Psychology: A Self-Teaching Guide

(a) If scores on one variable can be used to predict scores on a second variable, the variables

are said to.

(b) If increases in variable X suggest increases in variable Y, what kind of correlation is said

to exist?

Answers: (a) covary; (b) A positive correlation.

The magnitude of a correlation is measured with the use of the correlation

coefficient,a statistical tool developed by the mathematician Karl Pearson about

one hundred years ago in association with the researcher Francis Galton. Galton

used Pearson’s tool to measure the correlation between the eminence of fathers

and that of their sons. He found that eminent fathers tended to have eminent sons.

For example, a father who was a judge might have a son who was an army gen-

eral. Obscure fathers tended to have obscure sons. Galton used this evidence to

accept the hypothesis that heredity determines a person’s abilities. The research is

flawed by the simple fact that eminent fathers are in positions to help their sons

also attain eminence. In other words, the effects of environment may be as impor-

tant as heredity in determining a person’s vocational achievement. Although Gal-

ton’s research on heredity is not taken seriously today, it did produce the very

useful tool known as the correlation coefficient.

Correlation coefficients can range from −1.00 to +1.00. A perfect negative

correlation is −1.00, and +1.00 stands for a perfect positive correlation. There

can also be correlation coefficients between these two values. For example, −.85

indicates a high negative correlation, and +.62 indicates a moderate positive

correlation.

24 PSYCHOLOGY

Trend line

y

x

Graph for a positive correlation. The trend

line shows that in general as X increases, Y

also increases.