Research Design for a new Burglary Prevention ProgramI. Problem StatementBurglary is a frequent crime in most communities. The first line of defense to deter burglars from ransacking homes and properties lie in the homeowners themselves. There are certain practices and ways one can employ to prevent burglars from victimizing their homes.Burglars are, however, a resourceful lot, and would eventually find ways to get into our homes.
This is where the police come in. One of the responsibilities of the police is to ensure not just the safety of the residents in their jurisdiction but the safety of the homes and properties as well.Burglary prevention programs are deployed with the aim of preventing or decreasing the occurrences of burglaries and theft, but if untested prior to deployments, these government-sponsored programs can fall flat on their faces and end up as mere expenditures without any real and verifiable results.
All the efforts put into the programs’ deployment would be meaningless.Experimentation and research is necessary prior to deploying Burglary Prevention programs. Research stems from the need for good knowledge, and good knowledge is objective — it does not rely on the opinions or habits of individual observers. Good knowledge is replicable. This means that any results from research can be replicated should the research be performed again, or if the research is continued to cover larger or different areas. Lastly, good knowledge is communicable. This means that the results can be replicated, the knowledge gained applied, and others can draw information from the research to feed their own need for knowledge (Mitchell & Joley, 2007, ch.
2. lecture 2.2).
To verify the effectiveness of a Burglary Prevention Program means to conduct comprehensive research and experimentation on its effects on the crime rate of communities. The most relevant and effective methods must be used. To ensure this, the research design must be extensive and effective in outlining the necessary research and experiments to be undertaken.The research design holds the research project together like glue. The design shows the structure of the research: the methods, the samples, the variables, and the analysis, that all work together to address the central questions of the research (Trochim, 2006, “Design”).
The research design is the backbone of any research and can either validate or invalidate the results, depending on how the design was conceptualized. The design can make or break a good research project.This paper aims to design a research that would gauge the effectiveness of the new Burglary Prevention Program by carefully studying its effects on the burglary and theft rates of specific communities. II. Literature ReviewLiterature review is done by consulting other works that might be of relevance to the problem that one is researching about. Information that could facilitate the research process might already be available and so conducting the literature review would be beneficial to shorten the research duration.
The literature review will also identify gaps in previous research so that the current research will be able to augment those gaps or avoid the pitfalls of the previous research.In literature review, the following sources can be consulted: journal articles, books, conference proceedings, government and corporate reports, newspapers, internet, and magazines. The researchers will read materials on previous attempts at burglary prevention or details and results of burglary prevention programs in other places.Literature review is necessary in order to understand and investigate a research problem.
When conducting a literature review, two things are actually being done at the same time.1) The research problem is further defined thru the identification of gaps, questioning, and continuing previous research.2) Every source relevant to the research problem is consulted (Asian Institute of Technology Language Center, “How can I write a good Literature Review”).By conducting the literature review, one is able to focus on the relevant information and consequently limit the research area, thereby limiting the factors that would invalidate the research. It could also help the researchers come up with a good research design and conceptualize how the research should be conducted by drawing on the methods employed by previous research done on the field.
III. The Research DesignA good research design will eventually determine the success of a research project, and will ultimately justify the validity of the results.To prove the effectiveness of the new Burglary Prevention Program, we will be employing a Quantitative Research Design as opposed to a Qualitative Research Design.Qualitative Research could be used in studying the behavioral aspects of the people covered by the Burglary Prevention Program, but our research focuses mainly on the effects or the causal relationships associated with the program, or the “what” and not the “how” and “why”. Qualitative Research relies on words or testimonies of respondents and not on mathematical data, and so it does not adequately serve the purpose of our research.According to Burns & Grove, Quantitative Research is “a formal, objective, systematic process in which numerical data are utilised to obtain information about the world” (as cited by Cormack, 1991, p. 140).
The aim of Quantitative Research is to determine relationships between variables in a population (Hopkins, 2000). In this research we will be relying mainly on statistical data to quantify the relationship between two variables. In this case, we shall determine the relationship between the Burglary Prevention Program and the burglary rate.In this research, we are trying to find a causal relationship between these two variables. We want to find out whether the Burglary Prevention Program has an effect on the burglary rate, and so, the Quantitative Research is the best way to go.Our research will basically follow the design outlined in bold dotted strokes in this diagram of Quantitative Research Design (Dereshiwsky, 1998, Lesson 4-1-1):The choice of a research design is dependent on the feasibility.
