The family model stress and maternal psychological symptoms: mediated pathways from economic hardship to parenting

Abstract

Although much of the extant research on low-income families has targeted parental depression as the predominant psychological response to economic hardship, the current study examined a range of maternal psychological symptoms that may mediate the relations between early economic pressure and later parenting behaviors. A family stress model was examined using data from 1,142 mothers living in 2 areas of high rural poverty, focusing on the infancy through toddlerhood period. Maternal questionnaires and observations of mother-child interactions were collected across 4 time points (6, 15, 24, and 36 months). Results from structural equation analyses indicated that early economic pressure was significantly related to a variety of symptoms (depression, hostility, anxiety, and somatization), but only depression and somatization were significantly related to decreased levels of sensitive, supportive parenting behaviors. In contrast, anxiety was positively associated with sensitive parenting. Depression and anxiety were both found to mediate the relations between economic pressure and sensitive parenting behaviors. Results further suggest that mothers did not experience change in objective economic hardship over time but did experience a small decrease in economic pressure. Discussion centers on the apparent indirect influence of early economic hardship on later psychological symptoms and parenting behaviors, as well as detailing the need for broader and more complex perspectives on maternal psychological responses that arise as a result of economic disadvantage.

Figure 1.

Path model of relations between economic hardship, economic pressure, symptoms, and parenting. Numbers indicate standardized path coefficients. Only significant paths are shown for ease of readability. Direct paths were included from all covariates to all exogenous variables in the model, but are not shown. Covariances among psychological symptoms are not shown. Model provides an acceptable fit to the data: χ²(2) = 3.92, p = .14; comparative fit index = 1.00; root mean square error of approximation = .03; standardized root mean residual = .01.