Adopt AI-based contextual similarity scoring for mathematical essay items to ensure consistency and save cost and time¶

Strategy · All strategies
Evidence · 5 claims (5 for) · 1 study (1 associational), q2 · 0 of 1 report an effect size · 5 claims rest on one study

Description¶

The article recommends that teachers, evaluators and examination bodies use AI software such as AEEGS to score mathematical computation essay items, because human scoring of high volumes is subjective, time-consuming and costly. It states that "teachers, evaluators, and various examination bodies can use this instrument to provide consistent scoring for mathematics essays and tests", eliminating inconsistencies and biases and enabling prompt release of results, with saved time redirected to other teaching and learning tasks.

Design Implications¶

Context¶

Requirements¶

• Examination bodies or institutions supply items and marking schemes; students answer extended essay items on computers

Constraints¶

• The article notes students' scores were low partly because they were unaccustomed to answering extended responses on computers

Target Learners¶

• Senior secondary school students sitting certificate examinations with mathematical essay components

Target Learning Goals¶

• Consistent, timely scoring of extended mathematical essay responses in Economics

Claims¶

• AEEGS scores show high absolute agreement with 12 human expert raters' scores on mathematical Economics essay items (average-measures ICC = 0.863) [+W]
• AEEGS and human raters produced nearly equivalent, low mean scores (5.94 vs 5.97 of 20), with AEEGS slightly higher [+W]
• AEEGS and human raters produced similarly shaped score distributions, with most students' scores clustered around the mean in the 'good' band [+W]
• Each of the 12 human raters' scores correlates significantly with AEEGS scores, with coefficients ranging from 0.604 to 0.864 [+W]
• The AEEGS software demonstrated internal consistency of Cronbach alpha 0.86 in trial testing and was validated for face and content validity by three software experts [+W]

Related Strategies¶

Examples¶

-

Key Sources¶

• Olaoye DD, Owolabi HO and Olaoye OT (2026). Automated software scoring of senior school certificate examination mathematical items in economics using a contextual similarity model. Front. Educ. 11:1669504. https://doi.org/10.3389/feduc.2026.1669504