감점제와 단답형을 도입한 컴퓨터화 변형 선다형 시험 방식이 역행풀이 감소에 미치는 영향

송민해(Min Hae Song),박주용(Jooyong Park) 한국교육평가학회 2020 교육평가연구 Vol.33 No.3

본 연구는 우리나라 교육 현장에서 광범위하게 사용되는 선다형의 단점인 역행풀이를 완화시키기 위한 노력의 일환으로 수행되었다. 역행풀이란 답지를 이용하여 답을 추측해내는 전략이다. 이를 막기 위해 문제만 먼저 제시하고, 학습자가 요청하는 경우에 선지를 짧게 제시하는 컴퓨터화 변형 선다형 시험 방식(CMMT)을 고려했다. CMMT는 선지가 짧은 시간만 제시되어 역행풀이를 일부 줄일 수 있지만 학습자가 선지가 보이는 짧은 시간을 이용하여 답을 추론할 가능성이 여전히 남아있다. 본 연구에서는 이 가능성을 방지하기 위해 CMMT 방식에 감점제와 선다형으로 풀기 전에 단답형으로 답을 쓰는 절차를 도입하였다. 대학생을 대상으로 두 실험을 진행한 결과 CMMT 방식에 감점 비율이 높을 때 역행풀이가 유의하게 줄어듦을 발견하였다. 전체 논의에서는 본 연구 결과를 토대로 실제 교육 현장에서의 적용 가능성과 함께 이와 관련된 후속 연구 방향을 다루었다. This study was conducted to find out a way of reducing the problem of multiple-choice test, which is widely used in current Korean educational setting. The problem of multiple-choice test is that students could find the correct answer from the options by guessing when they do not know the answer. In order to prevent the problem, we considered the modified multiple-choice testing system (CMMT). CMMT could lessen the possibility of guessing by showing options briefly upon their request. However, even with the CMMT system, some students tried to select options during short time when they do not know the correct answer. The aim of this study was finding out if penalizing wrong answers and combining short answer could lower the possibility of guessing. Two experiments were conducted by students who were tested with the directions that they will be penalized for wrong answers, but not for no response. In experiment 1, two factors were manipulated: levels of penalizing (high and low) and formats of testing (traditional multiple-choice format and CMMT forma)t. The number of no response and probability of guessing were compared. Results showed that the number of no response was significantly higher and probability of guessing were significantly lower when using the CMMT with high penalizing percentage. In Experiment 2, we replicated the results of Experiment 1 using higher penalizing percentages than those of Experiment 1. Implications of these results and directions for future research in educational psychology were speculated in General discussion.

국가수준 학업성취도 평가의 서답형 학생 답안 분석 및 발전 과제

장의선 한국사회교과교육학회 2018 사회과교육연구 Vol.25 No.4

The purpose of this study is to derive directions for improvement and development of descriptive evaluation items. To this end, the researchers analyzed the students' answer sheets to geography related items of the National Assessment of Educational Achievement. As a result of the analysis, the students' answers varied and different than expected, including the ways of combinations of content elements that constitute the correct answers and unexpected answers. The wrong answers are affected by the unclear understanding of concepts, inaccurate memory of concepts, and the complexity of evaluation elements. Based on these results, the implications for future improvements and developments in descriptive evaluation items are as follows: First, in terms of essay items, the elements of the assessment should be clarified. In addition, the structure of the question should be refined to measure the overall thinking power. Second, in terms of short-answer item, specific examples of real life should be presented as questionnaires, not as a prior definition of concepts. Third, the information for analyzing student answers can be useful for teaching learning strategies by achievement level. Therefore, the data should be processed and provided for easy use by teachers. 본 연구는 국가수준 학업성취도 평가의 사회과 지리 영역 서답형 문항에 대한 학생 답안을 분석하고 향후 서답형 문항의 개선 및 발전을 위한 시사점과 과제를 도출한 것이다. 연구 결과, 학생들의 답안 진술 양상은 정답을 구성하는 내용 요소 조합에 대한 경우의 수와 예상 밖의 답안 등으로 인해 문항별로 차이는 있지만 출제자의 예상보다 다양하게 나타났다. 오답군의 경우에는 개념의 명확한 이해와 개념어 또는 용어의 정확한 인출, 그리고 평가 요소의 복합성 등이 영향을 미친 것으로 나타났다. 이러한 분석 결과를 토대로 한 향후 서답형 문항의 개선 및 발전에 대한 시사점은 다음과 같다. 첫째, 서술형 문항 출제의 측면에서는 평가 요소의 명료화와 함께 종합적인 사고력 측정에 적합하도록 문항 구조의 정교화 작업이 출제 과정에서 지속적으로 견지되어야 한다. 둘째, 단답형 문항 출제의 측면에서는 개념이나 용어의 사전적 정의보다는 실생활의 구체적 사례를 문항 자료로 제시해야 함에 유의해야 한다. 셋째, 향후 문항별 통계 정보와 분석 자료는 성취수준별 교수학습 전략에 유용하게 활용될 수 있는 정보이므로 수업 활용에 용이하도록 다양한 형태의 가공 작업과 적극적인 홍보를 통하여 학교 현장에 제공될 필요가 있다.

