Perceived Organizational Support for Enhancing Welfare at Work: A Regression Tree Model

Area of Science:

• Organizational Psychology
• Data Science
• Statistical Modeling

Background:

• Traditional research often examines limited variables, potentially missing complex interactions affecting employee welfare.
• Advanced statistical techniques are needed to model multifaceted outcomes like perceived organizational support.
• Recursive partitioning offers a method to analyze complex relationships within large datasets.

Purpose of the Study:

• To demonstrate the utility of recursive partitioning in predicting perceived organizational support.
• To identify key job characteristics influencing perceived organizational support in a large sample.
• To explore interactions between job factors and perceived organizational support.

Main Methods:

• Utilized recursive partitioning via the 'party' package in R for regression tree modeling.
• Analyzed data from over 6000 Italian bankers.
• Included numerous job characteristics such as job demand, control, supervisor support, and training as predictors.

Main Results:

• The regression tree model effectively identified significant predictors and interactions.
• Training was identified as the dominant factor influencing perceived organizational support.
• Perceived communication about safety and stress concerns (reporting) also significantly impacted organizational support.

Conclusions:

• Recursive partitioning is a valuable tool for uncovering complex predictors of organizational support.
• Training and effective communication channels are critical for enhancing perceived organizational support.
• The study highlights the importance of considering multiple job characteristics and their interactions.