Scholarlink

<p>The aim of the article is to develop a model for predicting organizational commitment based on passive quitting - a mechanism related to an unintentional decline in energy, motivation, and sense of purpose at work. The study employed a psychometrically validated Passive Quitting Scale (PQS) and the UWES-9 engagement scale. Data were collected through a questionnaire survey conducted using the CAWI method on a sample of 1,040 employees in Poland. Selected machine learning algorithms, including linear, kernel-based, and tree-based models, were applied for prediction. The best results were achieved using Gradient Boosting, which was adopted as the reference model. To interpret the model’s functioning, explainable artificial intelligence (XAI) methods were used: SHAP and Permutation Importance. The analyses indicated that the key predictor lowering engagement is PQS6 (“Work no longer gives me satisfaction…”), supported by PQS2 and PQS7. The article contributes theoretically by empirically confirming the role of passive quitting as a predictor of organizational commitment and highlights the value of ML and XAI methods in employee behavior research. The findings also have practical implications, providing managers with tools to identify early signs of declining engagement.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Application of the passive quitting concept in predicting organizational commitment using machine learning

  • 期刊论文

  • 发表日期：

    2026年4月13日

查看全文
  • Marcin Nowak

摘要

The aim of the article is to develop a model for predicting organizational commitment based on passive quitting - a mechanism related to an unintentional decline in energy, motivation, and sense of purpose at work. The study employed a psychometrically validated Passive Quitting Scale (PQS) and the UWES-9 engagement scale. Data were collected through a questionnaire survey conducted using the CAWI method on a sample of 1,040 employees in Poland. Selected machine learning algorithms, including linear, kernel-based, and tree-based models, were applied for prediction. The best results were achieved using Gradient Boosting, which was adopted as the reference model. To interpret the model’s functioning, explainable artificial intelligence (XAI) methods were used: SHAP and Permutation Importance. The analyses indicated that the key predictor lowering engagement is PQS6 (“Work no longer gives me satisfaction…”), supported by PQS2 and PQS7. The article contributes theoretically by empirically confirming the role of passive quitting as a predictor of organizational commitment and highlights the value of ML and XAI methods in employee behavior research. The findings also have practical implications, providing managers with tools to identify early signs of declining engagement.