Automated text analysis to examine qualitative differences in safety schema among upper managers, supervisors and workers

作者：Colley Sarah K; Neal Andrew^*

来源：Safety Science, 2012, 50(9): 1775-1785.

DOI：10.1016/j.ssci.2012.04.006

• Occupational safety
• Safety climate
• Schema
• Concept mapping
• Qualitative analysis

Differences in people's understanding of the concept of safety within an organization represent a barrier to communication, and may potentially undermine attempts to improve safety. The current study used a qualitative research design to examine whether safety schemas differed between individuals with and without leadership responsibilities. A representative sample of upper managers (N = 6), supervisors (N = 7) and workers (N = 12) were purposively sampled and interviewed. A machine learning algorithm was used to automatically extract concepts and themes from the interview transcripts. Results identified 10 emergent safety climate themes that formed the basis of the safety climate schema. Many of these themes aligned with dimensions of safety climate identified in the academic literature. Results also indicated that safety climate schema of upper managers, supervisors and workers differed. Upper managers were concerned more with themes relating to 'culture' and 'people'; supervisors were concerned more with themes relating to 'corporate values', 'management practices' and 'safety communication'; and workers were concerned more with themes relating to 'procedures' and 'safety training'. Results are discussed in relation to safety climate theory and in terms of how manager; can use this knowledge to improve safety communicate and align safety schemas.

• 出版日期2012-11