204 PART 5 Looking for Relationships with Correlation and Regression

correlation coefficient r, based on N observations, by the formula

t

r

r

N

/

/

1

2

2

. Because t fluctuates in accordance with the Student t dis-

tribution with N

2 degrees of freedom (df), it is useful for statistical testing (see

Chapter 11 for more about t).

For example, if r

0 500

.

for a sample of 12 participants, then t

0 5. /

1

0 5

12

2

2

.

/

, which works out to t

1 8257

.

, with 10 degrees of freedom. You

can use the online calculator at https://statpages.info/pdfs.html and calcu-

late the p by entering the t and df values. You can also do this in R by using

the code:

2 * pt(q = 1.8257, df = 10, lower.tail = FALSE).

Either way, you get p

0 098

.

, which is greater than 0.05. At α = 0.05, the r value

of 0.500 is not statistically significantly different from zero (see Chapter 12 for

more about α).

How precise is an r value?

You can calculate confidence limits around an observed r value using a somewhat

roundabout process. The quantity z, calculated by the Fisher z transformation

z

r

r

1

2

1

1

log

/

, is approximately normally distributed with a standard

deviation of 1

3

/ N

. Therefore, using the formulas for normal-based confidence

intervals (see Chapter 10), you can calculate the lower and upper 95 percent con-

fidence limits around z z

z

N

:

.

Lower

/

1 96

3 and Z

z

N

Upper

/

1 96

3

.

. You

can turn these into the corresponding confidence limits around r by the reverse of

the z transformation: r

e

e

x

z

2

2

1

1

/

for z

zLower and z

zUpper.

Here are the steps for calculating 95 percent confidence limits around an observed

r value of 0.05 for a sample of 12 participants (N = 12):

1.

Calculate the Fisher z transformation of the observed r value:

z

1

2

1

0 5

1

0 5

0 549

log

.

.

.

/

2.

Calculate the lower and upper 95 percent confidence limits for z:

z

z

Lower

Upper

/

/

0 549

1 96

12

3

0 104

0 549

1 96

12

3

1 203

.

.

.

.

.

.