Is getting the right answer just about choosing the right words? The role of syntactically-informed features in short answer scoring

TL;DR

This paper investigates whether incorporating syntactic features improves automatic short answer scoring, comparing knowledge engineering methods with lexical-based approaches, and introduces a large new dataset for the task.

Contribution

It introduces the largest available corpus for short answer scoring and explores syntactically-informed features as a less manual alternative to knowledge engineering.

Findings

Syntactic features can enhance scoring accuracy.

Lexical approaches dominate current methods.

The new corpus enables more comprehensive research.

Abstract

Developments in the educational landscape have spurred greater interest in the problem of automatically scoring short answer questions. A recent shared task on this topic revealed a fundamental divide in the modeling approaches that have been applied to this problem, with the best-performing systems split between those that employ a knowledge engineering approach and those that almost solely leverage lexical information (as opposed to higher-level syntactic information) in assigning a score to a given response. This paper aims to introduce the NLP community to the largest corpus currently available for short-answer scoring, provide an overview of methods used in the shared task using this data, and explore the extent to which more syntactically-informed features can contribute to the short answer scoring task in a way that avoids the question-specific manual effort of the knowledge engineering approach.