Assessing and Recommending Quantitative Research Designs

Determining the most appropriate research design involves taking into account a variety of factors related to the study. The researcher must consider the research questions, the hypotheses, and whether variables will be utilized. Although making the choice of design can seem daunting, the characteristics of the study will determine its design (Fort-Nachmias & Nachmias, 2008). The purpose of this paper is to assess the strengths and limitations of research designs, and to recommend a quantitative design for my research plan, with a rationale for that recommendation. In addition, its purpose is to explain why other research designs were not chosen.

Strengths and Limitations of Research Designs

Experimental Design
Strengths. Experimental designs are the original, perhaps exemplary models of research. This design allows the researcher to control independent variables and eliminate extraneous variables to a greater extent than in other research designs. Experimental design makes it possible to determine causal relationships since it involves manipulating exposure to independent variables. This allows researchers to observe cause and effect, or the effect of the independent variable on the dependent variable (Fort-Nachmias & Nachmias, 2008). The controlled conditions of experimental design allow for replication, which enables the researcher to verify results. This element of experimental design is crucial for future research as well; when other researchers can replicate an experiment, it provides the research community with a higher level of confidence in the study's results (Fort-Nachmias & Nachmias, 2008).

Limitations. Although experimental design seeks to eliminate or control extraneous variables, this is not always practical or possible. When the intention of research is to study real-world circumstances, an experimental design may not be applicable because of the inability to replicate a natural environment (Levy, Ellis, & Cohen, 2011). This has the potential to weaken external validity. In addition, the selection process necessary to maintain control over variables may not be random. The sample chosen may not be representative of a wider population, and as such, the results may not be generalizeable to the population of interest (Fort-Nachmias & Nachmias, 2008). In many cases, it can be impossible, impractical, or unethical to utilize random assignment to treatment and control groups.

Quasi-Experimental Design

Strengths. Quasi-experimental design allows researchers to investigate behavior in natural settings that may not be amenable to an experimental design. Using naturally existing samples increases the external validity of a study (Fort-Nachmias & Nachmias, 2008; Trochim, 2006). For example, Marques and Lima (2011) studied the effects of living in industrial neighborhoods. For this research, a quasi-experimental design did not require the researchers to utilize random assignment to comparison groups. In a study such as Marques and Lima, researchers could not control who lived in 

Each participant will be assessed with the Beck Depression Inventory Fast Screen for Medical patients.  Comparisons of the results will be made to determine if patients with rare cancer types who participate in online disease specific support groups experience lower levels of depression than patients with rare cancer types who participate in face-to-face support groups.

             

When researchers make group comparisons utilizing quasi-experimental designs, they are creating the comparison groups through non-random processes.  Quasi-experimental design is an alternative to experimental design, which always utilizes a random assignment to groups.  Because quasi-experimental design does not utilize a random selection process for assigning participants to comparison groups, the researcher must be aware of how the non-random selection process might affect the results of the study (Shannon, Goldenhar, & Hale, 2001).  For example, the researcher might ask whether there are inherent differences between participants in the different groups because of age, gender, marital status, socio-economic status, level of education, or other undeterminable differences.  In my study, some of these differences may include length of time since cancer diagnosis, age, geographical location, the particular cancer diagnosis, presence or severity of depression, and other factors that may be difficult to determine.  It will be necessary to note these differences because they have the potential to affect how participants respond to the support group to which they are affiliated (Shannon et al., 2001).  In addition, it will be important to describe and list information on these differences and account for them in the statistical analysis (Shannon et al., 2001). 

Designs not Chosen for Research Plan

Experimental Design
A primary difference between experimental designs and quasi-experimental designs is that the former always involves a control group to which participants are randomly assigned, exposed to treatments imposed by the researcher, and followed by an assessment of the effects of the treatment (Levy, Ellis, & Cohen, 2011). Experimental designs are investigations in which researchers control the impact of independent variables on the dependant variables (Goba, Balfour, & Nkambule, 2011; Levy et al., 2011). Further, experimental designs measure the relationship between the independent and dependent variables (Fort-Nachmias & Nachmias, 2008). The variables in my study cannot be controlled, only observed. My study would not be effectively accomplished with an experimental design because I cannot randomly choose participants. They have already been diagnosed with rare cancer types, and are either utilizing face-to-face support groups or online disease-specific support groups. I cannot exert control over the assignment of participants to comparison groups.

