate than the CTT and Rasch models for attributes with high correlations (>0.5) and few items. In the real-data analysis, the DINA and Rasch models generally showed better item fits and appropriate model fit. The consistency of mastery between the Rasch and DINA models ranged from 0.541 to 0.633 and the correlations of person attribute scores between the Rasch and DINA models ranged from 0.579 to 0.786. Conclusion: Although all 3 models provide a mastery decision for each examinee, the individual mastery profile using the DINA model provides more accurate decisions for attributes with high correlations than the CTT and Rasch models. The DINA model can also be directly applied to tests with complex structures, unlike the CTT and Rasch models, and it provides different diagnostic information from the CTT and Rasch models.
Characteristics of Item Parameter Estimation for the Multidimensional Item Response Theory (MIRT)
서동기,김좌근,김경태 한국심리학회 2015 한국심리학회지 일반 Vol.34 No.2
This study analyzes the three different estimation algorithms for recovering item parameters for the compensatory multidimensional IRT (MIRT) models. In particular, two- and four-dimensional models were investigated with different degrees of correlation between latent traits. The standards such as bias, standard error, and root mean square error were used to evaluate the recovery of item parameters for each program. The results indicated that in most conditions, Metropolis-Hasting Robbins-Monro (MH-RM) outperformed full information item factor analysis (FIIFA) and bivariate information item factor analysis (BIIFA) for a-parameters except for the independent and very low inter-trait correlation conditions where BIIFA outperformed the other algorithms. However, the MH-RM algorithm consistently produced the highest empirical standard errors compared to the other two methods for all conditions. FIIFA performed at a higher standard than BIIFA for a-parameters with moderately correlated latent traits. BIIFA is more suitable for a-parameters, especially when the levels of latent traits' independence or correlation are very low, and it is more suitable for d-parameters regardless of inter-trait correlations in the four-dimensional models. Overall, three estimation methods provided more accurate a- and d-parameter as the number of examinees increased, and less accurate a-parameter occurred as the inter-trait correlation increased. The inter-trait correlation condition did not have a dramatic impact on the recovery of d-parameter across all three algorithms.
미국의 부동산시장 - 펜더믹 이후의 미국과 우리나라 부동산시장 -
서동기 국제부동산정책학회 2020 토지와건물 Vol.35 No.-
The real estate market is so chaotic that it is hard to predict due to corona. The rise in housing prices is partly due to lack of supply and price formation due to location, but the downturn in commercial real estate seems impossible to predict a recovery time. Consumer prices rose only 0.6 percent in January, following 0.4 percent in 2019 and 0.5 percent last year. However, housing, stocks, cryptocurrency, oil prices, daily necessities, etc. are almost on the rise. Inflation is like an “invisible tax” since only salaries remain the same. It is time to make every effort to manage risks in preparation for the return of inflation in 30 years. The nature of the problem is that we will not artificially raise interest rates, which could drive the economy into a catastrophe, but we cannot just monitor the state of inflation.
