Timeline for Bayesian analysis of Plausible Values in large-scale education surveys
Current License: CC BY-SA 4.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 18, 2023 at 15:06 | vote | accept | Thomas | ||
| Nov 17, 2023 at 14:47 | answer | added | einar | timeline score: 2 | |
| Nov 16, 2023 at 14:02 | comment | added | Thomas | They are the response: math (and other) scores in case of PISA and TIMSS. | |
| Nov 15, 2023 at 13:17 | comment | added | einar | And what kind of model would you do five times for five plausible values? Are they predictors in a regression, or the response perhaps? | |
| Nov 13, 2023 at 13:06 | comment | added | Thomas | (from [1] Martin, M. O., von Davier, M., & Mullis, I. V. S. (2020). Methods and Procedures: TIMSS 2019 Technical Report. In International Association for the Evaluation of Educational Achievement. International Association for the Evaluation of Educational Achievement. eric.ed.gov/?id=ED610099 | |
| Nov 13, 2023 at 13:05 | comment | added | Thomas | Honestly, the "methods and procedures" documentation[1] goes over my head. :-) It says "In summary, the plausible values used in TIMSS and other large-scale assessments are random draws from a conditional normal distribution [...] that depend on response data x_n as well as context information z_n estimated using a group-specific model for each country g." | |
| Nov 13, 2023 at 12:37 | comment | added | einar | Do you know anything about the distribution from which these plausible values are drawn? | |
| S Nov 13, 2023 at 6:43 | review | First questions | |||
| Nov 13, 2023 at 7:42 | |||||
| S Nov 13, 2023 at 6:43 | history | asked | Thomas | CC BY-SA 4.0 |