Sample Research Paper on Theft from Vehicle in Retail Parking Lots

Introduction

In relation to the crime triangle, three things, namely the offender, location, and victim must be present at a crime scene. If any of these things lack in the crime triangle, a crime may not occur (Clarke, & Eck, 2014). While the three things in the crime triangle must be present at a crime scene, CPTED strategies on the other hand, strives to prevent the occurrence of crimes, especially in crime prone areas. Current literature on crime indicates that where the three aspects of crime triangle are present, the rates of crimes in such areas are high. Sherman (n.d) claims that majority of the crimes in Minneapolis concentrate in certain hot spots that are not secured. According to the crime triangle, lack of security in such areas exposes the areas to crime. On the other hand, Weisel (2003) proposes three steps towards solving problems in crime prevention strategies. According to her, crime prevention strategies should start by documentation of the problem then proceed to analyze data before concluding by interpreting the findings.

In relation to the prevalence of crime in areas that are not secured, this paper will present a descriptive statistics of the variables that are more likely to reduce crime rates in an area that is prone to crimes. The areas identified in this study did not have security signs and security personnel in year 1, but in year 2, there were security signs and security personnel. With regard to crime prevention, the study will evaluate whether the use of security guards and security signs at crime prone areas as means of reducing the number of crimes in such areas is effective. The study will evaluate some of the CPTED characteristics of the areas under study that expose them to crimes. The assumption is that the crime prevention strategy under study will be effective in reducing crime in the areas under study.

Descriptive Statistics

This study will focus its attention on four variables. The first variable will be the number of security personnel present at the parking lots. The second variable will be the number of security signs present at the parking lots. The third variable will be the crime rate for the first year per every one hundred parking spaces. The fourth variable will be the crime rate for the second year per every one hundred parking spaces.

Ratio variables

With regard to the third and fourth variables that represent the crime rate in the areas under study, I will collect the annual number of crimes at the parking lots from the local police stations in the areas under study. After collecting the number of crimes for each year, I will divide them by the number of parking spaces in the regions under study and multiply each of them by 100. I this case, I will divide the number of crimes by the number of parking spaces because by so doing, I will obtain the rate of crimes in the areas under study.

  Security personnel Rate 1 Rate 2
Total 129 129 129
Mean 0.58 34.8008 23.6692
Median None 17.1540 12.00
Minimum Range 0 0.92 0.46
Maximum Range 4 382.61 232.14
Standard deviation 0.890 54.95293 31.91119

The minimum value for the year 1 is 0.92 while the same value for year 2 is 0.46. On the other hand, the maximum value for year 1 is 382.61 while that for ear 2 is 232.14. At the same time, over 50 percent of the two distributions lie above the median. Finally, in both year 1 and year 2, the standard deviations are higher than a third of respective means. For this reason, the data are skewed and spread widely.

Categorical Variables

The first variable in this study, which is the number of security personnel present at the parking lots, will be measured by determining the number of security guards present at the shopping plazas and stand-alone retail stores. The assumption is that any security guard that will be present at these plazas and retails stores will provide security to the vehicles at the parking lots. The second variable that is the number of security signs present at the parking spaces will be measured by counting the number of security cameras present at the parking lots (Paynich, & Hill, 2011). Given that the number of the parking lots with security signs is not equal to the number of parking lots without security signs; this might affect the outcomes of the study. This might take place because there will be no reasonable comparison between the two groups.

Security signs Frequency Percentage (%)
No 73 56.6
Yes 56 43.4
Total 129 100

 

As it evident from the table, the frequency for the group with security signs is not equal to the frequency for the group without security signs. In this case, the frequency for the group that does not have security signs is 73 while that for the group with security signs is 56. This being the case, the frequency for the group without security signs might bias the results of the study.

Evaluating a crime prevention strategy

            The response being evaluated in this study is whether the presence of security guards and security cameras at crime prone retail parking lots would reduce the number of crimes in such areas. The assumption is that these measures are likely to reduce crime in such areas. This is in relation to what current literature in crime prevention field of study indicated. Accordingly, the researcher expects that the two crime prevention measures would reduce crime in the parking lots under study.

  Pre-test mean Post-test mean Mean difference
Treatment group 32.75 23.07 9.68*
Comparison group 37.16 24.35 12.80*
Mean difference  4.40 1.280  

 

Pre-test comparison of treatment and comparison groups

The purpose of this test will be to determine whether the means for year 1 and year 2 are the same before the implementation of the crime prevention strategy in question. In other words, the test will be trying to establish whether the mean for the treatment group will be similar to the mean for the comparison group before the implementation of the crime prevention strategy in question. Just remember that our intention in this study is to compare the means for the first year and the mean for the second year after the implementation of the crime prevention strategy in question. Accordingly, we want to know whether these two means are significantly different or not at the start of the study.

