Research design principles and methodologies

Both qualitative and quantitative research methods have the potential to yield reliable, valid, and important information that can be used to inform public policy. Criminal justice researchers use a wide range of research methods, which vary depending on the research questions, the purpose of the study (applied versus pure research) and the overall paradigm and theoretical framework. Research can be used to validate or disprove an existing theory, alter or enhance an existing theory, generate a new theory based on phenomenology, or evaluate the effectiveness of a policy, program, or intervention (“Criminology Research Methods,” n.d.). Qualitative research methods are best used when the researchers are interested in participants’ perceptions, or seek open-ended answers to questions. Quantitative research methods are best used when the researchers want hard data in numerical form, which can be used to generate statistical analyses. Mixed-methods approaches can be tremendously helpful for criminal justice and social science researchers because of the fluid nature of social reality. Within the two broad categories of qualitative and quantitative methods, researchers have options for how to design their study and test their hypotheses.
Qualitative and quantitative designs share some elements in common. Both qualitative and quantitative research begin with exploring topics of interest, perhaps within established theoretical boundaries. Furthermore, both qualitative and quantitative research starts by either taking an exploratory approach, in which no hypothesis is used, or experimental designs, in which the researcher posits a relationship between two or more variables. Both quantitative and qualitative research methods rely on sampling a population. Ideally, random sampling is used to enhance internal and external validity. Other methods of sampling, such as convenience sampling, can harm validity whether the research is qualitative or quantitative. The main difference between quantitative and qualitative design is that the former yields numerical and/or statistical data, but there are other differences between these two methods that should be apparent to criminal justice researchers. For instance, whereas almost all qualitative research will be values-driven, quantitative research can generate raw data, simply numbers that are not guided by any research questions or the need to apply the results of the study to public policy or criminal justice programming. While there are certainly qualitative research designs that are exploratory and therefore not necessarily linked to a particular research hypothesis, most qualitative studies will be conducted for some practical human interest or relevance for applied criminology (Higgins, 2009).
There are two broad types of criminal justice research, each of which can be sub-divided into different designs. Quantitative research designs include experimental, quasi-experimental, and non-experimental. Cross-sectional designs, cohort designs, causal designs, and longitudinal designs are also quantitative research methods commonly used in criminal justice. Qualitative research designs include case studies, descriptions, observational design, and historical designs (“Research Methods in the Social Sciences,” 2017). Although experimental research designs using double-blind random assignment are widely considered to be the “gold standard” in social science research methods, there are some potential drawbacks with overreliance on experiments (Weisburd, Lum & Petrosino, 2001, p. 53). The drawbacks can be traced to the initial development of the research hypothesis. Researchers can be blinded by their biases, leading to research questions that are either irrelevant or even unethical. Another drawback of the experimental approach is that researchers sometimes use designs that do not accurately measure either an independent or a dependent variable. Operationalizing the variables helps improve validity.
Other potential problems with experimental designs include the constraints on population sampling, plus the need to differentiate between causation and correlation. The experiment may have taken place in artificial settings to ensure adequate controls on the variables, and that artificial research setting might bear little resemblance to the ways people or variables react in a real life situation. There is always the potential for extraneous variables to cloud the validity of the research, and in some cases, the research might be internally valid but not externally valid—leading to results that do not carry over to a broader population or to populations comprised of different types of people.
However, experimental research presents many benefits to criminal justice researchers. An experiment is the only real way to determine causality between two variables. The same study can also be replicated in different situations, using different samples, and different researchers, to enhance the validity and reliability of the results. In one experimental research, Loughran, Wilson, Nagin & Piquero (2015) examined the effects of an intervention on population recidivism. Another study byBarnes, Hyatt & Angel (2013) examines the effect of reintegrative shaming, a restorative justice technique, to measure behavioral outcomes. These two recent criminal justice studies are similar and both use an experimental design, showing how researchers measure different variables using different means but still remaining true to the ethical and professional standards of reliable research.
Research plays an important role in the field of criminal justice. Both qualitative and quantitative research methods are important and can be used to base policies, programs, and practices. Experimental research designs lead to impressive quantifiable results, but there are some drawbacks to the experimental approach. When criminal justice policy makers want to base their interventions on evidence, they can rely on a broad range of research and are not limited to big data sets that are devoid of theory or values contexts.



References

Barnes, G.C., Hyatt, J.M. & Angel, C.M. (2013). Are Restorative Justice Conferences More Fair Than Criminal Courts? Comparing Levels of Observed Procedural Justice in the Reintegrative Shaming Experiments (RISE). Criminal Justice Policy Review 26(2): 103-130.
“Criminology Research Methods,” (n.d.). Criminal Justice. Retrieved online: http://criminal-justice.iresearchnet.com/criminology/research-methods/
Higgins, G.A. (2009). Quantitative vs. qualitative methods. Journal of Theoretical and Philosophical crimninology 1(1). Retrieved online: http://jtpcrim.org/January_Articles/Quantitative_Vs_Qualitative_Methods_George_Higgins.pdf
Loughran, T.A., Wilson, T., Nagin, D.S. & Piquero, A.R. (2015). Evolutionary regression? Journal of Experimental Criminology 11(4): 631-652.
“Research Methods in the Social Sciences,” (2017). Lynn University Library. Retrieved online: http://lynn-library.libguides.com/c.php?g=549455&p=3771805
Weisburd, D., Lum, C.M. & Petrosino, A. (2001). Does research design affect study outcomes in criminal justice? Annals of the AAPSS 578(Nov 2001). Retrieved online: http://cebcp.org/wp-content/publications/Does%20Research%20Design%20Affect%20Study%20Outcomes.pdf