Pati et al. (2010) carried out an empirical research study in Philadelphia to find out how maternal health literacy influenced child’s participation in social welfare programs. The sample under study showed that 20% of mothers had limited or no knowledge on their maternal health which in turn, influenced their children’s participation in social welfare programs. The empirical research study specifically aimed at shedding light on how decreasing maternal health knowledge could be linked to the decline in family participation in Temporary Assistance for Needy Families (TANF), Food Stamp Program and Special Supplemental Nutrition Programs. The above programs, out of which 50% of the families participated, were being offered to women and children. In addition, an estimated15% of families in Philadelphia received public housing and childcare subsidies. The study concluded that maternal health literacy affected child participation in social welfare programs either positively or negatively. More than 50% of children whose mothers were literate in health matters were likely to participate in social welfare programs. Conversely, participation in such programs was a mirage where maternal health literacy was low or completely absent (Pati et al., 2010).
In such case studies, there are myriad of risk factors which, if not controlled, are likely to affect outcome of a study (Mitchell & Jolley, 2001). Moreover, risk factors, also referred to as confounding factors, are very important in designing the method of study and also instrumental during statistical analysis of studies that deal with biological and human behavior. According to Shuttleworth (2008), confounding factors impact on results for both observational and controlled studies. From the case study, the researchers offer a contradictory statement when they argue that 38% of interviewed mothers had inadequate education despite having acquired formal learning beyond high school level. In addition, the above mothers had more than one child. The researchers failed to succinctly expound whether the number of children per each mother could also be related to their participation in social welfare programs. This is a confounding variable since making an assumption that a child had been enrolled to social program, while in real sense it did not happen may corrupt the outcome of such a study. In addition, the mother might have gained experience in child care and did not see the need to enroll for social welfare programs.
The researchers also fail to account for the relationship between a child’s age and the probability of being enrolled to a social welfare program. The study did not take into account the variable on the child’s age which by far and large, should be considered to be a confounding factor. The mothers, regardless of health literacy level, may have failed to enroll their children since they were still to be weaned. Shadish, Cook and Campbell (2002) deliberate that in case of a situation where spurious relationships among variables cannot be ruled out, another hypothesis maybe developed.
Finally, the researchers failed to provide an explanation why more mothers with poor health literacy enrolled for special welfare as compared to those with adequate health literacy. At the results analysis stage, the general literacy was equated to mean health literacy. If this is the case, we can identify a confounding factor, namely income. The relationship between literacy and income cannot be overlooked since it can account for a significant number of illiterate women who enroll for the programs. Mothers with higher literacy capability were likely to get well paying jobs and therefore, did not find the need to enroll to the social programs. The effect of health literacy as the independent variable on dependent variable (social welfare program) is not the sole variable that can affect enrollment to social welfare program. According to Levine and Parkinson (1994), a confounding variable also known as a third variable, affects the relationship between the two main variables (independent and dependent). This effect makes the result to be unauthentic as it flaws the analysis results stage.
References
Levine, G. &Parkinson, S. (1994). Experimental Methods in Psychology. Hillsdale, NJ:Lawrence Erlbaum.
Mitchell, M. & Jolley, J. (2001). Research Design Explained (4th Ed). New York: Harcourt.
Pati, S. et al. (2010). Influence of Maternal Health Literacy on Child Participation in Social Welfare Programs: The Philadelphia Experience. American Journal of Public Health, 100:1662–1665.
Shadish, W., Cook, T. Campbell, D. (2002). Experimental and Quasi-Experimental Designs for Generalized Causal Inference, Boston: Houghton Mifflin.
Shuttleworth, Martyn (2008). Confounding Variables. Web.