My research question is whether and to what extent participation in online disease-specific support groups evokes a greater sense of control over disease than face-to-face support group participation in patients with rare cancer types. My comparison is between two intact groups. I cannot randomly assign participants to a control and a treatment group. My research question requires that I seek individuals with a rare cancer diagnosis, then assign individuals who participate exclusively in online disease-specific support groups to one group, and those who only participate in face-to-face support groups in a second group. Participants cannot be randomly assigned to one group or the other, and as such, the design must be a quasi-experimental one. The variables, participation in face-to-face support groups and participation in online, disease-specific support groups cannot be supported by random assignment, therefore, an experimental design is not appropriate for my study. The hypothesis of my study, that participation in online, disease-specific support groups evokes a greater sense of control over disease than participation in face-to-face support groups in patients with rare cancer types is not amenable to an experimental and a control group. The study will measure sense of control in participants in each group and identify which independent variable is more effective at evoking a sense of control in participants.

Cross-Sectional Designs

Cross-sectional designs are observational, utilize a random sample of participants, and are often associated with survey research (Fort-Nachmias & Nachmias, 2008). Researchers record information obtained from the survey, although they would not manipulate variables or expose one or more groups to a treatment. For example, a researcher might utilize a cross-sectional design to measure inflammation in exercisers and non-exercisers. Since a cross-sectional design can provide researchers with an effective method of examining several characteristics simultaneously, a cross-sectional study might record characteristics such as age, socio-economic status, and gender of the exercisers and the non-exercisers. Rather than determining causal direction or relationships, this type of research design is descriptive. Unlike experimental design, variables are not manipulated in cross-sectional research designs.

This type of research design would not be amenable to my research question, my dependent and independent variables, and hypothesis. This research design does not support variables. In addition, my hypothesis is not amenable to a survey, and cross-sectional research design does not support observing how the independent variable affects the dependent variable. Cross-sectional designs do not suggest cause-and-effect relationships. I do not intend to manipulate variables, but instead to determine the effects of the independent variable on the dependent variable. This cannot be accomplished in a cross-sectional design.

Conclusion

Choosing a research design depends on the research problems, whether the study utilizes a treatment to which a group will be exposed, its dependent and independent variables, and its hypotheses. Each research design has strengths and limitations inherent in its design. The design which is chosen guides the researcher in the data gathering process, as well as in the analysis and interpretation of the observations that take place throughout the study. Although experimental control and randomization are cornerstones of experimental design, not all inquiries into the social sciences are amenable to such control (Fort-Nachmias & Nachmias, 2008). Quasi-experimental and cross-sectional designs provide alternative means of inquiry when it is not plausible, ethical, or possible to conduct a study in an experimental design.

References

Frankfort-Nachmias, C., & Nachmias, D. (2008). Research methods in the social sciences (7th ed.). New York: Worth.

Goba, B. B., Balfour, R. J., & Nkambule, T. T. (2011). The Nature of Experimental and Quasi- Experimental Research in Postgraduate Education Research in South Africa: 1995-2004. South African Journal Of Higher Education, 25(2), 269-286.

Levy, Y., Ellis, T. J., & Cohen, E. (2011). A Guide for Novice Researchers on Experimental and Quasi-Experimental Studies in Information Systems Research. Interdisciplinary Journal Of Information, Knowledge & Management, 6151-161.

Marques, S., & Lima, M. L. (2011). Living in grey areas: Industrial activity and psychological health. Journal of Environmental Psychology, 31(4), 314-322. doi:10.1016/j.jenvp.2010.12.002

Morgan, G. A., Gliner, J. A., & Harmon, R. J. (2000). Quasi-Experimental Designs. Journal of the American Academy of Child & Adolescent Psychiatry, 39(6), 794-796. doi: 10.1097/00004583-200006000-00020

Shannon, H. S., Goldenhar, L. M., & Hale, A. R. (2001). Chapter 4 Quasi-experimental and experimental designs: More powerful evaluation designs. In L. S. Robson (Author), Guide to evaluating the effectiveness of strategies for preventing work injuries: How to show whether a safety intervention really works. (pp. 29-42). Cincinnati: NIOSH.

Trochim, W. M. (2006). Quasi-experimental design. The Research Methods Knowledge Base. Retrieved December 21, 2013, from http://www.socialresearchmethods.net/kb/quasiexp.php