Data:
Original intention was to use data from the city-data website. On further investigation it was found that the data is too fragmented. Then I looked at the FBI public data website and found relevant information for the whole of USA. To limit the scope of the project, I decided to stick to the state of NY and the city of NYC. While reading an article I found a reference to data.gov and looked it up for crime data. I found some really good datasets on this website and fortunately found a county wise crime dataset for the state of NY.
The data was very good for population, county, total crime count, violent crime count and property crime count. The data also gave the total crime rate, violent crime rate and property crime rate. The data had missing values in the fire arms count. If the fire arms data was good, it would have been possible to look at correlation with the crime data.
To compare NYC with rest of the state, the data was separated into urban and rural data frames. This was done by separating the counties related to the 5 boroughs of NYC. Also, for studying the time trends, the data was grouped by the year.