More Than Just “Stressful”? Testing the Mediating Role of Fatigue on the Relationship Between Job Stress and Occupational Crashes of Long-Haul Truck Drivers

Figures & data

Table 1 Socio-Labor Features of the Study Sample

Feature	Category	Frequency	Percentage
Gender	Female	17	3.3%
	Male	504	96.7%
Hours driving a day	<5 hours	53	6%
	5–8 hours	280	53.8%
	9–12 hours	207	39.7%
	>12 hours	3	0.5%
Days driving a week	3 or less	4	0.8%
	4	11	2.1%
	5	381	73.1%
	6	103	19.8%
	7	16	3.1%
	There’s no regularity	6	1.1%
Shift working	Yes	219	42.5%
	No	302	57.5%
Type of vehicle driven	Light-mid truck	42	7.5%
	Heavy truck	479	92.5%

Table 2 Goodness-of-Fit Indexes Used for Assessing the Structural Model

Index	Description	Range	Optimal Values
RMSEA^a	RMSEA is a badness-of-fit measure, ie, an absolute fit index allowing to assess how far a hypothesized model is from a perfect model.^Citation66,Citation67	0–1	< 0.080 (better if < 0.060).
CFI^b	Introduced by Bentler (1990).^Citation68 Constitutes a normed fit index comparing the fit of a hypothesized model with that of a baseline model.	0–1	> 0.900 (better if > 0.950).
TLI^b	Measures a relative reduction in misfit per degree of freedom.^Citation67	0–1
NFI^b	The Bentler-Bonett Normed Fit Index assesses model fit through a comparison of the model of reference to a model of completely uncorrelated variables.^Citation69	0–1
IFI^b	The Incremental Fit Index (Δ2) adjusts the normed fit index to the sample size and degrees of freedom of the retained model.^Citation70	0–1
CMIN/df^a	It is the ratio between the Chi-square test value and the degrees of freedom used.	None	< 5.0

Notes: ^aLower values are indicative of a better fit; ^bhigher values indicate a better fit.

Table 3 Bivariate Correlations (Pearson) Between Study Factors

Variable		Mean	SD	2	3	4	5	6	7
1	Age	47.16	8.19	0.079	−0.205**	−0.026	−0.088*	−0.030	−0.146**
2	Daily Shift Intensity (hours)	7.90	1.81	–	−0.089*	0.114*	0.227**	0.179**	0.019
3	General Health Index^a	7.24	1.84		–	−0.468**	−0.235**	−0.441**	−0.075
4	Psychological Distress^b	24.68	4.71			–	0.407**	0.550**	0.178**
5	Job Strain^c	1.05	0.32				–	0.433**	0.135**
6	Work-related Fatigue (NFR)^d	7.67	1.78					–	0.101*
7	Work Traffic Crashes (2 years)	0.36	1.04						–

Notes: ^aScale 0–10; ^bscale 12–48; ^cscores greater than 1.0 indicate job strain/stress; ^dscale 0–11; *correlation is significant at 0.05 level (2-tailed); **correlation is significant at 0.01 level (2-tailed).

Figure 1 3D Graph for assessing the linear relationships among job stress (job strain), fatigue and work-traffic crashes suffered by long-haul truck drivers (LHTD).

Table 4 Variables Included in the Model, Estimates and Significance Levels of the SEM Paths for Explaining Work-Traffic Crashes in a Period of 2 Years

Variable			SPC^a	S.E.^b	C.R.^c	P^d
Age	→	Work-Related Fatigue (NFR)	−0.065	0.016	−1.883	0.060
Daily Shift Intensity (Hours)	→	Work-Related Fatigue (NFR)	0.087	0.071	2.534	*
General Health Index	→	Work-Related Fatigue (NFR)	−0.265	0.078	−6.883	***
Psychological Distress	→	Work-Related Fatigue (NFR)	0.350	0.031	8.793	***
Job Strain	→	Work-Related Fatigue (NFR)	0.205	0.437	5.501	***
Age	→	Work Traffic Crashes	−0.134	0.006	−3.038	**
Daily Shift Intensity (Hours)	→	Work Traffic Crashes	−0.025	0.026	−0.549	0.583
Job Strain	→	Work Traffic Crashes	0.011	0.191	0.188	0.851
Work-Related Fatigue (NFR)	→	Work Traffic Crashes	0.271	0.025	2.988	**

Notes: ^dp-value: *significant at the level p< 0.05; **significant at the level p< 0.01; ***significant at the level p< 0.001.

Abbreviations: ^aSPC, standardized path coefficients (can be interpreted as linear regression weights); ^bS.E., standard error; ^cCR, critical ratio.

Figure 2 Standardized parameter estimates. Solid lines represent significant paths.

Notes: All estimates listed in solid lines are significant (as shown in Table 3); ellipses represent inferred (latent) variables; squares represent observed ones.

Xia Y, Yang Y. RMSEA, CFI, and TLI in structural equation modeling with ordered categorical data: the story they tell depends on the estimation methods. Behav Res Methods. 2019;51(1):409–428. doi:10.3758/s13428-018-1055-2

 PubMed Web of Science ®Google Scholar

Shi D, Lee T, Maydeu-Olivares A. Understanding the model size effect on SEM fit indices. Educ Psychol Meas. 2019;79(2):310–334. doi:10.1177/0013164418783530

 PubMed Web of Science ®Google Scholar

Bentler PM. Comparative fit indexes in structural models. Psychol Bull. 1990;107(2):238–246. doi:10.1037/0033-2909.107.2.238

 PubMed Web of Science ®Google Scholar

Ullman JB. Structural equation modeling. In: Tabachnick BG, Fidell LS, editors. Using Multivariate Statistics. 5th ed. Boston, MA: Pearson Education Inc; 1996:676–780.

 Google Scholar

Bollen KA. A new incremental fit index for general structural equation models. Sociol Methods Res. 1989;17(3):303–316. doi:10.1177/0049124189017003004

 Web of Science ®Google Scholar