The test will be an independent t-test because the researcher will be evaluating the outcomes of two separate groups namely the comparison and treatment groups. In other words, the researcher will be evaluating the outcomes of the comparison group as well as the outcomes of the treatment group separately. By so doing, the researcher will be dealing with two independent groups; thus, conduct an independent t-test.

In this case, the null hypothesis will state that there will be no considerable variation between the mean of theft cases for the comparison group and treatment group in the first year and second year. On the contrary, the alternative hypothesis will state there will be considerable difference between the mean of theft cases for the comparison group and treatment group in the first year and second year. If considerable variation will be there between the two means, then we should discard the null hypothesis. On the other hand, if no considerable difference will be there between the two means, then we should retain the null hypothesis (Paynich, & Hill, 2011).

In our case, the mean for the treatment group in the pre-test case is 32.7527 while the mean for the comparison group in the same test is 37.1563. The mean difference between these two groups is 4.40360 whereby the mean for the comparison group is higher than the mean for the treatment group.

In order to interpret the above results, we shall consider the Levene’s test for variance that is on the data results table. If the significance level will be less than 0.05, which is our significance level in the study, then we shall use the unequal variance option. Conversely, if the significance level will be more than 0.05, then we shall use the option of equal variance. With this in mind, then we can see that our p-value is 0.293. Given that this figure is greater than our significance value of 0.05, then we shall use the option of equal variance.

In relation to the above analysis, then the p-value for this test will be 0.652 while the t-test will be 0.453. In this case, we should discard the null hypothesis when the p-value is equivalent or less than 0.05. In contrast, this hypothesis should be retained when the p-value is more than 0.05. In our case, the p-value of 0.652 is more than 0.05. Therefore, we should retain the null hypothesis and conclude that the means for the two groups are the same (Paynich, & Hill, 2011).

The above findings are important because they indicate that the two groups are similar at the start of the study. Accordingly, any mean difference that might be detected later on in this study will be attributed to the crime prevention strategy in question.

Post-test comparison of treatment and comparison groups

While pre-test is an important test before the study, post-test is an important one in the last part of the study. Accordingly, this test will be important in this study because it will establish the effects of implementing the crime prevention strategy in the study. In other words, this test will help in evaluating the outcomes of the crime prevention strategy in the study. By so doing, the researcher will be able to determine whether the crime prevention strategy in question has been effective or ineffective (Gordon, 2012). This explains the purpose of this test in the study.

Once again, this will be an independent t-test because the researcher will be evaluating the outcomes of two separate groups. The two separate groups will be the comparison group that will not include the crime prevention strategy in question while the treatment group will include the crime prevention strategy in question.

The null hypothesis for this test claims that there is no considerable difference between the mean for the comparison group and the mean for the treatment group. Conversely, the alternative hypothesis claims that there is considerable difference between the mean of the comparison group and the mean of the treatment group.

As in the pre-test case, the mean for the treatment group in the post-test case is 23.0738 while the mean for the comparison group is 24.3540. The mean difference between these two groups is 1.28020 whereby the mean of the comparison group is higher than that of the treatment group.

To interpret the above results, we shall consider the Levene’s test for variance that we have on the data results table. If the significance level will be less than 0.05, which is our significance level in the study, then we shall use the unequal variance option. Conversely, if the significance level will be more than 0.05, then we shall use the option of equal variance (Colman, & Pulford, 2011). In our case, the p-value of 0.802 is more than 0.05. For that reason, we should use the option of equal variance.

In relation to the option of equal variance, our p-value will be 0.821 while the t-value will be 0.226. Once again, the researcher should do away with the null hypothesis for p-value less or equal to 0.05. Conversely, the researcher should retain this hypothesis for p-values that exceed 0.05. In this case, we can see that a p-value of 0.821 exceeds 0.05. As a result, we should retain the null hypothesis. This means that we should conclude that the mean for the two groups are the same (Colman, & Pulford, 2011).

In our case, the treatment group had 69 crime cases while the comparison group had 60 crime cases. This being the case, the high number of crime cases in the treatment group has affected the outcomes of the study. For this reason, it appears that the intervention process of the crime may not have impact on the study.

Evaluating results of the treatment group

In relation to the quasi-experimental model, the purpose of this test will be to determine whether the means for the pre-test and post-test will be significantly different from each or they will not be significantly different from each other. If they will be significantly different from each other, then we shall claim that the crime prevention strategy in question will be effective in reducing crime rate in the parking lot. On the contrary, if the means for two groups will not be significantly different from each other, then we shall claim that the crime prevention strategy in question will not be effective in reducing the crime rate in the parking lot (Gordon, 2012).

