Estimating Group Stress Level by Measuring Body Motion

Area of Science:

• Occupational Health
• Data Science
• Organizational Psychology

Background:

• Employee stress is critical for corporate growth and risk management.
• Current methods like annual employee satisfaction surveys (ESs) lack quantitative and continuous data collection.
• A frequent, low-burden measurement for group stress is needed.

Purpose of the Study:

• To propose and evaluate a practical method for frequently measuring group employee stress levels.
• To assess the effectiveness of using body motion data for stress estimation.
• To provide a quantitative management guide for companies.

Main Methods:

• Utilized body motion data to analyze activity/rest duration distributions.
• Developed estimation models on both individual and group bases.
• Evaluated four combinations of modeling approaches.

Main Results:

• Group-based modeling using activity duration distribution yielded optimal results.
• Achieved a high correlation (r = 0.928, p < 0.001) with a low estimation error (1.36%).
• Demonstrated the feasibility of quantitative and easy stress assessment.

Conclusions:

• The proposed method enables quantitative and easy assessment of employee stress.
• This approach can serve as a valuable management guide for companies.
• Body motion data offers a practical solution for continuous organizational health monitoring.