• Communications of Mathematical Education
• /
• v.14
• /
• pp.151-162
• /
• 2001

교육평가는 선발의 기능을 할 뿐 아니라 학습자의 학습상태나 오류경향에 대한 진단을 통해 학습자의 학습에 도움을 주어야 한다. 이르한 측면에서 오류분석은 평가가 지녀야 하는 가장 중요한 기능 중 하나라고 할 수 있다. 인터넷의 대중적 보급과 시간적, 공간적, 기술적 효율성은 평가환경에 대한 새로운 가능성을 제시하고 있다. 인터넷 환경에서의 평가문항은 선다형을 기본으로 하게 되는데, 선다형 평가문항을 통해서는 학습자들의 오류를 직접적으로 관출할 수 없다는 한계가 있다. 그러나 선다형 평가문항을 통한 오류분석이 불가능한 것은 아니다. 본 연구의 목적은 선다형 평가문항에서의 오류분석을 위한 구체적인 방법과 절차를 모색하는 것이다.

• Journal of The Korean Association For Science Education
• /
• v.22 no.1
• /
• pp.122-131
• /
• 2002

Recently, performance assessment that makes growing use of free response items in a large scale assessment has been emphasized. This study is an empirical examination of the effectiveness of free response items in comparison with multiple choice items. Using the information function in Item Response Theory (IRT) framework, item information of free response items and multiple-choice items from the Third International Mathematics and Science Study-Repeat (TlMSS-R) were obtained. Test information of the whole science area as well as each area of science contents was computed. On average, free response items yielded more information than multiple choice items, especially in earth science, physics, chemistry, and life science. This study also showed that free response items were appropriate for students in high science ability. Also, free response items estimated students' science ability more accurately than multiple choice items with smaller number of free response items.

• The Journal of Korean Association of Computer Education
• /
• v.23 no.1
• /
• pp.29-36
• /
• 2020

Various items were used as evaluation tools that identify the student's abilities accurately to confirm the completion of learning. Among them, the multiple choice item has the advantages of high objectivity and reliability in scoring, but it cannot remove the factor of guessing. In this paper, the multiple choice items are classified into two types according to the relationship between the questionnaire and the choices. One is the type used in the classical test theory with the probability of guessing 1/k for k choices and the other is the novel proposed type which introduces the concept of partial knowledge. In the proposed type, the probability of guessing when the number of knowledge i is (i+1)/k for k choices. Based on the assumptions of the previous theories about multiple choice items, we derive the guessing parameter about the proposed type. And we analyzed the effect of the guess on the correct answer in the existing type and the proposed type. This shows that the proposed type has more question guessing than the existing type.

• Journal of Science Education
• /
• v.37 no.2
• /
• pp.278-293
• /
• 2013

The purpose of this study was to analyze the characteristics of multiple-choice test items used in assessment of high school integrated science according to 2009 revised curriculum. For the analysis of the tendency of item setting, we devised an analytic framework specific to integrated science, and analyzed the characteristics of items by applying the devised framework and item response theory. The results of the tendency of item setting revealed that most of items run counter to the intent of integrated science in terms of item resource, integration extent, and cognitive level, which means teachers are stick to separative method in teaching-learning and assessment of integrated science. The results of the analysis applying item response theory showed that item difficulty was appropriate and item discrimination was considerably high. However, there was no relevance between the tendency of item setting and qualitative characteristics of the items. We also discussed some agendas to improve the teaching-learning and assessment of integrated science based on the results of this study.

• Communications of Mathematical Education
• /
• v.2
• /
• pp.171-179
• /
• 1997

• Korean Journal of Cognitive Science
• /
• v.20 no.3
• /
• pp.275-289
• /
• 2009

A new computerized testing system, called the Computerized Multiple-choice Testing (CMMT) system, was introduced. In this system, questions of multiple choice (MC) items are presented first without options, so that students must generate answers for themselves. They can click for the options when they are ready, and can respond within a brief, specified time period. The present study was performed to examine whether this system is effective in reducing backward reasoning, I. e., using the options of MC items as cues to find the correct answer. One hundred and seventy-seven 6th grade students (12 year olds) were divided into two groups so that mean scores from a prior test were equal: The experimental group took an intervening computerized test in the new format, and the control group in the MC format. Five days after the computerized intervening test, a short answer paper-and-pencil final test was given. Testing effect was greater in the new system than in the MC system. Analysis of the final test response in relation to the intervening test response showed that i) the students retained the correct answer in the new system more than in the MC testing system, and that ii) students corrected their previous failures in the intervening CMMT format more than those in the MC format. These results suggest that the new system is effective in reducing backward reasoning.

