Who do you ask? Predicting injury rates from survey responses

Area of Science:

• Occupational Health and Safety
• Industrial-Organizational Psychology
• Risk Management

Background:

• Limited research exists on whether manager or worker perceptions of safety policies alone predict injury rates less accurately than combined perceptions.
• Establishing the predictive accuracy of single-group perceptions versus combined perceptions is crucial for efficient safety assessments.

Purpose of the Study:

• To evaluate the predictive value of worker and manager perceptions of safety policies and practices on reported injury rates.
• To determine if surveying both workers and managers offers additional predictive value compared to surveying only one group.

Main Methods:

• Multilevel logistic regression analysis was employed to estimate the relationship between perceptions and worker injury.
• Three distinct safety scales were used to assess worker and manager perceptions of safety policies and practices.
• The study compared the predictive accuracy of worker-only, manager-only, and combined perceptions of safety.

Main Results:

• Worker perceptions of safety policies were significantly associated with injury rates on two of the three scales.
• Manager perceptions of safety policies were not significantly associated with injury rates.
• Including manager perceptions did not significantly improve injury rate prediction when worker perceptions were already considered.

Conclusions:

• Worker perceptions alone are a significant predictor of workplace injuries.
• Manager perceptions do not add significant predictive value to injury rates when worker perceptions are available.
• The findings support the use of worker perceptions as a primary measure for predicting injury rates, enhancing the efficiency of safety assessments.