모형명세화 오류와 소표본에서 구조방정식모형 모수추정 방법들 비교: 모수추정 정확도와 이론모형 검정력을 중심으로
서동기,정선호,Seo, Dong Gi,Jung, Sunho 한국데이터정보과학회 2017 한국데이터정보과학회지 Vol.28 No.5
구조방정식모형은 사회과학 및 행동과학 연구 분야에서 이론검정을 위해 주로 사용되는 통계방법이다. 최근 이 통계기법에 대한 방법론적 이슈로서 모형명세화 오류와 소표본 문제가 부각되고 있다. 그런데 이 문제들이 구조방정식모형의 대표 추정 방법인 최대우도법에 위한 이론검정에 어떤 영향을 주는지에 대해 여전히 명확하지 않다. 따라서 본 연구에서 최대우도법 그러고 이에 대한 대안으로 개발된 2단계최소자승법과 2단계능형최소자승법을 정확도와 검정력 관점에서 시뮬레이션을 통해 체계적으로 비교해 본다. 이 실험 결과에 따르면, 모형이 정확하게 설정된 경우, 정확도 기준에서 추정방법들 간의 차이는 미미했다. 하지만 모형오류가 발생한 경우, 2단계능형최소자승법은 다른 방법들보다 표본 크기가 작을 때 훨씬 더 정확한 모수추정치를 산출해 내었다. 그러고 이 방법은 명세화 오류에 관계없이 표본 크기가 작을 때에도 제 2종 오류 (Type II error) 수준이 상대적으로 작거나 만족할만한 수준의 검정력을 보여주었다. 이에 반해 다른 두 방법들은 표본이 작은 경우 또는 명세화 오류가 있는 경우 상당히 높은 수준의 제 2종 오류를 나타내었다. Structural equation modeling (SEM) is a basic tool for testing theories in a variety of disciplines. A maximum likelihood (ML) method for parameter estimation is by far the most widely used in SEM. Alternatively, two-stage least squares (2SLS) estimator has been proposed as a more robust procedure to address model misspecification. A regularized extension of 2SLS, two-stage ridge least squares (2SRLS) has recently been introduced as an alternative to ML to effectively handle the small-sample-size issue. However, it is unclear whether and when misspecification and small sample sizes may pose problems in theory testing with 2SLS, 2SRLS, and ML. The purpose of this article is to evaluate the three estimation methods in terms of inferences errors as well as parameter recovery under two experimental conditions. We find that: 1) when the model is misspecified, 2SRLS tends to recover parameters better than the other two estimation methods; 2) Regardless of specification errors, 2SRLS produces small or relatively acceptable Type II error rates for the small sample sizes.
단축형 심리검사 개발의 측정학적 방법과 타당화: 한국형 역기능우울척도를 중심으로
서동기,이순묵,김종남,최승원,채정민,정선호,조성경,김명기,Chad Ebesutani 한국심리학회 2019 한국심리학회지 일반 Vol.38 No.1
본 연구에서는 집단 연구용으로 개발된 한국형 역기능우울(dysfunctional depression)척도 20문항을 이용하여 단축형 검사 5문항을 선별하고 타당화하였다. 한국형 역기능우울척도의 단축형은 연구용 20문항 중 일부 문항(5문항)을 선별하여 제작된 것으로, 연구용 척도와 단축형 척도 모두 자기보고식 척도이다. 역기능우울척도는 역기능우울 상태를 측정하기 위한 것으로 일상우울의 상태가 심화되었거나 경우에 따라 일상우울에서는 없는 증상들을 측정하는 도구이다. 연구용 척도는 역기능우울에 대한 학술연구에서 집단수준에서의 추론을 위해 개발되었고, 단축형 척도는 주로 임상 및 상담 현장에서 개인의 역기능 우울 수준에 대한 신속한 진단을 위해 개발되었다. 즉 연구용 척도는 역기능우울에 대한 이론구축 및 관계구조 검증이 목적이 되고, 단축형 척도는 현장에서 개인의 역기능우울 상태를 신속하게 점검하는 것이 목적이다. 본 연구에서 개발한 단축형 척도의 신뢰도와 타당도는 척도의 내적 구조 분석과 기준점을 이용한 분류정확성과 분류일관성 지수에 의해 검정이 되었다. The purpose of this study was to develop a short form of the Korean dysfunctional depression scale (KDDS) and introduce methods for validation of the short form. The short form was developed by selecting five items from the original 20 items in the Korean dysfunctional depression scale. While the KDDS was developed to measure dysfunctional depression for the purpose of group research, the short form was developed for the practical use at individual levels. The scale for academic study was developed to infer the dysfunctional depression at the group level. So the short form can be widely used by practioners in industries, organizations, clinical/counselling settings for quick screening of potential patients suffering from dysfunctional depression. We demonstrated the reliability and validity of the short form by analyzing internal structure of the scale and providing classification accuracy and consistency.
