NOTE: 	Questions on the quiz will be worded as part b). There are several
	ways of tackling this problem but we may use the approach
	presented in class as illustrated below:
	
b)	We wish to predict 'average score at age 35' from 'average score at
	age 18' and so we should make 'average score at age 35' the
	dependent variable (i.e. y).

	We should first determine the regression equation. Denote 'average
	score at age 35' by y and 'average score at age 18' by x. If a
	represents the intercept and b the slope, then:

			y = a + b(x)

	Since slope is r*(SDy/SDx):
			
			b = 0.80*(15/15) 
		          = 0.8

	Now, the mean is 100 in each case and so we know that (100,100) is 
	on the regression line. This is because one of the properties of the
	regression line is that the point defined by the mean of both 
	variables is on the line. So, substituting in the regression 
	equation:

			100=a + 0.8(100)

	Rearranging terms:

			  a=100-80
			   =20

	Hence our regression equation is:

			y = 20 + 0.8(x) 

	We can now use the regression equation to determine the score at
	age 35 for individuals with a score of 115 at age 18:

			 y = 20 + 0.8(115)
			   = 20 + 92
			   = 112

	Hence the average score at 35 is 112 for individuals who score
	115 at age 18.