컴퓨터 모의구술능력 평가의 음성인식기술 활용 가능도

최인철 한국멀티미디어언어교육학회 2004 멀티미디어 언어교육 Vol.7 No.2

With the advent of ubiquitous era due to proliferation of network-based teaching and testing, the computerized adaptive language test (CALT) appears to be an inevitable method of language testing in the 21st century. The validation research of CALT is mostly conducted on the basis of the oral proficiency interview (OPI) done by human raters (최인철, 2004). This validation practice utilizing human raters, however, has the inherent problem with the internal validity of research, in that OPI ratings done by human raters tend to have relatively low inter-rater and intra-rater reliability. Thus, human ratings in many cases fail to meet the rigorous criteria required by high-stakes language test. Therefore, the automatic speech recognition (ASR) technology-based simulated oral proficiency interview (SOPI) merits our attention as a viable alternative approach to human rater-based OPI. In this respect, this paper attempts to investigate the plausibility and feasibility of implementing ASR-based Oral Test by exploring the relationship between ASR-based Oral Test and native speaker raters, and finally CALT that has been validated by human rater-based OPI. The findings are so revealing that it warrants more large-scale experiment in which more sophisticated ASR-based test employs more valid tasks.

제도, 제도개혁과 경제성장

최인철 한국경제연구원 2000 제도연구시리즈 Vol.- No.1

Computerized Adaptive Language Testing 알고리즘 개발 및 타당도 검증

최인철 서울대학교 언어교육원 2004 語學硏究 Vol.40 No.1

In the era of information and communication technology, where computerized adaptive language testing continues to draw a keen attention from language testers, the present research attempts to develop a reliable and efficient algorithm of computerized adaptive language testing (CALT) and to investigate the extent to which the algorithm results in reliable and valid outcome. Employing an eclectic approach to analyzing real data as well as simulation data, the research has developed a highly reliable and valid algorithm on the basis of fundamental considerations including the difficulty level of initial items, the estimation method of the ability parameter, the item selection, the test length, the measurement error, the context effect, the item-based response time, etc. Focusing on the concurrent validity, the overall research findings suggest that the CALT version of the TEPS (Test of English Proficiency developed by SNU) accommodating the item- based response time factor is superior to the paper-based language test (PBLT) version of the TEPS in terms of measurement accuracy of overall English proficiency. Furthermore, the correlational analyses based on test-takers' performance of CALT, PBLT, and TOP (the Test of Oral Proficiency) strongly support the plausibility of the test methods of the TEPS in which the grammar test is administered in a speeded manner and the listening comprehension test input is presented in a purely oral mode.

신제도주의 이론의 비판적 검토

최인철 한국사회경제학회 1996 사회경제평론 Vol.- No.9

Dexmedetomidine for sedation in pediatric patients who received more than 20 sessions of radiation therapy -two cases report-

최인철,김선기,송명희,이일정,이재혁 대한마취통증의학회 2016 Korean Journal of Anesthesiology Vol.69 No.6

Dexmedetomidine is a highly selective α2-adrenoceptor agonist that demonstrates anxiolytic and analgesic properties without inducing respiratory compromise, which makes it a suitable agent for procedural sedation and imaging studies. In our current case reports, intravenous dexmedetomidine infusion was used to provide sedation to 2 pediatric patients over more than 20 sessions of radiation therapy. On both occasions, dexmedetomidine provided adequate sedation without respiratory depression. However, the required dosage increased with repeated radiation therapy sessions.