The best way we can ensure the success of our research is by controlling as many confounding variables as possible in order to reduce errors in our assumptions (Heffner, 2004, ch. 5).The researchers feel that the above-illustrated design is the most feasible way to go for the conduct of this research. Group ComparisonGroup Comparison or Experimental Research uses highly controlled procedures to ensure that the observed changes in the dependent variable is caused by the manipulations of the researcher. All scientific disciplines employ this method to understand the laws of nature (cause-and-effect relationships) (Maricopa Center for Learning and Instruction, 2001).In experimental research, the researcher attempts to maintain control over all variables that would have an effect on the results of the experiment. Simply put, the researcher attempts to predict the outcome of the experiment (Leedy, 1997, p. 185).
For experimental research, we need to have or do 4 things (Campbell & Stanley, 1963, pp. 9-12):1) We need to have a hypothesis for causal relationship. In this research, our hypothesis is that the new Burglary Prevention Program is effective in reducing and preventing the occurrences of burglaries in the community.2) We need to have a Control Group and a Treatment or Test Group. The Control group is the group that will not be receiving the program, thus remaining conditionally unchanged, while the program will be implemented to the Test Group. The Control and Treatment groups will be randomly assigned and must be “statistically equal”, meaning, even if the selection had been random, there variances in the groups’ properties such as population or size, and economic and social standing must not be very large (Cook & Campbell, 1979, pp. 64-65).The assignment of the control and test groups will be discussed further into this design.
3) We need to eliminate the confounding variables that might prevent the display of a causal relationship. A confounding variable is an unforeseen variable that may jeopardize the validity of the experiment’s outcome. For our research, we will attempt to eliminate these variables by employing the Regression Point Displacement (RPD) method of quasi-experimental research.
4) We need to have larger randomized groups so that accidental differences can become negligible and does not invalidate the results. This need will be addressed by the sampling and grouping methods that we shall use in the later part of this design. Quasi-Experimental ResearchStill under the Group Comparison design are research methods that serve individual purposes dependent on what the research aims to achieve. Although the true experimental method is considered the most accurate, it is a very rigid method, and is seldom applicable to societal research.The difference between true experimental and quasi-experimental research is that in true experimental research, individuals who would be subjects in the research are randomly selected. In quasi-experimental, pre-existing groups, instead of individuals, are selected (Dereshiwsky, 1998, lesson 4-1-1).
Quasi-experimental research mimics true experimental research, but only to a certain degree. Randomization is a property inherent in true experimental research that is considered optional in quasi-experimental research.In the case of the Burglary Prevention Program, we cannot randomly select individual homes, due mainly because if we do this, a very big sample size of individual subjects would be needed, else the experiment would become invalid. There is also no guarantee that the selection of individual homes to serve as samples would be representative of the entire population.
For this reason, we shall be utilizing the Quasi-Experimental method of research.There are several quasi-experimental designs available from which we can pattern our experiment. However, the most effective design for our purposes would most likely be the Regression Point Displacement (RPD) Design.The RPD design has important implications especially in the case of community-based research. The problem with community-level research is that there are other factors that can come into play and have an effect on the results of the experiment. To remedy this, the RPD design compares the performance of the treatment or test unit with a large set of comparison units or several control groups (Trochim, 2006, “Other Quasi-Experimental Designs”). Unit of AnalysisThe main unit of analysis we shall be using in our research is the burglary statistics or rate since the effectiveness of the Burglary Prevention program would be best measured by this value.
The VariablesA variable is anything that can vary. In experimental research, mainly two types of variables come into play, the independent variable and the dependent variable. The independent variable serves as the input variable while the dependent variable is the output variable. The independent variable is manipulated to determine its effect on the dependent variable (Heffner, 2004, ch. 7.
2).For the Burglary Prevention Program research, the dependent variable is the burglary rate while the independent variables can consist of time, and the burglary prevention program itself. Another independent variable that could be manipulated to further validate the research is the size of the police force attending to the sample. SamplingInstead of using probability or random sampling, a “matching” method of sampling could be used, wherein sample groups with similar characteristics as the test group are identified and used as the basis for comparison. In this sense, the other groups are not technically control groups, but comparison groups (Neuman & Wiegand, 2000, p. 213).For this experiment, we shall divide the city into several sectors, taking into consideration the demographics of each sector.
The divisions should have almost equal or similar demographics. This can be done by taking the population of each identified sector and comparing it with the others to verify whether the samples are “statistically equal”. Racial distribution should also be considered, and factored into the study.The sectors should then be numbered and a test sector can then be randomly selected. Using the demographics and properties of the selected test sector, the remaining sectors should then be evaluated and sectors “matching” the test sector should be identified. These sectors will then be the comparison sectors. Data CollectionPre-implementationTo have baseline data to serve for the comparison after implementation, the burglary rates of the test and comparison sectors for the last 6 months should be taken. This information can be taken from existing police records.
