Comparison Group Only the confidential hazard reporting system described above (#3)
Data Collection, Data Transformation, and Measurement Properties:
6 Intermediate Outcomes
Intermediate1: Airline Safety Culture Index (ASCI; p. 284) scores. ASCI is a survey-based instrument consisting of 25 positively worded statements requiring a response on a 5-point Likert scale ranging from strongly agree to strongly disagree. Content covered the following concepts: management commitment, management action, employee commitment, level of perceived risk, beliefs about accident causation, emergency procedures, provision of safety training and safety communication. Seemed to be based on existing safety climate tools. Was pre-tested with 10 aviation safety professionals. High internal consistency measures (0.94 (T1) and 0.96 (T2)). Test-retest reliability established (r = 0.67). Author gives two citations supporting the link between climate and accidents and there are others that the author didn't cite; i.e. the predictive validity of climate scores has been shown elsewhere, but not shown for this particular tool. Apparently there was an 80% response rate at T1 and T2, though the numbers given for # of respondents at T1 (p. 286) is the same as that given for total employees (p. 283). Intermediate2a and b: Staff hazardousness perception scores and staff hazard likelihood perception scores. p. 284. A list of 22 safety hazards developed from accident reports from 1990-1996. (i.e. content validity). Respondents were asked to rate each hazard as to its hazardousness (2a) and likelihood (2b) using a 9 point scale. The measurement method was based on some citations lacking peer review. No further information on validity and reliability of this particular tool. However, in the discussion (p. 292), it says that “research evidence suggests that there is little correlation between risk perception and actual accident or incident frequencies,” suggesting a lack of predictive validity. Intermediate 3: Staff willingness to report hazards. Number of hazards reports submitted to the confidential hazard reporting system. The period of collection was unspecified and is therefore assumed to be the 8-month intervention period.
Intermediate 4: Action taken on identified safety hazards. Number of hazards identified and addressed by each group.
Intermediate 5: Staff comments on safety management. Employees in both intervention and comparison groups were asked to comment on company safety management both before and after the intervention (method of data collection unspecified).
Barriers Statistical Tests:
Intermediate 1, 2a, 2b: Repeated measures ANOVA. Intermediate 1 was also tested for baseline differences using t-test. Intermediate 3 and 4: No statistical tests.
Effectiveness of Occupational Health & Safety Management Systems: A Systematic Review