| dc.description.abstract | This study examined the variability in the observed relationship between safety
climate and injuries in the extant literature by meta-analytically examining possible
moderators of the safety climate-injury relationship at both the individual and group
levels of analysis. Hypotheses were posited regarding the effects of six moderators:
study design (i.e., retrospective or prospective), the time frame for gathering injury data,
the degree of content contamination and deficiency in safety climate measures, the
source of injury data (i.e., archival or self-report), and the operationalization of injury
severity. Results revealed that the safety climate-injury relationship is stronger at the
group level (? = -.23) than at the individual level of analysis (? = -.18). Meaningful
moderators included the time frame between the measurement of safety climate and
injuries for prospective group-level studies, safety climate content contamination for
group-level studies, and safety climate content deficiency for individual-level studies.
Longer time frames for gathering injury data and safety climate content deficiency were
found to decrease effect sizes while content contamination was associated with stronger
effect sizes. Methodological recommendations are proposed for future research of the safety climate-injury relationship including prospective longitudinal study designs with
data collected and analyzed at the group-level of analysis and injuries operationalized at
a greater level of severity. | en |
