Research Methods and Statistics in Criminal Justice
Published 27 Sep 2017
The Criminal Justice System in the United States is charged with the responsibility of executing justice in the society to ensure orderliness. The system is involved in extensive research programs that enable them to understand the society better and offer solutions to the offenders. The Criminal Justice System collects large volumes of data to enable them know the causes that lead to people being involved in criminal activities. They engage in such research to enable them take necessary steps to prevent people from falling into crime.
Data are also collected to determine whether the System is serving justice. The police force and other law enforcement agencies have been mentioned as mistreating the prisoners under their care or general harassment.
The Criminal Justice System employs several research designs to enable them come up with conclusive results. The design that is applied during research determines the validity of the findings. There are several factors that determine the validity of the findings. Among these factors is the sample size which determines whether it is representative of the population. The amount of error allowable is also an important consideration. The variability of the population characteristics need to be carefully evaluated to eliminate the chances of deflating the findings. The research design applied should allow the researchers to answer the research questions with no ambiguity.
One of the research designs that are used in the Criminal Justice research is the classical experimental design. In this research design, the subjects to be studied are grouped into two: the experimental group and the control group, (David & Sutton, 2004). The term experiment in this design means that the researchers apply some stimulus to a group and they evaluate how the group responds to the stimulus. The control group on the other hand does not receive any treatment but is evaluated at the end of some period. The results are later compared for the two groups and if there are differences found, this may mean that the stimulus applied has good or bad effects.
In this design, the observation made after the application of the stimulus or manipulation to the experimental group is the dependent variable. The stimulus applied is the independent variable. The argument in this design is to enable the researcher to know the effect the independent variable has on the dependent variable.
Some defining characteristic of the classical research design is that it has a treatment and a control group. Another characteristic of this design is that random assignment of subjects to treatment and control groups have to be observed. The researcher must be able to manipulate the stimulus and apply them to the treatment group only. The control group is only for comparison purposes and therefore no stimulus is applied to the group. The other characteristic of the classical experiment research design is that must allow the researcher to perform a pretest of the variables of concern and a posttest at the end of the experiment (Worrall, 2005).
In the studies whose objectives are to identify cause-effect relationship, internal validity is of utmost importance. The researchers should limit the interaction of other factors with the subjects to reduce variability. The classical experimental design offers researchers with a strictly controlled experiment such that other factors may not affect the study. This design requires the treatment and control groups to be strictly monitored to eliminate other factors interaction.
The results of such a study would therefore conclusively be said to be due to the stimulus applied. If a cause-effect study allowed other factors to interact with the subjects, researchers may erroneously associate their results to the applied stimulus. Since the Criminal Justice System research is based on testing the workability of certain programs such as probation programs, internal validity is emphasized and classical experimental design is the best suited design for them.
External validity is concerned with whether the results of a study can be generalized for the whole population. The intent of causal relationship research is to assist in understanding the factors that can be changed and by which level to achieve certain results.
The fact that the classical experimental design uses strictly control makes it hard for the design to meet external validity test. The question that persists is whether the effects observed under a controlled experiment would be observed under uncontrolled setup. Most research in the Criminal Justice System takes place in the natural setting of the society which lacks that control. Classical experimental design does not attain much in the external validity test (Maxfield & Babbie, 2005). However, it does not mean that the design should not be used for such studies but that more control should be applied to guard against carrying out a study that cannot be applied anywhere else.
The classical experimental design also has to be evaluated on whether it meets the statistical validity test. The statistical validity test is concerned with the sample size and whether it is representative. It is also concerned with the tests that are used to test the differences. The tests applied for the analysis of classical experimental design are powerful enough to detect the small differences between the treatment and the control groups. This means that the design gives very accurate results. The design is also less costly requiring only a representative small sample to be included.
Though there are a few inadequacies associated with the use of this design for social matters, they do not disregard its importance. The various programs proposed by the Justice System can only be authenticated through such an authoritative design. The design also performs well in validity tests and should only be expanded to cover the few shortcomings.
- David, M and Sutton, C. D (2004). Social Research: The Basics. London: Sage Publications.
- Maxfield, M. G and Babbie, E. R, (2005). Research Methods for Criminal Justice and Criminology. Belmont: Wadsworth Thomson Learning, Inc.
- Worrall, J. L, (2005). Teaching Criminal Justice Students How to Choose Between ResearchDesigns. Academy of Criminal Justice Sciences Today, Vol xxx, No. 2. California State University.