Predicting Reported Job Risks Among Latino Day Laborers: A Structural Equation Model

Background and Objective(s):

The purpose of this presentation will be to describe and discuss a psychosocial model predicting job risks tested in a sample of Houston Latino day laborers (LDL). 

Methods:

A community survey was conducted with 327 participants in the Houston metropolitan area in 2008 to explore health and working conditions of LDL. Structural equation modeling was used to test theoretical linkages between psychosocial, mental health and economic conditions and job risks commonly experienced by these workers. Mplus 7.11 was used to conduct all analyses. Latent variables were used to model the constructs of Depression, Daily Hassles, Deportation Fear, SES and Job Risk (outcome variable). Direct and indirect paths were assessed for significance in models that met accepted criteria of fit. 

Results:

Preliminary analyses show that the measurement models for the constructs investigated in the model are adequate; fit is within the acceptable range on CFI and RMSEA and the factor loadings are high and significant on their hypothesized constructs.  There are significant direct relationships between all upstream variables and Job Risk. Both depression and deportation fear also partially mediate the relationship between daily hassles and work risk. SES is not related to any other upstream variables.

Conclusion:

Psychosocial and mental health factors are significantly related to work risk among LDL. SES plays a minor role in predicting work risk. Implications for the development of work-related risk reduction programs will be discussed.