Abstract

The turnover literature is expansive, however empirical evidence on turnover using data science tools remains limited. We propose a novel examination of turnover antecedents--competencies, commitment, trust and values--using big data tools to develop a granular, case-dependent measure of turnover. Using archival data from 700,000 employees of a large organization collected over a decade, we find that turnover changes according to its antecedents' levels. However, in more finegrained analysis, their effect on turnover is contingent upon role, person and cultural background. We discuss turnover implications and the potential of data science methods in the implementation of managerial and HR initiatives.
Original languageEnglish
Pages (from-to)1-6
Number of pages6
JournalAcademy of Management Annual Meeting Proceedings
Volume2021
Issue number1
DOIs
StatePublished - 1 Jan 2021

Fingerprint

Dive into the research topics of 'A MACHINE LEARNING EXAMINATION OF TURNOVER: HIDDEN PATTERNS AND NEW INSIGHTS.'. Together they form a unique fingerprint.

Cite this