ABSTRACT
Besides, in many instances it is impossible to determine whether these are causes or effects
14.1 Introduction In the previous chapter we discussed tests for the independence, or lack of association, of two categorical variables. If, instead, we are interested in the association o f two quantitative (num erical) variables measured on a random sample of individuals from a population, we may:
(a) Summarize the sample data graphically in a scatter diagram (see Fig. 3.9), where the two variables are 'height’ and 'distance from home’;
(b) Calculate a numerical measure of the strength or degree of asso ciation, called a correlation coefficient;
(c) Carry out a test of the null hypothesis that there is no correlation in the bivariate population from which the sample data were drawn, and interpret the conclusion of this test with great care!
