This explored the links between patterns of construction crew interactions on how they identify and communicate hazards. Social network analysis was used to model the social interactions.
18 active US construction crews were included in the study. The study specifically:
- Collected hazard recognition and communication data from the crews
- Modelled the safety interactions between the crew members and other situational awareness capability of the crews
The authors cover the rationale of why social interactions are important to study. Evidence is highlighted suggesting that crews that communicate more frequently using formal and informal means and higher quality safety interactions across the organisational hierarchies have been found to have superior safety performance.
Other research found that methods like SWMSs may largely be completed by supervisors and only sent around to the crews for a cursory look and signature.
For hazard recognition, several studies highlight that hazard recognition in construction may miss up to 50% of hazards within the environment. Overall, previous research indicates that hazard recognition levels within construction may be suboptimal.
Results
Based on the network analysis, the authors found “strong evidence suggesting that well-connected crews with higher network density outperform poorly-connected crews in identifying and communicating hazards” (p1752).
That is, crews that had higher network densities were able to identify and communicate a statistically significant higher proportion of hazards linked with their daily work (p1760).
Supervisors with a higher number of connections to other members wasn’t found to have a statistically significant relationship with higher hazard recognition and communication. The study then explored the findings for “betweenness”. Betweenness is defined as the frequency that a member in the network connects two distinct members, e.g. members B, C & D are connected to each other only via member A, so if you removed member A, the others would have no way to communicate (p.1759). For the effect of crews with higher supervisor betweenness versus crews with low betweenness, no statistically significant difference was observed. This suggests that “higher supervisor … betweenness is not necessary for higher [hazard recognition and communication] performance” (p.1760).
It’s suggested that training or coaching interventions focus not just on hazard elements but also “promote efficient and frequent interaction among workers of all trades” (p1752), given what we know about the criticality of social safety.
Authors: Alex ALBERT, Matthew R. HALLOWELL, 2014, Construction Research Congress
Study link: https://doi.org/10.1061/9780784413517.179
Link to the LinkedIn article: https://www.linkedin.com/pulse/modeling-role-social-networks-situational-awareness-ben-hutchinson
SafetyInsights.org 