There are three classifications related to the group research designs. Based on the number of times the variables are measured, they can be cross-sectional, pretest-posttest, and longitudinal (Yegidis et al., 2018). The overall nature distinguishes research designs into exploratory, descriptive, and explanatory. Social work requires three types, which are pre-experimental, experimental, and quasi-experimental.
The article “What’s the relationship got to do with it? Understanding the therapeutic relationship in therapeutic foster care” by Southerland et al. is a study on the relationship between children in Therapeutic Foster Care (TFC) and their treatment parents. Possible study limitations include the inability to establish connections between therapeutic measures and their outcomes due to the cross-sectional research design (Southerland et al., 2009). The internal validity of the research is high as the independent variable was how children would react to the behavioral and emotional assessment, which is only possible within this type of study. External validity was low because the sample size was small, and it is doubtful that findings will apply to larger cases. Subsequently, generalizability is also limited as the participants are specific children not corresponding with the general population.
While the design is cross-sectional, there are indications in the article, which would allow analyzing a pretest-posttest in a case study. Before the research, the authors used the measurements of the mental health status of children from centralized MIS files (Southerland et al., 2009). There was no evaluation after the treatment, although the article was steadfast in articulating the positive impact of a higher quality of the relationship between treatment parents and children on the latter’s healthy functioning.
The effectiveness of the research design is directly associated with study limitations. A cross-sectional study does not allow researchers to make key observations more than once. This means that the study is not effective when it comes to numerous observations of the same group sample, which is admitted by the authors of the research (Southerland et al., 2009). Although there was no loss of study units, it is viable to suggest that attrition would significantly lower the confidence of researchers in the effect of treatment on children. The subsequent lesson is that losing subjects can compromise the study’s internal validity.
When it comes to analyzing the study from the perspective of other classifications, it is pre-experimental. Southerland et al. (2009) specifically label their research as cross-sectional. Pre-experimental designs do not have a comparison group (Yegidis et al., 2018). The same is true for this study, which analyzes only one group sample at a fixed time. The study units themselves are children in TFC and treatment parents.
The researchers are fully aware of the limitations of the study. Mainly, they “were not able to establish causal linkages between the relationship measures and child outcomes” (Southerland et al., 2009, p. 57). Moreover, the article states that further longitudinal studies are necessary to confirm the observations over time. This is not possible by the current study, as it is cross-sectional and limited in terms of available observations.
The major concern about internal validity stems from the lack of longitudinal evaluation. According to Yegidis et al. (2018), behaviors can change over time, thus observations should be repeated. This study assesses children who are prone to change as they mature. This dynamic limits the ability of the study to draw conclusions about its causality. The researchers cannot conclude that causative linkages stand the test of time, as observations are made only once. This constitutes the major limiting aspect of the cross-sectional research design used in the study.
In its turn, the generalizability of the study is compromised by the deficit of existing literature on the treatment method used in the research. Authors admit that Therapeutic Foster Care had not been extensively used for research before. Subsequently, the findings of the study are difficult to replicate due to the low popularity of TFC. Furthermore, as has already been mentioned, the sample size does not represent the general population, negatively affecting generalizability.
References
Southerland, D. G., Mustillo, S. A., Farmer, E. M., Stambaugh, L. F., & Murray, M. (2009). What’s the relationship got to do with it? Understanding the therapeutic relationship in therapeutic foster care.Child and Adolescent Social Work Journal, 26(1), 49-63. Web.
Yegidis, B. L., Weinbach, R. W., & Myers, L. L. (2018). Research methods for social workers (8th ed.). Pearson.