Research Design


Prepare a 1,050- to 1,400-word paper describing classical experimental research designs. Address each of the following in your paper:
?Describe the various types of experimental research designs.
?Identify the strengths of experimental research designs.
?Identify the limitations of experimental research designs.
?Provide a summary of two contemporary examples of experimental research conducted within the field of criminal justice within the past 10 years.


Research Design

Research experiments are conducted to determine whether there are casual relationships or not and also whether a certain independent variable determines a certain dependent variable. This is done by carrying out a comparison between a group (with effects) and a certain control group (without effects) and the two groups are compared statistically. Design comprises of several elements such as observations, treatments, groups, the group assignment and time. A design is so useful in structuring the research and shows how groups, treatments and methods solve the research question. (Knight, 2010)

Observations, also known as measures are either a single observation, a single observation with several items or a complex instrument with multiple observations. It can also refer to series of measures observed in one instance.  It is denoted by O. Treatments also referred to as programs refer to either a simple or complex effects. It is denoted by X. The design structure is made up of lines that represent groups.  A design notation with two lines means that there are two groups in the design. Assignment to group shows how groups are organized in the design. These are assignments done at random, cut off or nonequivalent groups. Time shows the order of occurrence. It is from left to the right.

Internal validity shows the truth about casual relationships. It is useful in assessing the effects of interventions or programs. It addresses the question whether the changes observed can be explained by the cause or the intervention and not to alternative explanations for that particular outcome.  In other words, internal validity provides evidence that the study done has caused the outcome to occur or confirms the observation. The relevance of internal validity applies to the study in question only. External validity determines how the effects can be generalized in regard to population, treatment and measurement variables among others.

There are several steps that are taken while conducting experimental study. These are identifying and defining the problem, stating the hypotheses and examine the possible consequences. This is followed by building an experimental design that relates to the elements, the conditions and the outcomes of the study. These include selecting samples of the subjects, grouping the subjects and pay attention to non-experimental factors. In addition to this, build a mechanism to measure the outcomes, perform a pilot survey and take note of the time, place and how long the experiment will last. After this, the experiment is conducted, the data is analyzed and a test of significance is applied appropriately. (Tommola & Lindholm, 1995)

Describe the various types of experimental research designs.

There are several experiments researchers can do to test the hypothesis on whether there is a relationship between independent and dependent variables. These are two group experimental designs, Pre-test/Post-test control group design, and the Solomon Four-Group design.

Two group experimental design. This type is also known as posttest only random experiment. This is one of the simplest experiments. It has two conditions or lines, one for each group. At the beginning of every line there is an R which shows that’s the groups were assigned randomly. One of the groups, which is the treatment is labeled X. this is the independent variable. The other group is the comparison group or the dependent variable. In this particular design, there is no need to carry out a pretest. Normally, a pretest is used to test whether the two groups are comparable or not. In this case, these groups are assigned randomly with the assumption that the groups are equivalent probabilistically. An advantage of this design is that it can incorporate additional groups if required.

Pre-test/Post-test control group design. This type of design is also known as classic controlled experimental design or the randomized pre-test/post-test design. This design uses a table of random numbers to control how subjects in the treatment and control groups are assigned. This makes sure that all the subjects in the experiment have an equal chance of being in either the treatment or control groups. In this design, groups are also assigned randomly. This means that the groups are equal and there are systematic differences between them. The design also controls the groups the independent variable is exposed to and also its timing. The two groups are under the same conditions except for the experimental group which is influenced by the independent variable.  Finally, the design controls all the other factors under which the experiment happens. The design has minimal internal validity and no external validity. (Knight, 2010)

Solomon Four-Group design. This design avoids any challenges brought about by the pretest and posttest designs. The design has two additional control groups which reduces the confounding effect and enables the researcher to determine if the pretest affects the subjects. Although this design is much more complex to come up with and analyze, it deals with issues of internal validity. The design also allows the researcher to control the variables and check the influence of the pretest.

Identify the strengths of experimental research designs.

Experimental design has a number of advantages. One of these is that the researcher has control over the variables. The research controls the independent variables which enables a researcher to remove unwanted variables. Another advantage is that it is easy to determine the relationship of the cause and effect. In addition to this, experimental research designs produce better and reliable results. This only happens when the control set ups and research conditions are met. This can be done over and over again to cross check the results. Good result gives the researcher confidence in his work (Windsor, 2004).

Identify the limitations of experimental research designs

One disadvantage is that it is not easy to do the experiments. There are several issues such as ethical issues that can affect the success of an experiment. There are also artificial situations that come up while the researcher is controlling unwanted variables. These experiments are also subject to human error which might greatly affect the results of the study. There is also a possibility of bias and having samples that are not reliable. In addition to this, the environment can influence the participants. Participants might give the answers they think the researcher wants rather than what they think or feel. Lastly, researchers manipulate variables and observe the results. The manipulation of variables is not objective (McGraw, 1996).

Provide a summary of two contemporary examples of experimental research conducted within the field of criminal justice within the past 10 years.

One of the experimental research designs in the field of criminology and criminal justice is the Domestic Violence Experiment (Sherman). Here, offenders were assigned to three interventions at random. These are arrest, medication, and separation. In another case, a program by the Jersey City Police Department was done in an experimental design method (Braga et al). This program was meant to counter violence. In the study, twenty-four places with high activity and a lot of violence were randomly grouped into twelve groups.  A member of each group was assigned to treatment conditions in a randomized experiment.


Knight, K. L. (January 01, 2010). Study/experimental/research design: much more than statistics. Journal of Athletic Training, 45, 1.)

McGraw, K. (January 01, 1996). Political methodology: research design and experimental methods. A New Handbook of Political Science / Edited by Robert E. Goodin and Hans-Dieter Klingemann.

Tommola, J., & Lindholm, J. (January 01, 1995). Experimental research on interpreting: Which dependent variable?. Topics in Interpreting Research, 121-133.

Windsor, W. L. (January 01, 2004). Data collection, experimental design, and statistics in musical research. Empirical Musicology, 197-222.

Related: Ethical Issues in Criminal Justice Research