Standardizing National Tests Using Testlet Model
| Project/Area Number |
15330145
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Educational psychology
|
|---|
| Research Institution | The National Center for University Entrance Examinations (NCUEE) |
|---|
Principal Investigator |
ISHIZUKA Tomoichi NCUEE, Research Division, Professor, 研究開発部, 教授 (00168238)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
UCHIDA Teruhisa NCUEE, Research Division, Associate Professor, 研究開発部, 助教授 (10280538)
NAKAUNE Naoko Niigata University, Admission Center, Associate Professor, 入試センター, 助教授 (00321563)
SHOJIMA Kojiro NCUEE, Research Division, Research Associate, 研究開発部, 助手 (50360706)
|
|---|
| Project Period (FY) |
2003 – 2005
|
| Project Status |
Completed (Fiscal Year 2005)
|
| Budget Amount *help |
¥6,000,000 (Direct Cost: ¥6,000,000)
Fiscal Year 2005: ¥1,200,000 (Direct Cost: ¥1,200,000)
Fiscal Year 2004: ¥1,200,000 (Direct Cost: ¥1,200,000)
Fiscal Year 2003: ¥3,600,000 (Direct Cost: ¥3,600,000)
|
|---|
| Keywords | Center Test / Testlet / Monitor Survey / Item Response Theory / Standardization / 等化 |
|---|
| Research Abstract |
Testlet model in item response theory is expected to standardize large scale public test in our country. Many of our public test are constituted of some item bundles. Local independence is suspected to hold in item bundles because of the similarity of item implication. Testlet model is expected to be effective when local independence does not hold between items. Graded response model, partial credit model, continuous response model are effective models within testlet theory.
Standardization implies normalization and equating. Normalization means to transform obtained score into a common scale where scaled scores are mutually comparable. Equating is a procedure of making multiple test forms into a common scale so to be comparable. We paid our attention to equating especially. We worked on a continuous response model, and results are reported in the Japanese Journal of Behaviometrics. We also tried to equate between unbalance size of samples like the National Center Test and its supplementary test. The result was reported in Japan Journal of Educational Technology.
|
Report
(4 results)
Research Products
(19 results)
