The purpose of this assignment is to give you a chance to practice performing and interpreting a statistical analysis using any methods you know together with the material in Regression in R
The purpose is not to produce a polished report but if you are interested in doing that, have a look at Structure of a Data Analysis Report copied from a course at Carnegie-Mellon.
Use the data in the ‘Arrests’ data set in the ‘effects’ package. You can get some information on the data and its variables with ?Arrests.
The data was obtained from police records of arrests for marijuana possession between 1997 and 2002. The exact dates are not recorded and the initial and terminal dates are not precisely known. Thus the data for 1997 and 2002 may not cover the entire year.
When police arrest someone for marijuana possession, they can either issue a summons on the spot and immediately release the ‘suspect’ or they can bring the suspect to jail and release them later after additional processing, perhaps appearing before a justice of the peace, for example.
Each record in the Arrests data set represents one arrest. The variable ‘released’ records whether the suspect was released immediately or not. There is concern that the pattern of releases might reflect racial bias. The variable ‘colour’ takes on the values ‘Black’ or ‘White’. I don’t know whether suspects whose race would not normally be considered black or white are omitted or are somehow placed in one of those two categories. You are welcome to do some detective work if you wish or to make a reasonable assumption.
Your task is to explore this data set and see what you can come up with.
Pretend that you have been retained by lawyers for the Toronto Police Association (the police union) who want to have a better understanding of the potential implications of this data.
Prepare an informal report with a technical appendix. Although informal, the report should avoid making rash statements that could not be substantiated. There’s always a chance that your report could be compelled as evidence in court and you wouldn’t want it to be used to embarrass you or to impugn the credibility of your evidence.
Use verbal descriptions and graphics effectively. Put all your code in a separate file or in an appendix.
As you do the analysis, think of the real-world narrative behind the data. For example, during the period in which the data were collected, the police were being criticized in editorial for the apparent differential treatment of different races. They might have changed their behaviour as a result.