_______________________________________

Correlation
_______________________________________

Greg C Elvers
_______________________________________

_______________________________________
1

_______________________________________
What Is Correlation?
Correlation is a descriptive statistic that _______________________________________
tells you if two variables are related to each
other
E.g. Is your GPA related to how much you _______________________________________
study?
When two variables are correlated, knowing
the value of one variable allows you to _______________________________________
predict the value of the other variable
2

_______________________________________
Perfect Correlation
When two variables are perfectly correlated, _______________________________________
knowing the value of one variable allows
you to exactly predict the value of the other
variable _______________________________________

_______________________________________
3

_______________________________________
Perfect Correlation
For example, if the only 4 _______________________________________
thing that determined your
GPA was the amount of 3
time that you studied, then
_______________________________________
GPA

the two would be perfectly 2

correlated
1
If you know the value of
one variable, you can 0
_______________________________________
exactly determine the value 0 5 10
of the other variable Study Hours per Day
4
 _______________________________________
Perfect Correlation
That is, all the 4 _______________________________________
variability in one
variable is explained 3
by the variability in
_______________________________________

GPA
the other variable 2

0
_______________________________________
0 5 10
Study Hours per Day
5

_______________________________________
Perfect Correlations
Few, if any, psychological variables are _______________________________________
perfectly correlated with each other
Many non-psychological variables do have
a perfect correlation _______________________________________
E.g. Time since the beginning of class and the
time remaining in the class are perfectly
correlated _______________________________________
What are other examples of perfectly
correlated variables? 6

_______________________________________
Less Than Perfect Correlations
Even if two variables are correlated, most of _______________________________________
the time you cannot perfectly predict the
value of one variable given the other
E.g., other variables besides amount of time _______________________________________
spent studying influence your GPA
Some of the variability is people’s GPA is due
to the amount of time spent studying, but not all _______________________________________
the variability is due to it

_______________________________________
Less Than Perfect Correlations
Study 4 _______________________________________
Time IQ GPA
3
3 80 2.0
3 100 2.5 _______________________________________
GPA

2
3 120 3.0
5 80 2.5 1

5 100 3.0 _______________________________________

0
5 120 3.5 0 5 10
Study Hours per Day
8
 _______________________________________
Less Than Perfect Correlations
With a less than perfect correlation, we can _______________________________________
no longer perfectly predict the value of one
variable given the other variable
We cannot explain all the variability in one _______________________________________
variable with the variability in the other
variable
_______________________________________
9

_______________________________________
The Correlation Coefficient
Correlation coefficients tell us how _______________________________________
perfectly two (or more) variables are related
to each other
They can also be used to determine how _______________________________________
much variability in one variable is
explainable by variation in the other
variable. _______________________________________
10

Pearson’s Product Moment _______________________________________

Correlation Coefficient
Pearson’s product moment correlation _______________________________________
coefficient, or Pearson’s r, for short is a
very common measure of how strongly two
variables are related to each other _______________________________________
Pearson’s r must lie in the range of -1 to +1
inclusive
_______________________________________
11

_______________________________________
Interpretation of Pearson’s r
To interpret Pearson’s r, you must consider _______________________________________
two parts of it:
The sign of r
The magnitude, or absolute value of r _______________________________________

_______________________________________
12
 _______________________________________
The Sign of r
When r is greater than 4 _______________________________________
0 (I.e. its sign is
positive) the variables 3

are said to have a

_______________________________________

GPA
direct relation 2

In a direct relation, as
1
the value of one
variable increases, the
0 _______________________________________
value of the other 0 2 4 6
variable also tends to Study Hours per Night
13
increase

_______________________________________
The Sign of r
When r is less than 0 4 _______________________________________
(i.e., its sign is negative)
the variables are said to 3

have an indirect relation

_______________________________________
GPA

2
In an indirect relation, as
the value of one variable 1
increases, the value of the
other variable tends to 0
_______________________________________
decrease 0 2 4 6
Hours of TV per Night
14

_______________________________________
Is the Sign of r + or -?
As the number of cigarettes smoked per day _______________________________________
increases, GPA tends to decrease
As the number of cats in a farm yard
increases, the number of mice tends to _______________________________________
decrease
As the weight of a cat increases, the length
of its whiskers tends to increase _______________________________________
15

