Program Evaluation

Next, you try the following: Using the STATES10 data set, examine the relationship between states’ rate of U.S. Military Fatalities in Iraq and Afghanistan as of January 2010 (DFS90) and their 2006 Public High School Graduation Rate (EDS131) 1. What is the research hypothesis? Run a scatterplot and describe what you see. Save your complete output (all exercises) when you are done as a .pdf and submit along with your assignment. Regression Coefficient B for EDS131 ___________________ Significance Level ___________________ Is the relationship significant? YES NO R Square ___________________ In your own words, describe the relationship between DFS90 and EDS131. How would you explain these findings? Note on R squared or Adjusted R Square for multiple regression, do not assume that a low R square is meaningless given that there could be numerous factors that might impact the dependent variable, and if you can explain some small part, it is still useful. Any issues of temporal order or spuriness? Next, try the following: Using the STATES10 data set, examine the relationship between property crime rate, 2008 (CRS48) as predicted by the percent of a population in a state that is living below the poverty level, 2008 (PVS519). 2. What is the research hypothesis? Run a scatterplot and describe what you see. Regression Coefficient B for PVS519 ___________________ Significance Level ___________________ Is the relationship significant? YES NO R Square ___________________ In your own words, describe the relationship between DFS90 and EDS131. How would you explain these findings? Any issues of temporal order or spuriness? Up to this point all you have done is bivariate regression (essentially correlations) with but two variables: one dependent variable and one independent variable. Now, it is time to take full advantage of multiple regression, which allows one to control for other factors that might also impact a dependent variable. Using the STATES10 data set, we shall use multiple regression to predict the crime rate, 2008 (CRS31) using such indicators as homeownership rate, 2008 (ECS445), the divorce rate, 2008 (DMS506), and the personal bankruptcy rate, 2009 (ECS105). Using the STATES10 data set, we shall use multiple regression to predict the teenage pregnancy rates, 2007 (DMS397) using such indicators as poverty, 2008 (PVS519), divorce rate, 2008 (DMS506), high school dropout rate, 2007 (EDS134), and violent crime rates, 2008 (CRS35). Constant ___________________ Adjusted R Square ___________________ Regression Coefficient B for PVS519 ___________________ Significance Level ___________________ Is the relationship significant? YES NO Regression Coefficient B for DMS506 ___________________ Significance Level ___________________ Is the relationship significant? YES NO Regression Coefficient B for EDS134 ___________________ Significance Level ___________________ Is the relationship significant? YES NO Regression Coefficient B for CRS35 ___________________ Significance Level ___________________ Is the relationship significant? YES NO (Note: note that the coefficient does not necessarily mean 0, if it says .00) In your own words, describes these relationships and why they might be (or not be) associated with teenage pregnancy rates.