Research Details

Abstract

Natural language is a key form of communication in the capital markets. Textual analysis is the application of Natural Language Processing (NLP) to textual data for automated information extraction or measurement. We survey publications in top accounting journals and describe the trend and current state of textual analysis in accounting. We organize the available NLP methods into a unified framework. Accounting researchers have often used textual analysis to measure disclosure sentiment, readability, quantity, and forward-looking information; compare disclosures to determine similarities or differences; and detect topical themes. For each of these tasks, we explain the conventional approach and the newer approaches, which are based on machine learning, especially deep learning. Then we discuss the typical decisions faced by researchers in implementing the above approaches and the importance of construct validity of text-based measures. The final part of our article discusses future research opportunities. Our article concludes that (1) textual analysis has grown as an important research method for accounting researchers and (2) accounting researchers should increase their knowledge and use of machine learning, especially deep learning, for textual analysis.

Type

Article

Author(s)

Andrew J. Leone, Khrystyna Bochkay, Stephen V. Brown, Jennifer Wu Tucker

Date Published

2023

Citations

Leone, Andrew J., Khrystyna Bochkay, Stephen V. Brown, and Jennifer Wu Tucker. 2023. Textual Analysis in Accounting: What's Next. Contemporary Accounting Research.(2): 765-805.