SORA

Advancing, promoting and sharing knowledge of health through excellence in teaching, clinical practice and research into the prevention and treatment of illness

Identifying linguistic markers of CEO hubris: a machine learning approach

Akstinaite, V; Garrard, P; Sadler-Smith, E (2022) Identifying linguistic markers of CEO hubris: a machine learning approach. BRITISH JOURNAL OF MANAGEMENT, 33 (3). pp. 1163-1178. ISSN 1045-3172 https://doi.org/10.1111/1467-8551.12503
SGUL Authors: Garrard, Peter

[img]
Preview
PDF Accepted Version
Available under License ["licenses_description_publisher" not defined].

Download (568kB) | Preview

Abstract

This paper explores the potential of machine learning for recognizing and analysing linguistic markers of hubris in CEO speech. This research is based on three assumptions: hubris is associated with potentially destructive leader behaviours; linguistic utterances are a way of distinguishing between leaders who are likely to exhibit such behaviours; identifying hubris at‐a‐distance using machine learning techniques provides a reliable, automated and scalable method for the identification and prevention of destructive outcomes emanating from CEO hubris. Using machine learning techniques, we analysed spoken utterances from a sample of hubristic CEOs and compared them with non‐hubristic CEOs. We found that machine learning algorithms have the ability to identify automatically hubristic versus non‐hubristic speech patterns. One of the main implications of this study is building a foundation for future studies that are interested in the application of machine learning in the fields of hubristic and other forms of destructive leadership, and in the study of the role that language plays in management and organizations more generally. We discuss the implications of automated data extraction and analysis for the prediction of CEOs’, and other employees’, category membership, intentions and behaviours. We offer recommendations for how hubristic and destructive leadership in organizations can be managed and curtailed more effectively, thereby obviating their negative consequences.

Item Type: Article
Additional Information: This is the peer reviewed version of the following article: Akstinaite, V., Garrard, P. and Sadler-Smith, E. (2022), Identifying Linguistic Markers of CEO Hubris: A Machine Learning Approach. Brit J Manage, 33: 1163-1178, which has been published in final form at https://doi.org/10.1111/1467-8551.12503. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions.
Keywords: 1503 Business and Management, 1505 Marketing, Business & Management
SGUL Research Institute / Research Centre: Academic Structure > Molecular and Clinical Sciences Research Institute (MCS)
Journal or Publication Title: BRITISH JOURNAL OF MANAGEMENT
ISSN: 1045-3172
Dates:
DateEvent
5 July 2022Published
5 April 2021Published Online
15 March 2021Accepted
Publisher License: Publisher's own licence
URI: https://openaccess.sgul.ac.uk/id/eprint/113172
Publisher's version: https://doi.org/10.1111/1467-8551.12503

Actions (login required)

Edit Item Edit Item