삶과 물음에 비춰본 시험 : 물음, 시험, 그리고 삶

이경숙 한국교육개발원 2008 한국교육 Vol.35 No.1

The purpose of this article is to investigate a meaning of the question and the examination in life. The questions in life are the exploration toward the world. The questions begin with the known world, and their growth is accomplished by attention, posing problems, and immersion in it. Questions and answers are structured in form of ascent. So the questions lead to the answers, the answers lead another questions. This is the process of a dialectic in a sense. In this courses, the quality of questions and answers is deepened and widened. The examination is the institutionalization of the activity of questions and answers in human life. It in life may be either completion of the activity or restriction of the body and thinking. In view of distance of examiners and examinees, it may be either the short-distance examination or the great-distance examination. The results of examinations may be either data of reflection for teaching activities or an instrument for controlling examiners and examinees. The activity of questions and answers is the condition of human life. As the institutionalization of the activity of questions and answers, the examination must not restrain thinking of human, but stimulate the activity in human life. 인간에게 물음은 세계를 향한 탐험이다. 기지의 세계에서 미지의 세계로 확장되는 것이다. 물음의 출발은 마음의 기울어짐이며, 문제의 명료화와 몰입을 통해 물음은 진화된다. 물음에 대한 답은 정답으로서 먼저 존재하는 것이 아니라, 물음과 답이 상승하는 구조를 갖고 있다. 시험은 물음과 답의 제도화로서, 인간 삶에서 물음과 답을 완성하는 경험을 줄 수도, 인간의 정신과 신체를 구속하는 경험을 줄 수도 있다. 그리고 시험결과는 반성자료로서 존재할 수도, 권력의 도구가 될 수도 있다. 삶 속에서 시험은 그 경계지점에 있지만, 물음과 답의 완성경험으로, 반성자료로 사용되는 것을 지향해야 한다.

Ways of Replying to Negative Questions in English and Korean

박명관 한국영어학회 2015 영어학 Vol.15 No.3

Park, Myung-Kwan 2015. Ways of replying to negative questions in English and Korean. Korean Journal of English Language and Linguistics 15-3, 397-421. This paper investigates the syntax of negative question sentences and answers to them. By definition, yes/no or polarity questions are ones whose expected answer is either yes or no. We suggest that the functional category, termed ∑ or Pol(ity), which determines the polarity of a clause is generated with an unvalued feature, whose value is determined either structurally by the polarity particles yes and no or by default. We argue that the inverted n't before a subject is a realization of the Pol valued as minus by default, and the negation not inside TP functions as either a lexicalized item of the minus valued Pol or a VP-modifying adverb. We show that the inverted n't, the TP-internal negation not, and genuine adverbs like never or hardly behave differently but systematically, accounting for short answer replies as well as full sentence answers. We then move on to examine the patterns of replies to negative sentences in Korean, and investigate why Korean learners of English make an error in replying to negative questions in English. It will be revealed that despite the fact that the inverted n't before a subject is a default realization of the minus valued Pol, Korean learners of English mistakenly identify it with the negation not, resulting in making such an error.

