Research Question

The aforementioned experimental patrol strategy has theoretical support but there is a lack of empirical evidence to justify widespread adoption. The focus of this evaluation is to establish whether a pilot study provides evidence to support further investment in the patrol strategy. The research question we sought to address was

Is there a difference in the amount of crime before implementation compared to after implementation

Description
A total sample of 60 police beats were used in order to determine the effectiveness of a new patrolling style. The variables of interest in the study are the average crime rates from different police beats which is the dependent variable and the time period (before and after period) as the independent variable. From the 60 police beats, the researcher obtained the following descriptive statistics. The obtained descriptive statistics include the mean, standard deviation and standard error of the data. The following are the descriptive statistics for the data obtained by the researcher.
Table 1

Descriptive Statistics
MeanNStd. DeviationStd. Error Meanaverage monthly crime count in the pre period40.242602.0850.2692average monthly crime count in the post period35.584601.7573.2269The researcher found no missing values from the obtained data. However, the researcher found two extreme values which may hinder in the results of the data analysis.

Methods
In order to conduct the data analysis, the researcher used dependent samples t-test. Dependent samples t-test is used to determine whether there is a difference between the mean of two dependent samples (Howell, 2008). The data of concern involves the dependent variables pre and post average crime rate for different police beats. The null hypothesis for the study is that the average crime rate of pre and post period in different police beats is equal. The alternative hypothesis is that the average crime rate in different police beats for post period is smaller than the pre period.

Before conducting the analysis, the researcher should check for certain assumptions regarding dependent samples t-test. The following assumptions must be checked the data should approximate normal distribution and the data must be dependent (Howell, 2008).

The data of interest for the study is the average crime rate. In order to determine whether the data is dependent, one will look at the correlation between pre and post period average crime. The correlation returned an insignificant result. Thus, the two data variables where not related to each other. The next assumption is the normality of the data. In order to check for normality of the data, the researcher looked at the histogram for pre and post average crime rate. The researcher has found out that data does not approximate normality of the data. Thus, assumptions of the dependent samples t-test were not followed. However, the test will be conducted for the sake of discussion.

Results
Dependent samples t-test was conducted to test the null hypothesis that the mean pre and post period average crime rate is equal. The test was conducted at 0.05 significance level. The decision is to reject the null hypothesis when the p-value of the t-statistic is less than the significance level. Otherwise, the researcher will fail to reject the null hypothesis. After conducting dependent samples t-test, the researcher obtained the following results.

Table 2
Descriptive Statistics
MeanNStd. DeviationStd. Error Meanaverage monthly crime count in the pre period40.242602.0850.2692average monthly crime count in the post period35.584601.7573.2269Table 2
Dependent Samples T-test

Paired DifferencesdfSig. (2-tailed)MeanStd. DeviationStd. Error Mean95 Confidence Interval of the DifferenceLowerUppertaverage monthly crime count in the pre period - average monthly crime count in the post period4.65832.8432.36713.92385.392812.69159.000 From the table, one can see that the t-statistic is equal to 12.691. The corresponding p-value of the t-statistic is less than 0.001 (p  0.001). Since the p-value of the t-statistic is less than the significance level, the researcher rejected the null hypothesis that the mean average crime rate for pre and post period is equal.

Interpretation
The results of the evaluation suggest that the observed differences in the monthly averages between the before and after periods are too large to be explained by chance. To that end, we are confident that the observed reduction in crime is a result of the new patrol strategy.

0 comments:

Post a Comment