• The Mathematical Education
• /
• v.43 no.2
• /
• pp.151-162
• /
• 2004

For the past several years, performance assessment has been widely used by mathematics teachers. The superiority of performance assessment items compare to multiple choice items has been discussed by many researchers, however these discussions tend to be lack of empirical data. Thus, this study aims to examine the effectiveness of tree response items in comparison with multiple choice items. Using the information function in Item Response Theory(IRT), item information of free response items and multiple choice items from the Third International Mathematics and Science Study-Repeat(TIMSS-R) were obtained and compared. Test informations of the whole mathematics area as well as each content area of mathematics were computed. On average, tree response items yielded more information than multiple choice items, especially in measurement and data interpretation. This study also revealed that free response items estimated students' mathematics ability more accurately than multiple choice items with smaller number of items.

• Journal of The Korean Association For Science Education
• /
• v.18 no.4
• /
• pp.643-650
• /
• 1998

In order to assess students' higher mental abilities, such as scientific inquiry thinking skills, the essay type items would be more adequate than the multiple choice itmes. However, due to the present condition in which a huge number of students take the examination at the same time, it is inevitable to use the multiple choice type. For this reason, it is necessary to develop a new type of multiple choice items which can reduce the disadvantages of the traditional multiple choice type and can achieve a similar level of validity as subjective type assessment. The three-tier multiple choice test items which can be used for a large sample of students and especially for scientific inquiry thinking abilities, are proposed and examined. The three-tier multiple choice test items asked firstly conclusion or the results of calculation or experimental apparatus, secondly the processes of calculation or of developing conclusion, thirdly asking relevant scientific concepts. For the item analysis, 1 point was given to the correct answer, while 0 point was given to the wrong one. The data were processed through the computer program developed in Turbo C 2.0 language with an IBM compatable personal computer. The average score in the sub-items asking for scientific concepts was lower than that in the sub-items asking for results or processes. The score of guessing by chance in the three-tier multiple choice items was only 0.13%, so that the probability of making correct answers by just guessing would be extremely low. The three-tier multiple choice items, even if they are objective items, are thought to assess thinking skills of the scientific inquiry meaningfully excluding the possibility of guessing by chance.

• Journal of Science Education
• /
• v.37 no.3
• /
• pp.458-475
• /
• 2013

According to instruction of standard-based assessment system, teachers with higher level of competence in student assessment are required. The purpose of this study is to offer the basic data to enhance pre-service physics teachers' competency in student assessment. 17 pre-service physics teachers, who are sophomore in department of physics education, participated in the study for a whole semester. To do this, we gave the assignment of multiple choice question construction in physics conceptest for formative assessment to pre-service teachers and observed review questions constructed by them. We also analyzed multiple choice questions of 17 pre-service teachers. They had 5 classes for the review process of the question and improved their questions through it. 16 of 17 pre-service teachers completed test design. Our results show that pre-service teachers had a little low competence in test design for classroom assessment. In addition to, constructing and revising questions, almost of them didn't have various perspectives and didn't consider to get information about the pre-conceptions of students. These findings indicate that they have difficulties with constructing multiple choice question and more specific plan needs to enhance the overall student-assessment competency levels of pre-service teachers for successful settlement of the standard-based assessment system.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.6
• /
• pp.3609-3615
• /
• 2014

This study examined the difference in the item characteristics in multiple-choice aptitude tests focusing on the response position, response length and question types. A university aptitude test consisting of 80 questions was used for this study. The subjects were 3120 senior high school students from 80 schools nation-wide (liberal arts-1650, natural sciences-1467 patients). The results suggest that item prediction is higher for numbers 2 and 3 (located in the middle) than numbers 1 and 4. The item discrimination was higher for pick-the-'wrong'-items than pick-the-'right'-items. In addition, longer choices are preferred. The suggestions for future research are provided based on these findings.