There is a newer version of the record available.

Published March 18, 2019 | Version 1.5
Software Open

Open Language Chief Executive Personality Tool (OLCPT)

  • 1. Texas Christian University
  • 2. Utah State University
  • 3. Texas A&M University
  • 4. University of Georgia

Description

The Open Language Chief Executive Personality Tool (OLCPT) is a natural language processing (NLP) application that was developed to estimate personality scores of top executives in U.S. public companies based on their speech. Version 1.5 runs in R using supervised machine learning (ML) models built on Word2Vec word embeddings and eXtreme Gradient Boosting (XGBoost) algorithms.

The program works by reading in texts of executives’ speech from a comma-delimited file (.csv) or a set of .txt files, and then applying the ML models to those texts to estimate the executives’ Big Five personality traits. The Big Five traits include openness to experience, conscientiousness, extraversion, agreeableness, and neuroticism (Costa & McCrae, 1985). Traits are scored on a scale from 1 to 7, where 1 indicates low values and 7 indicates high values on each trait.

The program is based on models developed and validated by Harrison and colleagues (2019, 2020). It was trained by identifying key features of CEOs’ language in their responses during the Q&A section of earnings calls with equity analysts and comparing those with valid ratings of the CEOs’ Big Five traits, as scored by trained psychology students from videos of the CEOs using the international personality item pool (Goldberg, 2000). The tool has been shown to be valid and reliable for predicting the Big Five traits of S&P 1500 CEOs. Thus, unless it is validated for use in other samples in the future, it should only be used to assess the Big Five personality traits of top executives of major public, U.S. companies.

References:

Costa, P. T. & McCrae, R. R. 1985. The NEO Personality Inventory. Odessa, FL: Psychological Assessment Resources.

Goldberg, L. R. 2000. International Personality Item Pool. http://www.ipip.ori.org

Harrison, J. S., Thurgood, G. R., Boivie, S., & Pfarrer, M. D. 2019. Measuring CEO personality: Developing, validating, and testing a linguistic tool. Strategic Management Journal, 40(8): 1316-1330.

Harrison, J. S., Thurgood, G. R., Boivie, S., & Pfarrer, M. D. 2020. Perception is reality: How CEOs’ observed personality influences market perceptions of firm risk and shareholder returns. Academy of Management Journal, 63(4): 1166-1195.

 

Files

program-files_olcpt-v1.5.zip

Files (195.4 MB)

Name Size Download all
md5:6f33f22814d5e8a6e31e2bfac0a73ad6
34.0 kB Download
md5:c8d3f84a410f46d4d8b03a268e313e35
205 Bytes Download
md5:0a8cf8decc21b4596267ec419508c937
195.4 MB Preview Download