Given that the researcher is testing the difference between two measures of the same group, then the study is a paired test. In other words, the fact that the researcher is evaluating the mean difference between the pre-test and post-test that are in the treatment group, then the test is a paired test.

The null hypothesis for this test claims that there is no considerable difference between the mean for pre-test and the mean for post-test in the treatment group whereby year 1 is the pre-test while year 2 is the post-test. Conversely, the alternative hypothesis claims that there is considerable difference between the two means. If there will be considerable variation between the two means, then the null hypothesis will be rejected and instead the alternative hypothesis will be accepted. By accepting the alternative hypothesis, we shall be concluding that the crime prevention strategy in question has been effective in curbing crime rate at the parking lot. Conversely, if no considerable variation will be there between the two means, then the null hypothesis will be retained (Colman, & Pulford, 2011).

The mean for the pre-test is 32.7527 while the mean for the post-test is 23.0738. Consequently, the mean difference for the two tests is 9.67887. In this case, the mean for the pre-test is higher than the mean for the post-test.

The p-value for this test is 0.001 while the t-value is 3.498.

Given that the p-value of 0.001 is less than 0.05, then we should do away with the null hypothesis. By rejecting the null hypothesis, then we should accept the alternative hypothesis and conclude that there is significant difference between the theft cases for year 1 and year 2.

In the above case, the mean for year 2 is lower than the mean for year 1, and upon conducting the analysis, the results shows considerable difference between the two means. This indicates that the implementation for the crime prevention strategy in question has been effective in reducing the number of crimes at the parking lots.

Evaluating results of the comparison group

In relation to the quasi-experimental model, the purpose of this test will be to evaluate whether the means for the pre-test and post-test will be similar. If the two means will be similar, then we shall conclude that the crime prevention strategy in question has not been effective in reducing the crime rate at the parking lots, and vice versa.

Once again, this will be a paired test because we shall be testing the difference of two means in the same group namely the comparison group. This fact makes the test a two paired one or simply a two-tailed test.

In this case, the null hypothesis claims that there is no considerable difference between the mean of the pre-test in the comparison group and the mean of the post-test in the same group. Conversely, the alternative hypothesis claims that there is considerable difference between the two means. If there will be considerable difference between the two means, we should do away with the null hypothesis. By so doing, we shall conclude that the crime prevention strategy in question has been effective in curbing crime rate at the parking lots. Conversely, in case there will be no considerable difference between the two means, the null hypothesis will be retained. For such a case, the conclusion will be that the crime prevention strategy in question has not been effective in curbing crime rate at the parking lots (Weathington, Pittenger, & Cunningham, 2013).

The mean for the pre-test is 37.1563 while the mean for the post-test is 24.3540. Consequently, the mean difference for the two tests is 12.80227. In this case, the mean for the pre-test is higher than the mean for the post-test.

The t-value for this test is 2.239 while the p-value is 0.029.

Given that the p-value of 0.029 is less than 0.05, then the null hypothesis is rejected. Once again, by rejecting the null hypothesis, then we should accept the alternative hypothesis and conclude that there is significant difference between the theft cases for year 1 and year 2.

The purpose of conducting this test was to establish whether the mean for the pre-test would be similar to the mean for the post-test. However, the test has established that the mean difference between the two tests is significantly different. These findings indicate that the number of crimes in the area under study has increased.

Effectiveness of the crime prevention response

The fact that the study shows considerable variations between the means for the treatment group implies that the crime prevention strategy implemented in this study has been effective in reducing the crime incidences at the parking lots (Gordon, 2012). This means that the crime prevention strategy in question worked.

Testing hypothesis

The second hypothesis tests whether the presence of security signs at a parking lot reduces the number of crimes at the parking lots with the security signs. The assumption is that the presence of security signs enhances security at such parking lots. On the other hand, the third hypothesis tests the truth of the claim that the more security personnel at a parking lot reduce the number of thefts at a parking lot (Wetcher-Hendricks, 2014).

Hypothesis 2

The second hypothesis claims that if a retail parking lot has more security signs, then the number of thefts in such a parking lot is low. In other words, the more security signs at a parking lot, the less the number of thefts from such a parking lot.

In order to test the second hypothesis, I used an independent t-test to compare and contrast the outcomes of two groups. One group had security signs while the other group did not have the security signs. I used the independent t-test because my two groups were independent.

The null hypothesis stated that there would be no considerable difference between the parking lots at Palm Beach, Broward and Miami-Dade County with security signs and the parking lots in the same regions without the security signs. On the other hand, the alternative hypothesis stated that there would be considerable difference between the parking lots at Palm Beach, Broward and Miami-Dade County with security signs and the parking lots in the same regions without the security signs.

