Why did you choose this topic? What are you trying to find out?
The increasing crime rate in a region raises te question about the reason for the higher incidence of crime. It has been found that mostly young people get involved in criminal activities. In many cases it has been observed that school or college drop outs have a higher incidence of getting involved in crime.
The objective of this study is to find the factors that influence the crime rate. Specifically we try to find out whether lack of education, remaining out of school or college triggers more criminal activity. We also find the role of police in the rising crime rate in the area of study.
What you think the model will predict. (What do you expect the “b” values for each variable to be? (positive? Negative? Why?)
We have taken one dependent variable, the crime rate. The independent variables are:
X2= Annual police funding in $/resident
X3= Percentage of people 25 years + with 4 years of high school
X4= Percentage of 16 to 19 year olds not in high school and not high school graduates
X5= percentage of 18 to 24 year old in college
X6= percentage of people 25 years + with at least 4 years of college
For X3 and X4 I expect the b values to be positive as lower educational level or being school drop-out increases the possibility of being involved in criminal activities.
X2, X5 and X6 are expected to have negative values. More police funding will give the police more incentive to work intensively and be more vigilant, Thus crime rate will be reduced. X5 and X6 are also expected to be negative because being engaged in studies is supposed to reduce the crime rate among youth.
What are the coefficient parameters (b’s) and what do they mean?
X1= 10.98X2 - 6.09X3 + 5.48X4 +0.38X5 + 5.55X6
We can see from the values of the coefficients that the coefficient of X2 is quite high but it is positive. That means the increase in police funding will actually increase the crime rate. X3 has a negative coefficient suggesting that increase in the percentage of 25+ years of population with 4 years of high school will actually reduce the crime rate. All the other coefficients are positive suggesting that even education in college can increase the crime rate.
Which of these coefficient parameters are individually significant? Are they economically significant?
Only the coefficient of X2 is statistically significant as it has a t value of 3.57. All the other coefficients have a t value of less than 2 thus they are insignificant. Thus we see that the years of high school education or college education does not have any significant impact on the crime rate.
The result is economically significant as well because the increased expenditure on police can only increase the crime rate. Thus the government funding should be used more prudently to control crime rate rather than increasing the police funding.
How well do the predictor variables explain the dependent variable?
The value of multiple R square is 0.57. The R square and adjusted R square are 0.33 and 0.26 respectively. Thus the model has very low predictability. The independent variables can explain only 57% of the dependent varable.
Is the model significant as a whole? How do you know?
The model is not significant as a whole as the F statistic is not significant.
Did the model and the signs and values of the parameter agree with your prediction? If no, why do you think it didn’t? (Was there perhaps a violation of the 5 assumptions?)
The signs of the coefficients were not as expected. It suggests that I have made wrong assumptions and should include some other variables to make the study more comprehensive and meaningful. The independent variables chosen here are not appropriate in explaining the crime rate.
How do you think you might improve your model? Be specific. If you were to add another variable in there, which one would it be and why?
Other variables like the strength of the police force and the percentage od unemployed youth in the area would be relevant variables to be added to explain the dependent variable.
How do you think you might improve your research next time?
I would use a greater sample size and a more comprehensive set of variables. I would also like to include the time dimension here to understand the change in the crime rate over time.
Submit the actual data set as well (and tell me the link where you got the data)
Result: