Conclusion

This chapter set out techniques for studying the relationship between two variables, especially a linear relationship.

• The scatter diagram gives a visual picture of the relationship and, unlike the numerical coefficients, is not restricted to linear relations.
• Karl Pearson's coefficient of correlation and Spearman's rank correlation both measure linear relationships numerically. Spearman's method is the one to use when the variables cannot be measured precisely and can only be ranked.
• None of these measures implies causation — a correlation tells us only about covariation.

What correlation does give us is a sense of the direction and intensity of the change we can expect in one variable when the variable it is correlated with changes. …