PLS 209                                                                                 Name___________________________
Research Methods
Homework10 Bivariate Analysis of Ratio Level Variables  

COMPUTER INSTRUCTIONS

These instructions work on version 13 of SPSS. Different steps may be required for other versions.

    Step 1 Open PD84USS in SPSS

• From the PLS 209 homepage, Go to the SPSS Download Page
• On the SPSS download page, click-on PD84USS 
• Follow the directions on the download page for unzipping the file and opening it in SPSS

  Step 2 Use SPSS to generate bivariate correlations for 4 variables

From the Data Editor Window select the Analyze Menu

Select Correlate on the Analyze Menu 

Select Bivariate   

Highlight the following variables and move them to the right window:

divindex       Diversity Index
povertpc       Percent of Persons in Poverty
unemploy    Percent of Labor Force Unemployed
viocrime       Violent crime rate per 100,000 inhabitants, 1999

Click on the OK

Print the resulting correlation matrix

Close the output window

    Step 3 Generate the basic univariate descriptive statistics for the 4 variables

From the Data Editor Window select Analyze, Descriptive Statistics, and Descriptives.

Select the options button and add the skewness coefficient to the statistics to be calculated.

Highlight the 4 variables and move them to the right window.

Click on OK and then print the resulting descriptive statistics

    Step 4 Generate Scatterplots with Regression Lines 

Repeat the following instructions (1) first using viocrme and unemploy , (2) then viocrme and divindex, and (3) then viocrme and povertpc.

• Produce a simple scatterplot with viocrme Violent Crime on the Y Axis and unemploy Unemployment on the X Axis
  • Select in sequence Graphs, Interactive, Scatterplot
  • Assign Variables:  Drag unemploy to the X axis, drag viocrme to the Y Axis
  • Fit: Under Method select Regression from the box with the options
  • Click OK
• If the label for the regression line obscures the plot
  • Double click on the scatterplot
  • Drag the label box to the right
  • Close the chart editor window
• Print the output window and then close it
• Repeat Step 4 using povertpc and divindex as the independent variables

QUESTIONS

Use the correlation matrix to answer the following questions:.

1. The Correlation between Violent Crime Rate and divindex is    _____ with a significance level of ______

2. The Correlation between Violent Crime Rate and povertpc  is   _____ with a significance level of ______

3. The Correlation between Violent Crime Rate and unemploy is  _____ with a significance level of ______

4. Which independent variable(s) has/have statistically significant relationship(s) with Violent Crime Rate?
    ________________

5. Which independent variable has the strongest significant relationship with Violent Crime Rate? ___________

6. Which independent variables has a negative relationship with Violent Crime Rate?  _________

Use the three scatterplots and  descriptive statistics to answer the following:

7. The coefficient of determination for Violent Crime Rate and divindex is   ______

8. The coefficient of determination for Violent Crime Rate and povertpc is   ______

9. The coefficient of determination for Violent Crime Rate and unemploy is ______

10. Which independent variable explains the most variation in Violent Crime Rate?  _______  
      How much is explained? ____%

11. Which independent variable appears to have the steepest slope? _____________

12. Do the skewness coefficents suggest outliers are a problem for any variable? __________

13.  If large values are affecting any reported relationships ( r squared, or the slope) for each variable,
        it would be for the relationship between Violent Crime Rate and ___________

Return to PLS 209 Home Page

Larry Aspin    aspin at bradley.edu
Dept of Political Science
Bradley University