Develop a math word problem corpus to establish a semantic space for evaluating new problems' readability¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The authors recommend building a dedicated corpus of math word problems so that new problems can be evaluated against a domain-specific semantic space, rather than relying on general-text embeddings or traditional readability formulas. They write that "Readability assessment of MWPs may require the development of an MWP corpus of problems to establish a semantic space against which new problems can be evaluated." This follows their finding that custom LSA embeddings from 31,008 MATHia problems outperformed general embeddings in variable importance. They further call for research testing interventions that optimize specific linguistic factors.

Design Implications¶

Context¶

Requirements¶

• A corpus of MWP problems with at least one sentence and more than seven words of text, as used for the custom LSA embeddings

Constraints¶

• The article frames this as a direction for future work, not a tested intervention

Target Learners¶

• less-skilled readers in middle and high school math

Target Learning Goals¶

• improving accessibility and comprehension of math word problems

Related Strategies¶

Examples¶

-

Key Sources¶

• Kole Norberg, Husni Almoubayyed, and Stephen Fancsali. (2025). Linguistic Features Predicting Math Word Problem Readability Among Less-Skilled Readers. Proceedings of the 18th International Conference on Educational Data Mining, Palermo, Italy. https://doi.org/10.5281/zenodo.15870274