Finding the right fuel for the analytical engine: Expanding the leader trait paradigm through machine learning?

Brian R. Spisak, Paul A. van der Laken, Brian M. Doornenbal

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

Using self-report personality data and 360-degree performance evaluations of 973 managers across various contexts, we investigated the leader trait paradigm using a range of machine learning methods. We found that a relatively simple linear ordinary least squares model incorporating direct effects of traits and context performed equally as well as our best performing complex machine learning alternatives (e.g., lasso and random forests) at predicting leader effectiveness under low-dimension conditions (i.e., a small number of predictors). We then increased dimensionality and found that newer machine learning methods excelled. Overall, our computationally intensive approach supports the argument that (a) direct effects (not interactions) of traits and context are important predictors of leader effectiveness and (b) appropriately matching combinations of methods, models, and data (from simple and conventional to complex and novel) creates a powerful machine learning engine for investigating leadership. We end with opportunities for future research, discuss practical implications, and provide a list of resources for those interested in learning more about this analytical future.

Original languageEnglish
Pages (from-to)417-426
Number of pages10
JournalLeadership Quarterly
Volume30
Issue number4
Early online date11 Jun 2019
DOIs
Publication statusPublished - 1 Aug 2019

Fingerprint

leader
paradigm
learning method
learning
Least-Squares Analysis
Self Report
Personality
personality
Learning
manager
leadership
Machine Learning
Machine learning
Paradigm
interaction
evaluation
resources
performance
Direct effect
Learning methods

Keywords

  • Context
  • Leader trait paradigm
  • Machine learning

Cite this

Spisak, Brian R. ; van der Laken, Paul A. ; Doornenbal, Brian M. / Finding the right fuel for the analytical engine : Expanding the leader trait paradigm through machine learning?. In: Leadership Quarterly. 2019 ; Vol. 30, No. 4. pp. 417-426.
@article{b72b934a5e764f1eae28948cf3c06925,
title = "Finding the right fuel for the analytical engine: Expanding the leader trait paradigm through machine learning?",
abstract = "Using self-report personality data and 360-degree performance evaluations of 973 managers across various contexts, we investigated the leader trait paradigm using a range of machine learning methods. We found that a relatively simple linear ordinary least squares model incorporating direct effects of traits and context performed equally as well as our best performing complex machine learning alternatives (e.g., lasso and random forests) at predicting leader effectiveness under low-dimension conditions (i.e., a small number of predictors). We then increased dimensionality and found that newer machine learning methods excelled. Overall, our computationally intensive approach supports the argument that (a) direct effects (not interactions) of traits and context are important predictors of leader effectiveness and (b) appropriately matching combinations of methods, models, and data (from simple and conventional to complex and novel) creates a powerful machine learning engine for investigating leadership. We end with opportunities for future research, discuss practical implications, and provide a list of resources for those interested in learning more about this analytical future.",
keywords = "Context, Leader trait paradigm, Machine learning",
author = "Spisak, {Brian R.} and {van der Laken}, {Paul A.} and Doornenbal, {Brian M.}",
year = "2019",
month = "8",
day = "1",
doi = "10.1016/j.leaqua.2019.05.005",
language = "English",
volume = "30",
pages = "417--426",
journal = "The Leadership Quarterly",
issn = "1048-9843",
publisher = "Elsevier Inc.",
number = "4",

}

Finding the right fuel for the analytical engine : Expanding the leader trait paradigm through machine learning? / Spisak, Brian R.; van der Laken, Paul A.; Doornenbal, Brian M.

In: Leadership Quarterly, Vol. 30, No. 4, 01.08.2019, p. 417-426.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - Finding the right fuel for the analytical engine

T2 - Expanding the leader trait paradigm through machine learning?

AU - Spisak, Brian R.

AU - van der Laken, Paul A.

AU - Doornenbal, Brian M.

PY - 2019/8/1

Y1 - 2019/8/1

N2 - Using self-report personality data and 360-degree performance evaluations of 973 managers across various contexts, we investigated the leader trait paradigm using a range of machine learning methods. We found that a relatively simple linear ordinary least squares model incorporating direct effects of traits and context performed equally as well as our best performing complex machine learning alternatives (e.g., lasso and random forests) at predicting leader effectiveness under low-dimension conditions (i.e., a small number of predictors). We then increased dimensionality and found that newer machine learning methods excelled. Overall, our computationally intensive approach supports the argument that (a) direct effects (not interactions) of traits and context are important predictors of leader effectiveness and (b) appropriately matching combinations of methods, models, and data (from simple and conventional to complex and novel) creates a powerful machine learning engine for investigating leadership. We end with opportunities for future research, discuss practical implications, and provide a list of resources for those interested in learning more about this analytical future.

AB - Using self-report personality data and 360-degree performance evaluations of 973 managers across various contexts, we investigated the leader trait paradigm using a range of machine learning methods. We found that a relatively simple linear ordinary least squares model incorporating direct effects of traits and context performed equally as well as our best performing complex machine learning alternatives (e.g., lasso and random forests) at predicting leader effectiveness under low-dimension conditions (i.e., a small number of predictors). We then increased dimensionality and found that newer machine learning methods excelled. Overall, our computationally intensive approach supports the argument that (a) direct effects (not interactions) of traits and context are important predictors of leader effectiveness and (b) appropriately matching combinations of methods, models, and data (from simple and conventional to complex and novel) creates a powerful machine learning engine for investigating leadership. We end with opportunities for future research, discuss practical implications, and provide a list of resources for those interested in learning more about this analytical future.

KW - Context

KW - Leader trait paradigm

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=85067027617&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067027617&partnerID=8YFLogxK

U2 - 10.1016/j.leaqua.2019.05.005

DO - 10.1016/j.leaqua.2019.05.005

M3 - Article

VL - 30

SP - 417

EP - 426

JO - The Leadership Quarterly

JF - The Leadership Quarterly

SN - 1048-9843

IS - 4

ER -