대구경 볼트와 와이어텐션을 이용한 슬래브 보강공법에 관한 실험적 연구
서동기,노현섭,김상모,이창남,김상섭 대한건축학회 2003 대한건축학회 학술발표대회 논문집 - 계획계/구조계 Vol.23 No.1(구조계)
As time goes up, structural capacity of members decreases in reinforced concrete structures. The deterioration of the capacity is mainly due to the poor construction, corrosion of steel bars and the neutralization of concrete due to environmental pollution, etc. This deterioration is unavoidable, thus it is necessary to develop the method for repair and retrofit of structures in order to ensure structural safety of members. The purpose of this paper is to investigate the load-deflection, ultimate strength, failure mode, and energy absorption capacity of reinforced concrete slabs subject to load experimentally for several variables and grasp the problem of concrete slabs strengthened with large diameter bolts and wire tensioning. And propose foundational research data for4 strengthening reinforced concrete slabs. The test result shows that the capacity of anchorage is controlled by bearing capacity of concrete under the anchorage or punching shear around the anchorage. And the gap between bolts and core hole must be filled to prevent slip and rotation of anchorage.
한국판 PCI 검사의 타당화 연구 : 국내 직무장면을 중심으로
서동기,김예실,유대열,임다현 한국경영컨설팅학회 2018 경영컨설팅연구 Vol.18 No.2
본 연구의 목적은 성격의 5요인을 기반으로 한 Personality Characteristics Inventory(PCI) 검사를 한국판으로 타당화 하는 것이다. PCI는 서구 직무 장면에서 성격이 업무성과를 예측하는 것을 입증한 검사로서, 기업조직과 학계에서 활발하게 사용되고 있다. 본 연구에서는 국내 컨설팅 업체 및 바이오벤처회사 직원 145명을 대상으로 PCI 검사를 타당화 하였다. 번역 및 역번역 과정을 통하여 번안타당도를 확보하였고, 중간 수준 이상의 내적일치 신뢰도 및 5요인 12하위요인 구조를 확인하였다. 상사가 부하직원의 과업수행 및 조직시민행동을 평정하게 하여 이를 준거로 회귀분석을 실시함으로써 준거타당도를 확보하였다. 성격의 5요인에서는 원만성이 과업수행 및 제안활동을 정적으로 예측하고, 성실성이 대인지향 및 조직지향 조직시민행동을 정적으로 예측하는 것으로 나타났다. 반면 외향성은 조직지향 조직시민행동을 부적으로 예측하고, 안정성은 과업수행, 대인지향 및 조직지향 조직시민행동을 부적으로 예측하는 것으로 나타났다. 하위 12요인에서 보면, 자신감이 과업수행, 제안활동, 대인지향 및 조직지향 조직시민행동을 부적으로 예측하는 것으로 나타났다. 마지막으로 본 연구의 제한점을 논의하고, 추후 보다 완성도 있는 연구를 위한 제안을 하였다. The purpose of this study is to validate the Personality Characteristics Inventory (PCI) in Korean version. PCI which is proved to predict performances in Western work settings has been actively used in business organizations and academia. In this study, the PCI was validated for the subject of 145 consultants and bio-venture employees in Korea. The translation validity was obtained through translation and back-translation processes. The internal consistency reliability at the intermediate level and the construct of the 5 factors and 12 sub factors were confirmed. The criterion validity of the in-role behavior and OCB with supervisor ratings was obtained by regression analysis. In the 5 factor of personality, Agreeableness predicted in-role behavior and voice in the positive way, and Conscientiousness predicted toward individuals OCB and toward the organization OCB in positive way. On the other hand, Extroversion predicted toward the organization OCB in negative way, and Emotional Stability predicted in-role behavior, toward individuals OCB and toward the organization OCB in negative way. In the sub 12 factors, Self Confidence predicted in-role behavior, voice, toward individuals OCB, and toward the organization OCB as negatively. Finally, the limitations of this study were discussed and proposals for more complete research were made.