_______________________________________
Is the Sign of r + or -?
Create two examples of correlations and _______________________________________
determine if the sign of r is positive or
negative
_______________________________________

_______________________________________
16
 _______________________________________
The Magnitude of r
The magnitude refers to the size of the _______________________________________
correlation coefficient ignoring the sign of r
The magnitude is equivalent to taking the
absolute value of r _______________________________________
The larger the magnitude of r is, the more
perfectly the two variables are related to
each other _______________________________________
The smaller the magnitude of r is, the less
perfectly the two variables are related to 17
each other

_______________________________________
r=1
When r equals 1.0, 1.0 _______________________________________
there is a perfect
0.8
correlation between
the variables 0.6
_______________________________________
Hours

Knowing the value of

0.4
one variable exactly
predicts the value of 0.2
the other variable
0.0
_______________________________________
0 10 20 30 40 50 60
Minutes
18

_______________________________________
r=0
When r equals 0, 10 _______________________________________
either the assumptions
of correlation have 5
been violated or there
is no relation between _______________________________________
0
the two variables -10 -5 0 5 10
The points in a scatter
plot with r = 0 will -5
_______________________________________
tend to form a circular
cluster -10
19

_______________________________________
0<|r|<1
The larger the _______________________________________
magnitude or r is, the
more the scatter plot’s
points will tend to
0<|r|<1 _______________________________________
cluster tightly about a
line

_______________________________________
20
 _______________________________________
Magnitude of r
Cohen (1988) Correlation Negative Positive _______________________________________
recommends the
following values of r
Small -.29 to -.10 .10 to .29
for “small”,
“medium”, and “large” _______________________________________
effects Medium -.49 to -.30 .30 to .49

Large -1.00 to -.50 .50 to 1.00 _______________________________________

21

_______________________________________
Magnitude of r
List a couple of pairs of variables and guess _______________________________________
whether the magnitude of r is closer to 0 or
closer to 1
_______________________________________

_______________________________________
22

_______________________________________
Pearson’s r
Pearson’s r makes several assumptions _______________________________________
about the data
When these assumptions are violated, r
must be interpreted with extreme caution _______________________________________
Assumptions:
Linear relation
Non-truncated range _______________________________________
Sufficiently large sample size
23

_______________________________________
Linear Relation
Pearson’s r, in its simplest form, only works _______________________________________
for variables that are linearly related
That is, the equation that allows us to predict
the value of one variable from the value of the _______________________________________
other is a line:
Y = slope * X + intercept
Always look at the scatter plot to determine if
the two variables are approximately linearly
_______________________________________
related
24
 _______________________________________
Linear Relation
If the variables are not linearly related, _______________________________________
Pearson’s r will indicate a smaller relation
than actually exists
Often, non-linear relations can be _______________________________________
transformed into linear ones by taking the
appropriate mathematical transformation
_______________________________________
25

_______________________________________
Square Root of Y Transformation
60 8 _______________________________________
7
50
6
40
5
_______________________________________
30 4
3
20
2
10
1 _______________________________________
0 0
0 2 4 6 8 0 2 4 6 8
26

_______________________________________
Non-Truncated Range
A truncated range occurs when the range of _______________________________________
one of the variables is very small
When the range is truncated, Pearson’s r
will indicate a smaller relation between the _______________________________________
variables than what actually exists
Once a range truncation occurs, there is
little that you can do; be careful not to _______________________________________
design studies that will lead to a truncated
range 27

_______________________________________
Truncated Range
A linear relation 4.0
_______________________________________
clearly exists in this 3.5

data 3.0
CollegeGPA

Consider only the data

in the square (thereby
2.5
_______________________________________
2.0
truncating the range)
Is the linear relation as 1.5

clear as it was? 1.0

1.0 1.5 2.0 2.5 3.0 3.5 4.0
_______________________________________
No High School GPA

28
 _______________________________________
Sample Size
If the size of the sample is too small, _______________________________________
relations can appear due to chance
These relations disappear when a larger sample
is considered _______________________________________
Too large of a sample can make near 0
correlations statistically significant, even
though they have very little explanatory _______________________________________
power
29

_______________________________________
Sample Size
The magnitude of r does not depend on _______________________________________
sample size
The likelihood of finding a statistically
significant r does depend on sample size _______________________________________
The sample should be large enough to
generalize to the population of interest
_______________________________________
30