เปรียบเทียบวิธี
ดูวิธีที่เลือกเทียบกันแบบเคียงข้าง แถวที่ต่างกันจะถูกเน้นไว้
| Wright Map Analysis× | Many-Facet Rasch Measurement× | |
|---|---|---|
| สาขาวิชา | Education | Education |
| ตระกูล | Latent structure | Latent structure |
| ปีกำเนิด≠ | 2005 | 1989 |
| ผู้ริเริ่ม≠ | Benjamin Wright (Rasch measurement); construct-mapping framing by Mark Wilson | John Michael Linacre |
| ประเภท≠ | Graphical display aligning person abilities and item difficulties on one scale | Rasch model extension adding rater and other facets to person and item |
| แหล่งต้นตำรับ≠ | Wilson, M. (2005). Constructing Measures: An Item Response Modeling Approach. Lawrence Erlbaum Associates. ISBN: 9780805847857 | Linacre, J. M. (1989). Many-Facet Rasch Measurement. MESA Press. ISBN: 9780941938020 |
| ชื่อเรียกอื่น | Item-Person Map, Item Map, Construct Map (Rasch), Variable Map | MFRM, Many-Faceted Rasch Model, Facets Model, Linacre Facets Model |
| ที่เกี่ยวข้อง | 4 | 4 |
| สรุป≠ | A Wright map (item-person map) is the signature graphical output of Rasch measurement: it places persons and items on the same vertical scale, with examinee abilities on one side and item difficulties on the other, both in logits. Because a person succeeds on an item with probability one-half when their ability equals the item's difficulty, this shared scaling lets analysts see at a glance how well a test is targeted to its examinees, what the items reveal about the construct's order, and where measurement is sparse. Named for Benjamin Wright and central to Mark Wilson's construct-mapping approach, it is a primary tool for interpreting and validating measures. | Many-facet Rasch measurement (MFRM) extends the basic Rasch model to assessments mediated by raters. Beyond examinee ability and item difficulty, it adds explicit parameters for rater severity and for any other facet of the rating situation — task, occasion, rating criterion — placing them all on one common logit scale. Developed by John Michael Linacre, MFRM lets analysts estimate and adjust for the fact that some raters are systematically harsh and others lenient, producing 'fair' ability estimates that do not penalize an examinee for happening to draw a severe judge. |
| ScholarGateชุดข้อมูล ↗ |
|
|