Respond to a colleague’s post by suggesting two alternative ways that the study results could be generalizable. Please use the resources to support your answer.
This should be answered in 1-2 paragraphs
This program evaluation uses a quantitative, quasi-experimental, time series design to study program outcomes for participants who enrolled “in various forms of public assistance programs” offered through the new TANF (Temporary Assistance to Needy Families) initiative, set forth by the Personal Responsibility and Work Opportunity Reconciliation Act of 1996, which placed limitations on cash benefits and encouraged employment to deter dependency on financial public assistance (Plummer et al., 2014; Yegidis, et al, 2018, p. 108-115).
According to researchers Yegidis et al. (2018) a large sample population size helps to decrease “sampling error,” which is caused when the sample population within the study does not effectively represent the rest of the same population (p. 122). Researchers go on to report that using randomized sampling, coupled with a large sample size, further adds to the external validity to generalize to the greater population because of the probability of including representation of the whole population and lends toward the study’s credibility (pp. 122-124). Because the study was conducted amongst participants in “one of the largest counties in the San Francisco Bay Area” the use of such a large sample population increases the study’s external validity (Plummer et al., 2014; Yegidis, et al, 2018, p. 122).
Plummer et al. (2014) report that the sampling frame of research participants was pulled from recipients of the new TANF program and included “22,000 families… 10,000 [of which] elected to participate in one of the education and training programs, 9,000 elected to attend intensive job placement (Job Club) classes,” as the sample population and the remaining families did not participate in any program and were subjected to the reductions mandated by the new TANF regulations. Additionally, using the stratified random sampling method allowed for the disproportionate participants or dependent variables from subgroups who were exposed to the varying independent variable to be compared in a way that each has the potential for equal representation as not to overrepresent or underrepresent in any way, decreasing the opportunity for sampling errors (Yegidis et al., 2018, pp. 207-213).
Study Results Generalizability
The research results revealed successful outcome for those who participated in education and training programs over the first decade after implementation (although this approach was most expensive) and was subsequently affected by an economic crash that began in 2007 and continued through 2011 causing interference within the study (Plummer et al., 2014).
According to researchers Yegidis et al. (2018) this type of interference is an example of how “history” may distort the internal validity of a study; moreover, these occurrences could have “a major effect on the dependent variable” or sampling frame, which is unforeseeable and must be accounted for within the context of the study threatening its credibility (p. 119). Risks in conducting longitudinal studies include the potential for experimental mortality, which is in losing participants, or in this case historic interferences, which may overshadow any positive effects that the programs had on the participant outcomes; therefore, the results up to the point when threats to the internal validity occurred must be considered because the stratified random sampling would have been effected to a degree that the sampling size was effected by attrition (people losing jobs due to the economy) and the external validity would be jeopardized as well (Yegidis et al, 2018, p.116-124).
Plummer, S. -B., Makris, S., & Brocksen, S. M. (Eds.). (2014). Social work case studies: Foundation year. “Social work research: Program Evaluation.” Baltimore, MD: Laureate International Universities Publishing. [Vital Source e-reader].
Yegidis, B. L., Weinbach, R. W., & Myers, L. L. (2018). Research methods for social workers (8th ed.). New York, NY: Pearson.