ImplementationThe program should then be administered to the test sector for at least 4 months, during which burglary rates will be constantly monitored and recorded for this sector as well as the comparison sectors.There should be no changes in police procedures during this implementation period for the comparison sections. No changes in the burglary rates of the comparison sectors are to be expected, but other confounding variables beyond the control of the researchers might arise during implementation.
Such variables could include the population of the sectors (influenced by the migration of residents), the idiosyncrasies of the criminal minds at work (some burglars might decide to suddenly lie low or strike more than usual), and the attitude of the residents of the sectors towards burglaries. Some residents might take it upon themselves to upgrade their home security systems while the study is ongoing and so this becomes a factor that influences our dependent variable as well.To render such confounding variables statistically negligible, it is recommended that the number of samples or sectors for comparison be large enough to still be able to get a valid correlation between our independent and dependent variables. Data AnalysisUsing the Regression Point Displacement Design, the estimate of effectiveness of the program can be measured by comparing the displacement of the test sector’s post-implementation burglary rate with that of the comparison sectors’ pre-post regression line (Trochim & Campbell, in preparation).
The method of data analysis will not be discussed in detail here as this is simply a design and not the actual research report. III. Ethical ConsiderationsThere are various ethical factors to be considered when conducting research on humans or human behavior. It is only ethical to use scientific principles in the study of humans if the potential benefits exceed the potential harm, and if the potential harm has been minimized (Mitchell & Joley, 2007, ch. 2).In the case of the Burglary Prevention Program, the potential benefits that the program offers to the community should it be proven effective can reduce the harm caused by criminality, and so the researchers consider the study or experiment within the ethical bounds of the law.
Some ethical questions might be raised on why the program is administered only to one sector and not to the others, thereby denying the other sectors the potential benefits of the program. It should be kept in mind though, that the choice of the test sector was random and should the program be effective, it will be administered to the other sectors as well. There is no potential harm in denying the program to the other sectors since it does not alter the current situation of the comparison sectors, it only serves to continue the prevailing condition of the sectors. ReferencesAsian Institute of Technology Language Center. Writing up research: Using the Literature. Retrieved July 28, 2004 from http://www.languages.ait.
ac.th/EL21LIT.HTM Burns, N. & Grove, S.K. (1993).
The Practice of Nursing Research: Conduct, Critique, and Utilization (2nd Ed.) Philadelphia: W.B. Saunders. Campbell, D.
T., & Stanley, J. C. (1963). Experimental and quasi-experimental designs for research. Boston: Houghton Mifflin.
Cook, T. & Campbell, D. (1979). Quasi-Experimental Design. Chicago: Rand McNally. Cormack, D.F. (1991).
The research process in nursing (2nd Ed.). Victoria, Australia: Blackwell Scientific Publications. Dereshiwsky, Mary. (1998).
Introduction to Research. Northern Arizona University online course. Retrieved July 28, 2007 from http://jan.ucc.
nau.edu/~mid/edr610/class/design/part1/lesson4-1-1.html Heffner, Christopher. (2004). Research Methods. AllPsych Online.
Retrieved July 28, 2007 from http://allpsych.com/researchmethods/index.html Hopkins, Will G. (2000). Quantitative Research Design. Sportscience 4(1). Retrieved July 28, 2007 from sportsci.
org/jour/0001/wghdesign.html Leedy, P.D. (1997).
Practical research: Planning and design (6th ed.). Upper Saddle River, NJ: Prentice-Hall, Inc., p. 232-233.
Maricopa Center for Learning and Instruction. (2001). Reseach Methods Laboratory. http://www.
mcli.dist.maricopa.edu/proj/res_meth/login.html Mitchell, Mark L. and Jolley, Janina M. (2007).
Instructor’s Site for Research Design Explained. Retrieved July 28, 2007 from http://www.teachRDE.com;Neuman, L.
; Wiegand, B.. (2000). Criminal Justice Research Methods. Boston: Allyn ; Bacon.
;Trochim, William M.K. (2006). Research Methods Knowledge Base.
Retrieved July 28, 2007 from http://www.socialresearchmethods.net/kb/design.php;Trochim, William M.K. ; Campbell, Donald T.
(in preparation). The Regression Point Displacement Design for Evaluating Community-Based Pilot Programs and Demonstration Projects.;;;;