Imagine you're tracking two things at once — say, hours spent studying and exam scores. You notice a pattern: when study hours go up, scores tend to go up too. That's correlation in its simplest form: a measure of how two variables move together.
But here's the key intuition: correlation is not about causation. Just because two things move together doesn't mean one causes the other. Ice cream sales and drowning incidents both rise in summer — they're correlated, but ice cream doesn't cause drowning. Both are driven by a third factor (heat).
The Core Idea
Correlation quantifies the strength and direction of a linear relationship between two variables. It answers three questions:
Direction: Do they move in the same direction (positive) or opposite directions (negative)?
Strength: How tightly do they follow that pattern — is it a perfect line, or a loose cloud of points?
Form: Is the relationship linear (a straight line) or something else? Correlation only measures linear relationships.
Watch out
Correlation is blind to non-linear relationships. Two variables could have a perfect U-shaped relationship and still show zero correlation.
The Precise Statement: Pearson's Correlation Coefficient
The most common measure is Pearson's correlation coefficient, denoted by r. It's a single number between −1 and +1.
This is the same formula written more compactly: covariance divided by the product of standard deviations.
What the Numbers Mean
Value of r
Interpretation
+1
Perfect positive linear relationship (all points on an upward-sloping line)
+0.7 to +0.99
Strong positive correlation
+0.3 to +0.69
Moderate positive correlation
0 to +0.29
Weak positive correlation
0
No linear relationship
−0.29 to 0
Weak negative correlation
−0.69 to −0.3
Moderate negative correlation
−0.99 to −0.7
Strong negative correlation
−1
Perfect negative linear relationship (all points on a downward-sloping line)
Tip
| The sign tells direction, the absolute value tells strength. An r of −0.9 is just as strong as +0.9 — only the direction differs.
The Intuition Behind the Formula
The numerator ∑(xi−xˉ)(yi−yˉ) is the covariance. For each point, it asks: is this point above or below the mean in both variables?
If a point is above the mean in xand above the mean in y, the product (xi−xˉ)(yi−yˉ) is positive.
If a point is above in x but below in y, the product is negative.
Summing all these products tells you whether the overall pattern is positive or negative.
The denominator ∑(xi−xˉ)2∑(yi−yˉ)2 is just a scaling factor — it ensures r always falls between −1 and +1, regardless of the units of measurement. …
Ranking the 10 individuals' IQ and EQ scores and applying Spearman's formula gives ρ≈−0.115, indicating almost no linear relationship (a very weak negative tendency) between IQ rank and EQ rank in this sample.
ρ=1−n(n2−1)6∑di2, where di=R(xi)−R(yi) is the difference between the ranks of the ith pair, and n = number of pairs.
Rank the IQ values in descending order (rank 1 = highest IQ); all 10 values are distinct, so there are no ties:
Student
1
2
3
4
5
6
7
8
9
10
IQ
99
120
98
102
123
105
85
110
117
90
EQ
3
0
30
45
16
25
17
24
26
5
R(IQ)
7
2
8
6
1
5
10
4
3
9
R(EQ)
9
10
2
1
7
4
6
5
3
8
d=R(IQ)−R(EQ)
-2
-8
6
5
-6
1
4
-1
0
1
d2
4
64
36
25
36
1
16
1
0
1
Rank EQ the same way (rank 1 = highest EQ); all 10 values distinct, no ties.
Check: both rank columns sum to 2n(n+1)=210(11)=55✓ (7+2+8+6+1+5+10+4+3+9=55; 9+10+2+1+7+4+6+5+3+8=55). …