Posting/Due Date: 5:00Pm Monday Weighting: 20% You

Posting/Due date: 5:00pm Monday

Weighting: 20%

You need to complete the following request set by a fictitious professor (any similarity with persons living or dead is entirely deliberate). You can find the dataset on the course website in the Course Content section. You need to provide the Description, Method and Results paragraphs. It should be written in a style that is consistent with the other paragraphs, i.e. To academic journal level.

"Dear student,

I've been struck down by a debilitating illness that has rendered me unable to work or dispense statistical advice. Unfortunately, the evaluation of the targeted policing project is due shortly and I need you to write up some sections. This was the evaluation we were commissioned by QPS into their experimental patrolling strategy. They identified a number of police beats in which they tested a new style of patrolling. Our task was to tell them whether recorded crime went down in the post period compared to the pre-period. Pre and post periods were 12 months and we simply took the annual crime count and computed a monthly average for both pre and post periods.

All I have been able to compile is the Interpretation section of the report. Can you write the Description, Methods and Results findings, please? I've copied what I have below, so just make sure it fits in.

You are already well versed in writing the Description section, so I don't need to explain this. As for the Methods section, the main point is to convince the reader you have used the most appropriate technique for answering the research question, given the data at hand. In other words, you need to justify the particular technique used. This typically involves restating the research question but in a more abstract form (e.g. As we compared four groups, an ANOVA was used...). Also, make sure you make comment on the assumptions the technique requires to be valid. You might include descriptive statements in the Description section, but you'll need to comment in the Methods section to justify why such a technique was chosen.

The Results section is straight forward: you simply report the statistical findings generated by SPSS. Please report this in a form consistent with APA. While QPS probably don't mind how we report it, it is good form to be consistent and they do send reports out for expert review, so let's write this as if we where writing for a journal.

The evaluation is due 5 pm Monday of Week 8. If we (by we, I mean you) don't deliver on time, QPS will commence proceedings for breach of contract!

Pray for a speedy recovery....

Professor Arthur Workshy"

Existing Paragraphs

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

The data we collected is based on monthly averages of reported incidents in each beat for one year pre-implementation and one year post-implementation. This range gives us a wide enough sample to control for normal annual fluctuations in crime rate due to weather or cultural events. Our analysis uses a basic comparison of means to test whether crime in each precinct was different in the year after the implementation of the new crime prevention techniques. Average crime in the pre-implementation period was 40.27, while in the post-implementation period it was 35.52. This suggests that the new targeted policing techniques are having some effect, but further analysis is necessary in order to discover whether it is a statistically significant effect -- i.e. is it worth the community's money to implement targeted policing in other precincts?

Methods

Using the available crime statistics, we computed a monthly average for each precinct. All analyses described below are with respect to these averages. The analysis we performed tested each individual precinct's pre-implementation crime rate vs. its post-implementation crime rate using a paired-samples t-test. Since we had obtained data for each beat for every month and our comparison pitted two conditions against each other, a paired-samples t-test was most appropriate. The suitability of this test was also based on the normal distribution of the data, determined by a Q-Q plot (see below). This plot shows that deviations from expected values within the normal distribution are nonsignificant for either pre- or post-implementation crime rates.