NOTE: For these questions, the y and x variables have been identified for 
you. This will not always be the case. You will be expected to know how to
determine y and x from the wording of the problem.

Question #1:

Find the regression equation for predicting 'final score' from 'midterm
score' where y is 'final score' and x is 'midterm score'. The following 
sample statistics have been computed:
	midterm score:	average=70; standard deviation=10
	final score:	average=55; standard deviation=20
			r=0.60	

Solution:
We first compute the slope and then determine the intercept. 
		slope = 0.60*(20/10) = 1.2
Since (xbar, ybar), that is (70, 55), is on the regression line we can 
plug these values into the regression equation and solve for the 
intercept:
		   55 = a + 1.2(70)
       		    a = 55 - 84 = -29
Therefore the regression equation is:
		final score = 1.2(midterm score) - 29
The interpretation of these coefficients follows: 
   SLOPE:- For unit increase in midterm score an increase of 1.2 points
           can be expected in final score.
   INTERCEPT:- When midterm score is zero final score is -29. This is 
           not meaningful.

Question #2:

For men age 25-34 in the HANES sample, sample statistics for height and
income are given below. Let y be income and x height.
	height:	average=70in; standard deviation=3in
	income: average=$29,800; standard deviation=$14,400
	        r=0.20

Solution:
The approach is the same. In this case income is y and height is x:
		slope = 0.20*(14400/3) = 960
Since (xbar, ybar), that is (70, 29800), is on the regression line we can 
plug these values into the regression equation and solve for the intercept:
		   29800 = a + 960(70)
       		       a = 29800 - 67200 = -37400
Therefore the regression equation is:
		income = 960(height) - 37400
Typically you would interpret the coefficients as follows:
   SLOPE:- For unit increase in the height of men age 25-34 an increase of 
           $960 can be expected in income.
   INTERCEPT:- When height is zero income is $-37400. This is clearly not
           meaningful.
However by thinking carefully about the problem, not only is the intercept 
not meaningful, but the relationship between height and income is spurious.
Also, note the low correlation. This essentially means that only 4% of the
variability in income is explained by height. Given this fact, it is 
reasonable to conclude that this functional relationship is not useful.