Can LLMs Grade Short-Answer Reading Comprehension Questions : An Empirical Study with a Novel Dataset

TL;DR

This study evaluates GPT-4's ability to grade short-answer reading comprehension questions using a new dataset from Ghana, showing near-human performance and highlighting potential for scalable formative assessment.

Contribution

Introduces a novel dataset for short-answer grading in a low-resource context and empirically demonstrates GPT-4's high accuracy in grading student responses.

Findings

GPT-4 achieved QWK 0.92 and F1 0.89, near human performance.

The dataset is from Ghana, expanding evaluation beyond high-income countries.

Minimal prompt engineering sufficed for high-quality grading.

Abstract

Open-ended questions, which require students to produce multi-word, nontrivial responses, are a popular tool for formative assessment as they provide more specific insights into what students do and don't know. However, grading open-ended questions can be time-consuming leading teachers to resort to simpler question formats or conduct fewer formative assessments. While there has been a longstanding interest in automating of short-answer grading (ASAG), but previous approaches have been technically complex, limiting their use in formative assessment contexts. The newest generation of Large Language Models (LLMs) potentially makes grading short answer questions more feasible. This paper investigates the potential for the newest version of LLMs to be used in ASAG, specifically in the grading of short answer questions for formative assessments, in two ways. First, it introduces a novel dataset of short answer reading comprehension questions, drawn from a set of reading assessments conducted with over 150 students in Ghana. This dataset allows for the evaluation of LLMs in a new context, as they are predominantly designed and trained on data from high-income North American countries. Second, the paper empirically evaluates how well various configurations of generative LLMs grade student short answer responses compared to expert human raters. The findings show that GPT-4, with minimal prompt engineering, performed extremely well on grading the novel dataset (QWK 0.92, F1 0.89), reaching near parity with expert human raters. To our knowledge this work is the first to empirically evaluate the performance of generative LLMs on short answer reading comprehension questions using real student data, with low technical hurdles to attaining this performance. These findings suggest that generative LLMs could be used to grade formative literacy assessment tasks.