Natural language processing (NLP) is located at the intersection of data science, artificial intelligence and linguistics. In a nutshell, it is the process of computers organizing written or spoken human language and analysing it in such a way that the meaning of words is understood by them to a limited extent [1]. One application of NLP is basic sentiment analysis in which NLP scripts automatically assign positive, negative or neutral sentiment to a wide array input texts related to an organization, such as tweets, news articles or customer feedback forms. NLP can greatly reduce the work needed to manually code large amounts of data, enabling researchers and practitioners to analyse large quantities of natural language data related to organizations in an automated and cost-efficient manner. Free, open-source software packages to conduct NLP, such as the Natural Language Toolkit [2], have become more user friendly and offer a wide selection of pre-made functions to run with any given input texts. Since more data is available for data mining than ever before and NLP methods have become increasingly widespread and user-friendly, I would like to highlight some useful applications of NLP in organizational research, as demonstrated by three papers, below.

First, NLP has been used as an alternative to the traditionally survey based measures of organizational culture [3]. The authors successfully compiled and validated a dictionary containing six dimensions of organizational culture based on the annual letters to shareholders and the annual reports of Fortune 500 organizations. Their approach shows how NLP may be used to operationalize theoretical constructs in large scale datasets.

Second, NLP techniques have been applied to measure customer satisfaction [4]. By analysing several thousand natural language customer feedback forms, the authors were able to determine a range of recurring themes in the feedback of positive, negative and neutral opinionated customers. They analysed the relationship between satisfaction of customers and the number of themes addressed in order facilitate the development of automated customer feedback management systems that raise long-term customer satisfaction.

Third, NLP has been used to capture the moral content of texts [5]. The authors compiled the moral foundations dictionary, a list of words related to five distinct dimensions of morality for use with NLP techniques. In my ongoing research, I work with this moral foundations dictionary to capture the extent to which different organizations frame themselves as moral, based on the language used in annual reports and shareholder letters. I then contrast this information with data on an organizations’ financial performance, but also its performance in environmental, social and governance sectors to study the effects of moral framing on organizational performance.

In conclusion, interest in NLP has been rising both in academia and in business as NLP techniques enable researchers and management practitioners to efficiently structure large datasets containing natural language data and produce insights relevant for both academia and business practice as shown in the examples. Because NLP can be applied to any text document related to an organization or even to recorded speech, for example from board meetings or conference calls, I believe that it will benefit organizational behaviour research in the coming years by providing additional and, in combination with data mining, almost instantaneous data gathering options.

Peer Stiegert (p.stiegert rug.nl) is a PhD candidate at the Department of Human Resource Management & Organizational Behavior at the University of Groningen. His research focuses on public perception of organizational scandals, attribution theory and decision-making.

 References

1. Chowdhury, Gobinda G. "Natural language processing." Annual review of information science and technology1 (2003): 51-89.

2. https://www.nltk.org/book

3. Pandey, S., & Pandey, S. K. (2019). Applying natural language processing capabilities in computerized textual analysis to measure organizational culture. Organizational Research Methods, 22(3), 765-797.

4. Piris, Y., & Gay, A. C. (2021). Customer satisfaction and natural language processing. Journal of Business Research, 124, 264-271.

5. Hopp, F. R., Fisher, J. T., Cornell, D., Huskey, R., & Weber, R. (2021). The extended Moral Foundations Dictionary (eMFD): Development and applications of a crowd-sourced approach to extracting moral intuitions from text. Behavior Research Methods, 53(1), 232-246.