1. Open the dataset statedeath.jmp, available on the MathLab 1 server and on the website here. This data contains, among other things, homicide rates (number of homicides per 100,000 people) for each state and whether or not each state has the death penalty. We will investigate if the death penalty effects homicide rates
    
2. Analysis of the data
    
   1. Analyze -> Fit Y by X
       
      This means click on "Analyze" towards the top of the screen and then on "Fit Y by X"
       
   2. Click on "deathpen" and then on the ">> X >>" button
       
   3. Click on "homicide" and then on the ">> Y >>" button
       
   4. A graph should come up. On the "Display" drop down menus (below each graph) make sure that only following items are checked:
       

      	"Show Points"
      	"Quantile Boxes"
      	"Jitter"

   5. Under the "Analysis" drop down menu check:
       

      	"Quantiles"
      	"Means, Std Dev and Std Err"

   6. Summary statistics and graphics will be generated separately for death penalty and non-death penalty states. The graphs are similarly to box-and-whisker graphs. The box represents the IQR and the line across the middle of the box corresponds to the median. These graphs don't have whiskers though. Instead the have horizontal lines at the 10th and 90th percentiles.
        
3. Answer the following questions
    
   1. What are the median homicide rates for death penalty and non-death penalty states?
       
   2. What are the mean homicide rates for death penalty and non-death penalty states?
       
   3. Compare standard deviations and IQR widths between the two groups of states. Which group is more variable?
       
   4. We are considering if the death penalty effects homicide rates. What is the treatment variable and what is the response variable? Is this an observational or designed experiment?
       
   5. Do homicide rates seem to differ between death penalty and non-death penalty states? How do they differ? Does this data convince you that the death penalty effects homicide rates? Why or why not?
       

Hand in the JMP output you used to answer these questions along with your answers.

This page was last modified on October 17, 1999.