Quantitative Text Analysis: The Promises and Perils of Data Science Techniques for Organisation and Management Research

Ali Bayat, Peter Kawalek

Research output: Working paper

Abstract

Text data is ubiquitous. This type of data is not limited to organizational reports anymore and can come from a wide variety of sources (e.g. Internet Data). Even though beneficial, the traditional quantitative text analysis techniques simply fail to deal with the size and the dimension of the data. The advent of new data science techniques has proven to be promising in manipulating and making sense of highly-dimensional textual data. The aim of this research is to explore advances in text analysis techniques in the field of data science and how they can be used by organisation and management scholars. We draw our attention to automated text retrieval techniques, measure construction and validation, trend analysis and, high-level representation and visualisation of text data. We provide a thorough review of the extant methods used in the organization and management literature, namely in the domains of Computer-Aided Text Analysis and Corpus Linguistics and report on the potential advancements that can be made in each area. We illustrate the application of the methods using a corpus of text featuring letter to shareholders from a sample of S&P 500 list of corporations.
Original languageEnglish
Publication statusPublished - 2017

Keywords

  • Computer-Aided Text Analysis
  • Corpus Linguistics
  • Big Data Analytics
  • Topic Models
  • Machine learning

Fingerprint

Dive into the research topics of 'Quantitative Text Analysis: The Promises and Perils of Data Science Techniques for Organisation and Management Research'. Together they form a unique fingerprint.

Cite this