An Instructive Guide for Extracting Digital and IT Competence of Firms Using Natural Language Processing on Earnings Call Transcripts
摘要
Earnings Call Transcripts are a valued data source across the fields of accounting, finance, strategic management, corporate governance, and corporate social responsibility. Researchers in the field of information systems have asserted the importance of ECT as a source and resource to understand firms’ digital endowments and strategies. The objective of this chapter is to provide a description and instructive guide for extracting digital and IT-related constructs of firms using natural language processing on ECTs. We explore the implementation of the text-mining and natural language processing techniques of bag-of-words, word2vec, lexical, semantic, and combined analyses to interpret a firm’s proclivity towards the utilization of digital and information technology. We discuss the strengths and limitations of employing ECT. Finally, we provide an extensive explication of alternative approaches and avenues for future research related to the use of earnings call transcripts in the field of information systems.