Table 3.

Repeated measure multilevel modeling results: DV = safety compliance

		Coefficients (standard error)	95% confidence interval	df	p- value
b0	Intercept	4.98 (.14)**	4.71–5.25	19.26	< .01
b1	Time	− .08 (.03)**	− .14 to − .03	20.74	.01
b2	Safety climate	.27 (.12)*	.03 to .52	322.70	.03
b3	Emotional exhaustion	− .12 (.03)**	− .18 to − .05	551.50	.00
b4	Time × safety climate	.03 (.02)	− .01 to .07	176.30	.13
b5	Time × emotional exhaustion	.02 (.01)*	.00 – .03	207.70	.04
b6	Safety climate × emotional exhaustion	.02 (.01)**	− .17 to − .03	143.00	.01
b7	Gender	− .06 (.08)	− .21 to .09	653.50	.44
b8	Tenure	− .01 (.00)*	− .02 to .00	1119.00	.01
Random effects variance (standard deviation)
e0ijk	Individual-level random intercept = .06 (.25)
e1ijk	Individual-level random slope for time = .01 (.09)
r00jk	Station-level random intercept = .01 (.12)
r01jk	Station-level random intercept = .00 (.01)
u00k	Department-level random intercept = .01 (.12)
u01k	Department-level random intercept = .00 (.02)
Residual variance (standard deviation) = .16 (.41)

^** p < .01, ^* p < .05, degrees of freedom and statistical significance testing were based on Satterthwaite’s approximation method (Hrong-Tai Fai and Cornelius, 1996; Kuznetsova et al., 2017; Satterthwaite, 1946)