구성적 선다형 검사 방식에서 배점 조작의 효과

박주용 한국인지및생물심리학회 2010 한국심리학회지 인지 및 생물 Vol.22 No.4

The Constructive Multiple-choice Testing (CMT) system is a new computerized testing system developed to supplement the weaknesses of the multiple-choice (MC) format. The CMT system involves having the examinee respond to the stem first in the short answer format and then in the MC format. Therefore, one can see whether or not the examinee chose the correct option in the MC format because he or she actually knew the answer by checking the short answer portion of the examinee's response. The current study was carried out to examine whether there is any difference in the scores obtained from the CMT test as opposed to the short answer or the MC tests. Two hundred and twenty seven 6th graders in elementary school were randomly assigned to 2 groups. In Experiments 1 and 2, comparison was made between the performance of the group who took the test in the CMT format and that of the group who took the test in the short answer format. In Experiment 1, where the instruction for the CMT test was that the two portions would be graded separately, the mean of the short answer portion of the CMT group was lower than that of the short answer format group. However, in Experiment 2, the examinees were told that the short answer portion would be weighed 9 times as heavily as the multiple-choice portion (90% vs. 10%). There was no difference in the means of the short answer responses between the two groups. In Experiment 3, the means of the multiple-choice responses were compared between the CMT group and the multiple-choice group. The CMT group was given the same instruction as in Experiment 2. The result revealed that the mean of the MC portion of the CMT group was higher than that of the MC group. These results suggest that the performance is not affected by having the examinees respond twice if more points are allotted to the short answer portion of the CMT format. 구성적 선다형 검사 방식은 선다형 방식의 문제점을 보완하기 위해 개발된 새로운 컴퓨터화 검사 방식이다. 이 방식에서는, 하나의 발문에 대해 먼저 단답식으로 반응하도록 한 다음, 다시 동일한 발문에 대해 선다형으로 반응을 하도록 한다. 따라서 선다형에서 정답을 선택했더라도 단답식 반응을 통해 정답을 알고 선택했는지 모르고 선택했는지를 확인할 수 있다. 이 방식을 사용한 선행 연구에서, 구성적 선다형으로 본 집단과 단답식으로만 본 집단의 단답식 수행을 비교하였을 때, 단답식으로 본 집단이 더 높은 점수를 얻는다는 것이 반복적으로 관찰되었다. 본 연구는 구성적 선다형의 단답식 수행과 단답식으로만 보았을 때의 수행간 동등성을 확보하기 위해, 구성적 선다형의 단답식 점수 배점을 높이는 조작이 실제 수행에 어떻게 영향을 주는지를 알아보기 위해 수행되었다. 초등학교 6학년생 227명을 대상으로 3개의 실험이 수행되었다. 실험 1과2에서는, 무선적으로 나뉘어진 두 집단을 대상으로, 한 집단은 구성적 선다형 방식으로, 다른 집단은 단답식으로 보게 한 다음 수행을 비교하였다. 각각 따로 채점된다고 지시를 준 실험 1에서는 구성적 선다형에서의 단답식 반응이 단답식으로만 반응하게 한 경우보다 낮았다. 그렇지만 단답식과 선다형의 배점을 90% 대 10%로 했을 때에는 두 집단간에 단답식 점수의 차이가 사라졌다. 실험 3에서는, 점수 배점이 90% 대 10%일 때, 구성적 선다형 집단과 선다형 집단의 수행을 비교하였다. 그 결과 오히려 구성적 선다형 집단의 선다형 점수가 높아짐을 발견하였다. 이 결과는 구성적 선다형에서 단답식의 비중을 크게 하면, 두 가지 반응을 하게 하더라도 그로 인해 따로 따로 반응하게 했을 때에 비해, 수행이 저하되지는 않음을 시사한다.

On the Locus of the Interpretable Negative Feature in a Negative Sentence

Daeho Chung 서울대학교 언어교육원 2014 語學硏究 Vol.50 No.2

Kramer and Rawlins (K & R 2009) analyze short answers yes and no to a polar question as sentence fragments involving TP ellipsis along the similar lines of Merchant’s (2001, 2004) treatment of short answers to a content question. Their TP ellipsis analysis was mainly grounded on the following two factual claims in (i) and (ii) and the theoretical assumption in (iii): (i) Short answers to an outer negative question carry the same meaning as short answers to an affirmative question; (ii) Short answers yes and no to an inner negative question are neutralized as a negative meaning, which K & R (2009) dub as negative neutralization; and (iii) the interpretable negative feature of a negative sentence may reside in the higher or lower Σ. This paper, however, refutes the factual claim in (ii) (negative neutralization) and the theoretical assumption in (iii). It is shown that (ii) is not (at least not always) empirically supported, when corpus data is considered, and that (iii) faces non-trivial theoretical problems. This paper instead proposes that the interpretable negative feature resides in the higher Σ, treating an overt or covert negation morpheme in the lower Σ as pleonastic. The proposed higher negative theory draws further support from the scope widening effects of negation and polarity reversals displayed in NPI/N-word fragmental answers to a content question.