Out of the 129 parking lots in the study, 56 of them had security signs while 73 of them did not have security signs. The mean for the parking lots with security signs was 26.8495 while the mean for the parking lots without the security signs was 40.9005. On the other hand, the standard deviation for the parking lots with security signs was 43.33051 while the standard deviation for the parking lots with the security signs was 62.02785. The mean difference for the two groups was 14.05093. Assuming equal variances, the significance level for the test was 0.151 and the t-value was 1.445 while assuming unequal variances, the significance level for the same was 0.33 and the t-value was 1.513 (Veney, Kros, & Rosenthal, 2009). Therefore, applying Levene’s test, we shall use the option of equal variance assumed because the significance level of 0.151 is higher than 0.05.

At this point, we shall evaluate whether we should reject or accept the null hypothesis under the option of equal variance option. In this case, if the p-value will be less or equal to 0.05, then we shall reject the null hypothesis. On the contrary, if the p-value will be more than 0.05, then we should accept the null hypothesis.

In our case, the p-value of 0.151 is more than 0.05. Therefore, we should accept the null hypothesis that states that there is no considerable difference between the parking lots with security signs and those without the security signs. This means that there lacks sufficient evidence from the available data to conclude that parking lots with security signs have fewer theft cases than parking lots that do not have security signs.

Hypothesis 3

The third hypothesis states that there are fewer theft cases for the parking lots that have more security personnel. In simple terms, this hypothesis presumes that there are fewer theft cases for the parking lots that have security personnel.

To test this hypothesis, I used the correlation test because I was interested in establishing the linear relationship between the ratio variables I used in the study. In this case, I wanted to establish whether there was a linear relationship between these variables as well as determine the strength of that relationship (Carver, & Nash, 2009).

The null hypothesis stated that there would be no linear relationship between the parking lots that had security cameras and the parking lots that did not have the security cameras. On the other hand, the alternative hypothesis stated that there would be a linear relationship between the parking lots that had security cameras and the parking lots that did not have the security cameras.

The correlation coefficient for this test is 0.004 while the p-value for the same is 0.480. The researcher shall use these results to make the conclusion.

Given that the correlation coefficient of 0.004 is too small in comparison to one that indicates a perfect linear relationship between two variables, then we should retain the null hypothesis. This hypothesis states that there would be no linear relationship between the parking lots with security cameras and the parking lots without the security cameras (Bonate, 2000).

The above findings that there is no linear relationship between the parking lots with security cameras and the parking lots without the security cameras supports my original theory that there would be no linear relationship between the two. In simple terms, these findings support my theory that security cameras would reduce crime in crime prone areas.

Conclusion

The implication of this study is that changing the physical environment of a crime-prone area may reduce the number of crimes occurring at such an area. This is based on the actuality the study has established that increasing the number of security signs and security personnel at such areas reduces the number of crimes that take place at such areas. By indicating this, the study contributes to the current literature on crime by showing that the natures of the location at which crimes take place matters a lot in the occurrence of crimes. Future researches should focus their attention on the factors that could weaken the crime triangle with an intention of reducing the number of crimes at crime prone areas.

References

Bonate, P. (2000). Analysis of pretest-posttest designs. Boca Raton: CRC Press.

Carver, R., & Nash, J. (2009). Doing data analysis with SPSS version 16. Belmont, CA: Brooks/Cole Cengage Learning.

Clarke, R., & Eck, J. (2014). Become a problem-solving crime analyst. New York: Routledge.

Colman, A., & Pulford, B. (2011). A Crash Course in SPSS for Windows: Updated for Versions 14, 15, and 16. Hoboken: John Wiley & Sons.

Gordon, R. (2012). Applied statistics for the social and health sciences. New York: Routledge.

Paynich, R., & Hill, B. (2011). Fundamental of crime mapping: principle and practice. Ontario: Jones & Bartlett publishers.

Sherman, L. (n.d). Hot spots of crime and criminal careers of places.

Veney, J., Kros, J., & Rosenthal, D. (2009). Statistics for Health Care Professionals: Working With Excel, Epub Edition. John Wiley & Sons Inc.

Weathington, B., Pittenger, D., & Cunningham, C. (2013). Understanding and conducting research in the health sciences. Hoboken, N.J: Wiley.

Weisel, D. (2003). The sequence of analysis in solving problems. Crime prevention studies, 15; 115-146.

Wetcher-Hendricks, D. (2014). Analyzing Quantitative Data: An Introduction for Social Researchers. Hoboken: Wiley.