Psychology2016

The Science of Psychology 27

Notice that the closer the number is to zero, the weaker the relationship becomes.
Researchers would probably find that the correlation coefficient for the relationship
between people’s weight and the number of freckles they have is pretty close to zero,
for example.

Go back to the cigarette thing—if we found that the correlation

between cigarette smoking and life expectancy was high, does that

mean that smoking causes your life expectancy to be shortened?

Not exactly. The biggest error that people make concerning correlation is to assume
that it means one variable is the cause of the other. Remember that correlation does not
prove causation. Although adverse health effects from cigarette smoking account for

*inverse: opposite in order.

Figure 1.4 Five Scatterplots
These scatterplots show direction and strength of correlation. It should be noted that perfect correlations,
whether positive or negative, rarely occur in the real world.

Y

X

Perfect positive

correlation

Y

X

Perfect negative

correlation

Y

X

Modest negative

correlation

Y

X

Modest positive

correlation

Y

X

No correlation

Inter

act

ive

Interactive

goes up; as one decreases, the other also decreases. If negative, the two variables have an
inverse* relationship—as one increases, the other decreases. If researchers find that the
more cigarettes a person smoked, the younger that person was when he or she died, it
would mean that the correlation between the two variables is negative. (As smoking goes
up, life expectancy goes down—an inverse relationship.)
The strength of the relationship between the variables will be determined by the
actual number itself. That number will always range between +1.00 and −1.00.
The reason that it cannot be greater than +1.00 or less than −1.00 has to do with the
formula and an imaginary line on a graph around which the data points gather, a graph
called a scatterplot (see Figure 1.4). If the relationship is a strong one, the number will be
closer to +1.00 or to −1.00. A correlation of +.89 for example, would be a very strong pos-
itive correlation. That might represent the relationship between scores on the SAT and
an IQ test, for example. A correlation of −.89 would be equally strong but negative. That
would be more like the correlation researchers would probably find between smoking
cigarettes and the age at which a person dies.