2002 Fiscal Year Final Research Report Summary
Development and Evaluation of Adaptive Computer-Based Test for Students with Disabilities in the National Center Test for University Admissions
| Project/Area Number |
12480047
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Educational technology
|
|---|
| Research Institution | The National Center for University Entrance Examinations (2001-2002)
National Center Test for University Admissions (2000) |
|---|
Principal Investigator |
FUJIYOSHI Mamoru Research Division, Professor, 研究開発部, 教授 (20190085)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
SUZUKI Masakazu Kyusyu Univ., Faculty of Mathematics, Professor, 大学院・数理学研究院, 教授 (20112302)
OHTAKE Nobuyuki Tsukuba College of Technology, Research Center on Educational Media, Associate Professor, 教育方法開発センター, 助教授 (10223851)
FUJIYOSHI Akio Ibaraki Univ., Faculty of Engineering Research Associate, 工学部, 助手
OKAMOTO Masayuki Shinshu Univ., Faculty of Engineering Professor, 工学部, 教授 (50109196)
KIMURA Shinji Kwansei Gakuin Univ., School of Law, Associate Professor, 法学部, 助教授 (40225040)
|
|---|
| Project Period (FY) |
2000 – 2002
|
| Keywords | students with disability / braille format test / large print formfat test / the Center Test for University Entrance Admission / computerized adaptive test / item response theory |
|---|
| Research Abstract |
We developed a prototype of an adaptive computer-based test (CBT), which can be measured more effectively and promptly than ordinal paper-and-pencil test (PPT), in order to make fundamental changes of the methods of exams for students with disabilities such as the National Center Test for University Admissions (NCT).
As a result of quantitative and tentative study to analyze the ratio of extension of time of the NCT for students with disabilities, we found the necessity of extension of time for the subjects that consist of a large amount of reading materials such as Japanese. It is, however, difficult to extend testing time of NCT since it has already spanned two days. We also found the fact that we could compare the result of linear CBT and ordinal PPT through our experiments. On the basis of this result we studied adaptive CBT so as to improve the methods of test fundamentally.
We advanced the development of the CBT system with the user interface to fit the types and extensions of disa
… More
bilities. For test-takers with severe visual disabilities, we automatically print and set a braille format test including decoding figures. We request test-takers to mark electrical answer sheets for computer scoring on a braille display. For the partially sighted, we studied the most suitable font type and size to display large print format tests in the use of pen computers. We revealed that digital audio format tests can work on the newly blind and students with learning disabilities who have difficulty with character reading.
We pursued the construction of an item pool of the CBT with the use of past tests of NCT. We investigated the way to convert the past tests of three subjects, Japanese, mathematics, and English, to XML format text files. We analyzed the past result files with item response theory (IRT) and estimated the differences of difficulty among questions in each year. We collected about 120 freshmen for equating the difference among the operating years of NCT. As each section includes several questions in NCT, we carried out study of IRT for system of testlet.
We will study the operation of the CBT in days ahead. The assessment of the availability of comparison between the result of adaptive CBT and PPT also remains to be experimented. Less
|
