Math Isn't Neutral: Designing Word Problems with GPT-4 for Relevance.

Saved in:

Bibliographic Details
Title:	Math Isn't Neutral: Designing Word Problems with GPT-4 for Relevance.
Authors:	Trivedi, Mosum¹
Source:	Ohio Journal of School Mathematics. Spring2026, Vol. 102, p34-47. 14p.
Subject Terms:	Word problems (Mathematics), Educational relevance, Mathematics education, Teaching methods, Social context, Artificial intelligence in education, Geometry, Generative pre-trained transformers
Abstract:	Textbook word problems often miss students' lives; AI can fix that only when teachers stay in the driver's seat. This article shows a practical, repeatable way to use GPT-4 to design mathematically rigorous tasks that feel relevant to students. I present an iterative prompting approach that pairs content goals with two added parameters: [SC] Social Context and [PA] Pedagogical Approach, alongside the familiar task elements (object, shape, properties, target, position). Drawing on examples from a Detroit high school, the paper traces how simple geometry items were reshaped into modeling tasks, civic case studies, and a systems-based investigation of environmental data. For immediate classroom use, the article includes: (1) a step-by-step "try it next week" recipe, (2) a revision checklist for catching quantity/wording issues, and (3) a rubric for analyzing the results. A brief classroom pilot (a trigonometry quiz with local contexts) illustrates gains in student talk, diagramming, and flexible reasoning, as well as common pitfalls and how to revise AI drafts. The goal is not to automate curriculum, but to amplify teacher judgment and help answer the daily question, "Why are we doing this?", with tasks that connect mathematics to the world students inhabit. [ABSTRACT FROM AUTHOR]
	Copyright of Ohio Journal of School Mathematics is the property of Ohio Council of Teachers of Mathematics and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Education Research Complete

Description
Abstract:	Textbook word problems often miss students' lives; AI can fix that only when teachers stay in the driver's seat. This article shows a practical, repeatable way to use GPT-4 to design mathematically rigorous tasks that feel relevant to students. I present an iterative prompting approach that pairs content goals with two added parameters: [SC] Social Context and [PA] Pedagogical Approach, alongside the familiar task elements (object, shape, properties, target, position). Drawing on examples from a Detroit high school, the paper traces how simple geometry items were reshaped into modeling tasks, civic case studies, and a systems-based investigation of environmental data. For immediate classroom use, the article includes: (1) a step-by-step "try it next week" recipe, (2) a revision checklist for catching quantity/wording issues, and (3) a rubric for analyzing the results. A brief classroom pilot (a trigonometry quiz with local contexts) illustrates gains in student talk, diagramming, and flexible reasoning, as well as common pitfalls and how to revise AI drafts. The goal is not to automate curriculum, but to amplify teacher judgment and help answer the daily question, "Why are we doing this?", with tasks that connect mathematics to the world students inhabit. [ABSTRACT FROM AUTHOR]
ISSN:	24725986