Homework for Stats 343
Due: March 19, 2004
You are interested in modeling the relationship between (a) the degree to which teachers of a country approve of corporal punishment and (b) the homicide rate of that country. Let’s assume that a linear model, Y = a + bX, is an appropriate one for modeling this relationship.
1. Use the following data to obtain least-squares estimates of the parameters a and b.
|
Countries |
X (approval of corporal punishment) |
Y (homicide rate) |
|
A |
-1.00 |
1.5 |
|
B |
-0.75 |
1.5 |
|
C |
-0.50 |
1.6 |
|
D |
-0.40 |
1.8 |
|
E |
-0.75 |
2.3 |
|
F |
-0.50 |
2.3 |
|
G |
0.50 |
2.3 |
|
H |
0.70 |
2.3 |
|
I |
1.30 |
2.8 |
|
J |
1.50 |
2.3 |
2. Create a scatterplot of the data.
3. What is the correlation between these two variables?
4. Given the model, and the least-squares estimates of the model’s parameters, what does homicide rate does the model imply for a nation that has a corporal punishment approval score of 3.00?
5. What is the R-squared for the model?