Methods: Correlation
steven j. barnes
• Explain what is meant by the term ‘correlation.’
• Describe what a scatterplot is and why it can be useful.
• Explain the difference between a positive and a negative correlation.
• Explain how an artificial correlation might arise and how to be on the
lookout for artificial correlations.
Learning Goals
You are only interested in correlation when you have scores on
two variables measured on the same subjects.
Correlations
You are only interested in correlation when you have scores on
two variables measured on the same subjects.
Some examples:
-height and weight
-number of neurons and age
Correlations
You are only interested in correlation when you have scores on
two variables measured on the same subjects.
Some examples:
-height and weight
-number of neurons and age
Correlations make it possible to use the value of one variable to
predict the value of another.
Correlations
Scatterplot
Variable A
Variable B
Correlations
Scatterplot
Variable A
Variable B
Correlations
Positive Correlation
Variable A
Variable B
Correlations
Positive Correlation
Variable A
Variable B
Correlations
Negative Correlation
Variable A
Variable B
Correlations
Negative Correlation
Variable A
Variable B
Correlations
No Correlation
Variable A
Variable B
Correlations
Artificial Correlation
Variable A
Variable B
Correlations
Artificial Correlation
Group 1
Variable A
Group 2
Variable B
Correlations
Correlation and Causation
Can we be sure variable A causes variable B?
A B
A B
A
C
B
Correlations
Correlation and Causation
Can we be sure variable A causes variable B?
Low self esteem Depression
Low self esteem Depression
Low self esteem
Biological
Predisposition
Depression
Correlations