Difference between revisions of "National Origin or National Location"
Jump to navigation
Jump to search
| Line 10: | Line 10: | ||
*Speakers of Arabic and Hindi were given lower scores | *Speakers of Arabic and Hindi were given lower scores | ||
Ogan and colleagues (2015) [[https://link.springer.com/content/pdf/10.1007/s40593-014-0034-8.pdf pdf]] | Ogan and colleagues (2015) [[https://link.springer.com/content/pdf/10.1007/s40593-014-0034-8.pdf pdf]] | ||
* | *Multi-national model predicting learning gains from student's help-seeking behavior | ||
* | *Both U.S. and combined model performed extremely poorly for Costa Rica | ||
*U.S. model outperformed for Philippines than when trained with its own data set | |||
Li et al. (2021) [[https://arxiv.org/pdf/2103.15212.pdf pdf]] | Li et al. (2021) [[https://arxiv.org/pdf/2103.15212.pdf pdf]] | ||
* Model predicting student achievement on the standardized examination PISA | * Model predicting student achievement on the standardized examination PISA | ||
* Inaccuracy of the U.S.-trained model was greater for students from countries with lower scores of national development (e.g. Indonesia, Vietnam, Moldova) | * Inaccuracy of the U.S.-trained model was greater for students from countries with lower scores of national development (e.g. Indonesia, Vietnam, Moldova) | ||
Revision as of 05:00, 24 January 2022
Bridgeman, Trapani, and Attali (2009) [pdf]
- E-Rater system that automatically grades a student’s essay
- Inaccurately high scores were given to Chinese and Korean students
- System showed poor correlation for GRE essay scores of Chinese students
Bridgeman, Trapani, and Attali (2012) [pdf]
- A later version of E-Rater system for automatic grading of GSE essay
- Chinese students were given higher scores than when graded by human essay raters
- Speakers of Arabic and Hindi were given lower scores
Ogan and colleagues (2015) [pdf]
- Multi-national model predicting learning gains from student's help-seeking behavior
- Both U.S. and combined model performed extremely poorly for Costa Rica
- U.S. model outperformed for Philippines than when trained with its own data set
Li et al. (2021) [pdf]
- Model predicting student achievement on the standardized examination PISA
- Inaccuracy of the U.S.-trained model was greater for students from countries with lower scores of national development (e.g. Indonesia, Vietnam, Moldova)