On the Locus of the Interpretable Negative Feature in a Negative Sentence

정대호 서울대학교 언어교육원 2014 語學硏究 Vol.50 No.2

Kramer and Rawlins (K & R 2009) analyze short answers yes and noto a polar question as sentence fragments involving TP ellipsis alongthe similar lines of Merchant’s (2001, 2004) treatment of short answersto a content question. Their TP ellipsis analysis was mainly groundedon the following two factual claims in (i) and (ii) and the theoreticalassumption in (iii): (i) Short answers to an outer negative question carrythe same meaning as short answers to an affirmative question; (ii)Short answers yes and no to an inner negative question are neutralizedas a negative meaning, which K & R (2009) dub as negative neutralization;and (iii) the interpretable negative feature of a negative sentencemay reside in the higher or lower Σ. This paper, however, refutes thefactual claim in (ii) (negative neutralization) and the theoretical assumptionin (iii). It is shown that (ii) is not (at least not always) empiricallysupported, when corpus data is considered, and that (iii) facesnon-trivial theoretical problems. This paper instead proposes that theinterpretable negative feature resides in the higher Σ, treating an overt orcovert negation morpheme in the lower Σ as pleonastic. The proposedhigher negative theory draws further support from the scope wideningeffects of negation and polarity reversals displayed in NPI/N-wordfragmental answers to a content question.

정답 템플릿 작성 방식에 의한 한국어 서답형 문항 자동채점 시스템

박일남(Il-Nam Park),강승식(Seung-Shik Kang),노은희(Eun-Hee Noh),김명화(Myung-Hwa Kim),성태제(Tae-Je Seong) 한국정보과학회 2013 정보과학회 컴퓨팅의 실제 논문지 Vol.19 No.12

본 논문에서는 단어-구 수준의 한국어 서답형 문항을 컴퓨터 기반 채점 방식으로 처리하는 방법을 제안한다. 수천 명의 채점자가 채점 기준표를 보면서 수작업으로 채점하는 방식을 컴퓨터를 이용한 자동채점 소프트웨어로 대치하기 위하여 수동 채점을 목적으로 작성된 채점 기준표를 자동채점을 위한 XML 기반의 정답 템플릿으로 기술한다. 정답 템플릿을 기준으로 정답 일치 채점, 개념 기반 채점, 단서어 기반 오류 처리의 과정을 통해 학생 답안에 점수를 부여하는 자동채점 시스템을 설계하고 개발하였다. 자동채점 시스템의 성능을 평가하기 위해 자동 채점 결과와 실제 채점자가 부여한 점수를 비교, 분석하였다. 그 결과에 의하면 실험에 사용한 11개 문항의 평균 카파 계수는 0.83으로 실제 채점자와 비교했을 때 자동채점 시스템은 수작업 채점과 일치도가 매우 높은 성능을 나타내었다. This paper proposes a computer-based automatic scoring method for Korean short answer questions. The purpose of this research is to replace a traditional method of human raters by scoring guide-line with an automatic scoring by computer software. We transformed a hand-written scoring guide-line to a machine-readable answer template that is an XML format. Automatic scoring system consists of four steps: model answer scoring, concept-based scoring, cueword-based scoring, and manual scoring. Kappa coefficient showed an average 0.83 between automatic and manual scoring method.

함수 영역 문제해결 협력학습 과정에서 문제 유형에 따른 중학생의 인지부하 분석

김성경 ( Kim Seong-kyeong ),김지연 ( Kim Ji Youn ),이선지 ( Lee Sun Ji ),이봉주 ( Lee Bongju ) 한국수학교육학회 2018 수학교육 Vol.57 No.2

From the assumption that an individual’s working memory capacity is limited, the cognitive load theory is concerned with providing adequate instructional design so as to avoid overloading the learner’s working memory. Based on the cognitive load theory, this study aimed to provide implications for effective problem-based collaborative teaching and learning design by analyzing the level of middle school students’ cognitive load which is perceived according to the problem types(short answer type, narrative type, project) in the process of collaborative problem solving in middle school function part. To do this, this study analyzed whether there is a relevant difference in the level of cognitive load for the problem type according to the math achievement level and gender in the process of cooperative problem solving. As a result, there was a relevant difference in the task burden and task difficulty perceived according to the types of problems in both first and second graders in middle schools students. and there was no significant difference in the cognitive effort. In addition, the efficacy of task performance differed between first and second graders. The significance of this study is as follows: in the process of collaborative problem solving learning, which is most frequently used in school classrooms, it examined students' cognitive load according to problem types in various aspects of grade, achievement level, and gender.

Ways of Replying to Negative Questions in English and Korean

Park, Myung-Kwan(박명관) 한국영어학회 2015 영어학 Vol.15 No.3