L11: Correlation

Question 1

What is the definition of correlation?

Answer 1

Correlation is the quantification of the degree to which two random variables (continuous/ordinal) are related, provided that the relationship is linear.

Question 2

What do we need to do before performing correlation analysis?

Answer 2

Construct a scatterplot

Question 3

What is correlation coefficient?

Answer 3

• The correlation between two variables can be quantified using the correlation coefficient.
• Correlation coefficient is a quantitative measure of the strength and direction of the linear relationship between the two variables

Question 4

If both variables are continuous, normally distributed, what correlation analysis do we use?

Answer 4

Pearson product-moment correlation coefficient (r)

Question 5

If either variable is continuous, non-normally distributed or ordinal, what correlation analysis do we use?

Answer 5

Spearman rank correlation coefficient (rs)

Question 6

Do r and rs have units?

Answer 6

No units

Question 7

What is the range of r and rs?

Answer 7

-1 to 1

Question 8

What does the sign of r and rs indicate?

Answer 8

Direction of the linear relationship
Positive –> positive correlation (variables increase together)
Negative –> negative correlation (as one increases, the other decreases)

Question 9

What does the magnitude of r and rs indicate?

Answer 9

Strength of the linear relationship
If less than 0.5 –> weak linear relationship
0.5-0.7 –> strong linear relationship
0.7-1.0 –> very strong linear relationship
1 –> perfect positive correlation
-1 –> perfect negative correlation
0 –> no correlation (no linear relationship)

Question 10

Does correlation imply causation?

Answer 10

NO!

Question 11

What is the conclusion for correlation analysis?

Answer 11

There is/is no statistically significant negative/positive correlation (r = , p = ) between the two variables