Difference between revisions of "Automated Essay Scoring"

From Penn Center for Learning Analytics Wiki
Jump to navigation Jump to search
Line 3: Line 3:
*Essays written by Hispanic and Asian-American students over-graded than those by White and African American peers.
*Essays written by Hispanic and Asian-American students over-graded than those by White and African American peers.
*inaccurately give Chinese and Korean students significantly higher scores than human essay raters on a test of foreign language proficiency
*inaccurately give Chinese and Korean students significantly higher scores than human essay raters on a test of foreign language proficiency
*Correlate more poorly and bias upwards in terms of GRE essay scores for Chinese students,


Bridgeman, Trapani, and Attali (2012) [[https://www.tandfonline.com/doi/pdf/10.1080/08957347.2012.635502?needAccess=true pdf]]
Bridgeman, Trapani, and Attali (2012) [[https://www.tandfonline.com/doi/pdf/10.1080/08957347.2012.635502?needAccess=true pdf]]

Revision as of 03:25, 24 January 2022

Bridgeman, Trapani, and Attali (2009) [pdf]

  • E-Rater system that automatically grades a student’s essay
  • Essays written by Hispanic and Asian-American students over-graded than those by White and African American peers.
  • inaccurately give Chinese and Korean students significantly higher scores than human essay raters on a test of foreign language proficiency
  • Correlate more poorly and bias upwards in terms of GRE essay scores for Chinese students,

Bridgeman, Trapani, and Attali (2012) [pdf]

  • A later version of E-Rater system for automatic grading of GSE essay
  • Model gave lower scores to African American students than human-raters
  • Chinese students are given higher scores than human essay raters
  • Speakers of Arabic and Hindi were given lower scores

Ramineni & Williamson (2018) [pdf]

  • A later version of E-Rater system for automatic grading of GSE essay
  • for some types of essays, E-Rater gave African American students substantially lower scores than human raters did