Property Crimes Introduction: The main purpose of this analysis is to determine the primary factors that
Property Crimes
Introduction:
The main purpose of this analysis is to determine the primary factors that affect property crime rate in the United States. The statistical analysis required in this case, for this type of data, involves using a multiple-regression analysis. Multiple regression analysis creates a meaningful quantitative model which allows us to determine the significant drivers of property crime in the U.S.
The questions that need to be answered in this case are:
- Can a meaningful regression model be found from the sample date to predict the property crime rates in the United States?
- What are the predictors that significantly affect property crime rates in the U.S?
Methods:
As explained in the introduction, the main purpose of this analysis is to determine the primary factors that affect property crime rates in the United States. To this end, a dataset with 50 cases, corresponding to the states in U.S. was used. This dataset contains the following variables (for each of the states):
- STATE: ID variable to identify the corresponding state
- CRIMES: Property crime rate per hundred thousand inhabitants
- PINCOME: Per capital income for each state
- DROPOUT: High school drop out rate (%, in year 1987)
- PUBAID: Percentage of public aid recipients (in year 1987)
- DENSITY: Population/total square miles
- KIDS: Public aid for families with children, in dollars per family
- PRECIP: Average precipitation in inches, in the major city of each state (1950-80)
- UNEMPLOY: Percentage of unemployed workers
- URBAN: Percentage of the residents living in urban areas
All the above variables are thought to have a certain degree of association with the property crime rate, which is going to be the response variable in this analysis. A multiple regression model consists of a quantitative model which has the objective of estimating the value of the dependent variable based on the value of a certain number of predictors.
In this case, here we have eight candidates to be predictors of the variable CRIME in a multiple regression model.
Deliverable: Word Document
