Measuring What Matters -- or What's Convenient?: Robustness of LLM-Based Scoring Systems to Construct-Irrelevant Factors

TL;DR

This paper evaluates the robustness of LLM-based scoring systems in educational assessments, finding they are generally resilient to irrelevant factors like spelling errors but sensitive to duplicated text and off-topic responses.

Contribution

It provides empirical evidence on the robustness of LLM-based scoring systems against construct-irrelevant factors, highlighting their strengths and vulnerabilities.

Findings

Robust to spelling errors and writing style variations.

Lower scores for duplicated large passages.

Heavily penalized for off-topic responses.

Abstract

Automated systems have been widely adopted across the educational testing industry for open-response assessment and essay scoring. These systems commonly achieve performance levels comparable to or superior than trained human raters, but have frequently been demonstrated to be vulnerable to the influence of construct-irrelevant factors (i.e., features of responses that are unrelated to the construct assessed) and adversarial conditions. Given the rising usage of large language models in automated scoring systems, there is a renewed focus on ``hallucinations'' and the robustness of these LLM-based automated scoring approaches to construct-irrelevant factors. This study investigates the effects of construct-irrelevant factors on a dual-architecture LLM-based scoring system designed to score short essay-like open-response items in a situational judgment test. It was found that the scoring system was generally robust to padding responses with meaningless text, spelling errors, and writing sophistication. Duplicating large passages of text resulted in lower scores predicted by the system, on average, contradicting results from previous studies of non-LLM-based scoring systems, while off-topic responses were heavily penalized by the scoring system. These results provide encouraging support for the robustness of future LLM-based scoring systems when designed with construct relevance in mind.