In particular, you wish to examine the intensity or strength of the relationship between state characteristics (e.g., % impoverished, % uninsured, infant mortality rates, tobacco-related mortality rates, burglary rates)

Instructions: 

Professor Pericles is interested in how state characteristics end up predicting the homicide rate.   Consider the following question:  Is there a link between the unhealthy behaviors people engage in and the crime rate?

In particular, you wish to examine the intensity or strength of the relationship between state characteristics (e.g., % impoverished, % uninsured, infant mortality rates, tobacco-related mortality rates, burglary rates) and state homicide rates (your dependent variable).   In other words, what impact does a state’s social standing (health, crime, etc.) have on the most serious indicator of crime?  More importantly, which indicators appear to exert the strongest effect on homicide rates?  Fortunately, SPSS can again help us out here.

Unlike previous analytical procedures that have been done up to this point, the Pearson Product Moment Correlation Coefficient allows us a mechanism to assess the direction, significance, and strength of the relationship between two variables.  So, you are given the task of determining which measures of state characteristics are significant predictors of homicide, as well as the strength of the relationships.  Moreover, which are positive predictors?  Negative predictors?   Which predictors exert the strongest effect on homicide?  The variables that will be used in the bivariate correlation matrix are listed in the table below. 

In order to investigate these hypotheses, you will be employing the use of the 2012 states data,  which can be found within Learning Module IV in Canvas.   The states-level data set compiles official statistics from various official sources, such as the CDC, Census Bureau, and FBI.  It includes basic demographic data, crime rates, and incidence rates for various illnesses and infant mortality for entire states.

After performing the (ANALYZE/CORRELATE/BIVARIATE), answer the following discussion questions.

1.     Provide the correlation coefficients for ONLY the five state characteristics variables AND state homicide rate.        

2.     Are any of the state characteristics variables significantly correlated with homicide rates, and if so, which measures? 

3.     Specify the direction of the relationships between the five state characteristics variables and murder rates.     

4.     Which of the five state characteristics variables has the most potent (strong) effect on homicide rates?  (Hint:  I am not concerned about the correlations among the state characteristics items, as those will obviously be strong.)

a.     How did you make this determination? 

5.     Does multicollinearity seem to be a problem between any state characteristics items and homicide rates? 

a.     How did you make this determination, and why do you think this is the case?

6.     Talk about the strength of the relationship between various state characteristics and homicide, paying particular attention to the measures of state characteristics that has the strongest impact on homicide rates.  What is the larger takeaway of this as we consider a state’s social and health profiles as a predictor/correlate of the safety within a state?  Hint:  Don’t give me a short answer here – elaborate on these consequences, letting the data/results do your talking for you

7.     Though we assessed strength and direction of the relationship between state characteristics and homicide, what do the preceding analysis have to say with regard to causality?  Why? 

Students should include the syntax files for all procedures used.  Only the class project VIII assignment drop box will actually be evaluated.

Format: 

While there is no specific length/page requirement to this assignment, students are asked to answer the questions comprehensively.  Assignments MUST be typed, with 1” margins throughout the document.  All students must use 12-point, Times New Roman Font.  Students should single-space within each answer (including the letters – e.g., 1a, 1b), but double space between answers (e.g., 1, 2, 3).