TEPS 개인별 수험 결과 추이 분석 : Trend Analysis

최인철,이혜옥 서울대학교 어학연구소 2001 語學硏究 Vol.37 No.3

1999년부터 2년 이상 시행된 TEPS의 시험 방식의 타당도에 대한 수험자의 의견은 전문 설문조사 기관에 의한 설문조사 결과(1999)에 의하면, 매우 고무적이라고 할 수 있다. 그러나, "TEPS 점수는 들쑥날쑥"이라는 무책임한 일부 사람들이 퍼뜨리는 소문이 과연 사실에 근거한 것인지를 알아보기 위해서 본 연구에서는 2회 이상 중복 응시한 수험자의 수험결과의 개인별 변화 추이를 다음과 같은 점을 고려하여 체계적으로 분석하였다. 우선, 성별, 연령(학력), 점수 등급별로 수험자 집단 성격을 파악하였다. 또한, 응시 회수와 간격에 따라 각 변인별로 어떤 형태의 변화추이가 나타나며, 등급이 변화 폭이 얼마나 되는지 분석하고, 점수 변화 정도는 통계적으로 유의한 정도인지를 알아보았다. 연구 결과, TEPS는 3급이하 집단에서는 6회 이상(1년기간) 시험 치뤘을 때에 100점 정도의 상승 효과가 있었지만, 2급 이상의 상위권 집단에서는 6회 이상 시험을 치루었어도 상승 효과가 50점 미만(한등급의 반수준내)에 그쳤다. 모든 등급 수준의 응시자들에게 응시회수에 따른 점수 상승의 패턴이 있으나, 이처럼 등급수준이 높아지면서 그 상승 폭이 줄어든다는 것은 응시회수나 응시간격에 따라 변화되지 않는 안정적인 능력 수준으로 해석될 수 있다. 이로써, 특히 상위권 수험자들은 잘 변별하고, 요령이 통하지 않는 시험임이 입증되었다. 결론적으로, 본 연구를 통해 TEPS 점수의 신뢰성이 밝혀졌으며, 공신력 있는 시험으로서의 타당성이 검증되었다고 볼 수 있다. A survey was conducted by Korean Gallup and Marketing Research Korea on the test takers who took the first TEPS exam administered on January 31, 1999. The results indicated a strong consensus among English teachers and test-takers as to the validity of the test method facets of the TEPS exam. There are, however, some who feel that the TEPS exam does not yield reliable results. In order to investigate the extent to which such claims hold true, the present study attempts to conduct a systematic trend analysis based on the test-takers' performances. The study identifies those test-takers who took the TEPS more than twice within a two hear period (1999-2000), and explored the fluctuation pattern, if any, of the test-takers' performances while taking into account their ability level, gender, and age. The results show that those test-takers below TEPS Level 3 obtained about 100 points (which is equal to the difference between each level) taking the TEPS six times over a one year period, and that those above TEPS Level 2 obtained no more than 50 points (which is equal to half of the difference between each level) taking the TEPS six times over one year period. Given the fact that test scores among all test-takers of all levels showed a rising pattern, but that the trend of rising patterns diminished as the TEPS level increases, the belief that the TEPS exam is not reliable is unwarranted The overall findings of the present study reveal that the TEPS exam succeeds in producing reliable test results with high discriminability, especially among those test-takers of above average ability.

토끼에서 내독소로 인한 급성 폐손상에 미치는 Aprotinin 의 효과

최인철,고홍,조인해,심지연,최수경 대한마취과학회 2001 Korean Journal of Anesthesiology Vol.41 No.5

우유용종이팩 예치금 부과 비현실적

최인철 (사)한국포장협회 1993 包裝情報 Vol.2 No.-

한우 농가 수익성 향상을 위한 번식 및 출하 계획

최인철,조재성 한국수정란이식학회 2016 한국동물생명공학회지 Vol.31 No.3

Wholesale beef price is the critical factor for determining Korean native cattle, Hanwoo, farm’s income in short-term. Wholesale beef price has seasonality due to high demand in Korean traditional holidays such as Korean thanksgiving day and lunar new year’s day. Therefore, it is important to make reproduction and marketing plans for Korean Hanwoo farmers, in order to increase their farm income. However, there is no study available on changes in the expected farm income depending on reproduction and marketing schedules. This study analyzed the expected farm income per head depending on the monthly-based marketing schedules. The analysis was conducted based on the seasonality of wholesale beef price, reproduction efficiency, operating costs, relationship between carcass grade and slaughter age. The result shows that slaughter Hanwoo at the age of 29 months-old in August and January generating the highest expected farm income per head.

Test of Oral Proficiency (TOP)의 개발 연구

최인철 서울대학교 어학연구소 1998 語學硏究 Vol.34 No.1

Six important factors to be considered in exploring a test's usefulness include the content validity, reliability, authenticity, test--test taker interactiveness, practicality, and educational impact(or washback effect) (Bachman & Palmer 1996). It is virtually impossible to develop a perfect measurement tool which fulfills all of the six criteria, especially in measuring sophisticated communicative language ability such as oral proficiency. The needs analyses, however, strongly suggest that a performance test be developed to measure the general oral proficiency of Korean test-takers (Park et al. 1997). Thus, the present research is intended to develop a simulated oral proficiency interview test (tentatively entitled Test of Oral Proficiency, English [TOP, English]) designed to measure overall oral communicative skills on the basis of a sound theoretical framework. In order to investigate the extent to which the test method facets and the scoring schemes of the TOP are valid, the present study employs not only qualitative approaches but quantitative research methods including Multiple Regression, Factor Analysis, and GENOVA based on Generalizability Theory. The overall results reveal that the present test methods and the scoring schemes of TOP meet the requirements for valid measurement. Follow-up research is recommended to validate the slightly revised test methods of the TOP.