Meaning of Correlation Analysis
Correlation Analysis measures the strength and direction of the relationship between two variables. It helps businesses and researchers understand how one variable changes in relation to another.
For example:
- Sales and Advertising Spend – Does increased spending on ads lead to higher sales?
- Stock Prices and Interest Rates – Do higher interest rates impact stock prices negatively?
Correlation does not imply causation—it only shows whether two variables move together.
2. Types of Correlation
Based on Direction
✔ Positive Correlation: When one variable increases, the other also increases (e.g., more study hours → higher marks).
✔ Negative Correlation: When one variable increases, the other decreases (e.g., higher prices → lower demand).
✔ Zero Correlation: No relationship between variables (e.g., shoe size and intelligence).
Based on Strength
✔ Perfect Correlation (+1−1) – A direct relationship exists.
✔ High Correlation (0.75 to 1 or −0.75 to −1) – Strong relationship.
✔ Moderate Correlation (0.5 to 0.75 or −0.5 to −0.75) – Moderate relationship.
✔ Low Correlation (0 to 0.5 or −0.5 to 0) – Weak or no relationship.
3. Rank Correlation Method (Spearman’s Rank Correlation Coefficient)
Definition
Spearman’s Rank Correlation Method is used when data is in ordinal (ranked) form rather than numerical values. It measures the strength of association between two ranked variables.
Formula for Rank Correlation
Where:
- R = Spearman’s Rank Correlation Coefficient
- d = Difference between the ranks of corresponding values
- n = Number of observations
Example Calculation
A company compares Employee Performance Scores and Customer Satisfaction Ratings (Ranked Data).
| Employee | Performance Rank (X) | Satisfaction Rank (Y) | d=X−Y | d2 |
|---|---|---|---|---|
| A | 1 | 2 | -1 | 1 |
| B | 3 | 1 | 2 | 4 |
| C | 2 | 3 | -1 | 1 |
| D | 5 | 4 | 1 | 1 |
| E | 4 | 5 | -1 | 1 |
Since R=0.6, there is a moderate positive correlation between performance and satisfaction.
Advantages of Rank Correlation
✔ Useful for qualitative data (e.g., rankings in surveys).
✔ Works even when data is not normally distributed.
✔ Easy to calculate with small datasets.
Limitations of Rank Correlation
❌ Less accurate for large datasets with many tied ranks.
❌ Cannot be used for numerical data with precise measurements.
4. Karl Pearson’s Coefficient of Correlation
Definition
Pearson’s Correlation Coefficient (r) measures the linear relationship between two numerical variables. It shows how strongly two variables are related quantitatively.
Formula for Pearson’s Correlation
Where:
- r = Pearson’s correlation coefficient
- X,Y = Individual values of variables
- n = Number of observations
Example Calculation
A company wants to check the relationship between Advertising Spend (₹) and Sales Revenue (₹).
| Advertising Spend (X) | Sales Revenue (Y) | x2 | y2 | XY |
|---|---|---|---|---|
| 10 | 30 | 100 | 900 | 300 |
| 15 | 40 | 225 | 1600 | 600 |
| 20 | 50 | 400 | 2500 | 1000 |
| 25 | 60 | 625 | 3600 | 1500 |
| 30 | 70 | 900 | 4900 | 2100 |
Step 1: Compute Summations
Step 2: Apply Pearson’s Formular=
Since r=1, this indicates a perfect positive correlation between advertising spend and sales.
Advantages of Pearson’s Correlation
✔ Highly accurate for numerical and continuous data.
✔ Used in econometrics, business analytics, and finance.
✔ Helps in predictive modeling (e.g., sales forecasting).
Limitations of Pearson’s Correlation
❌ Assumes a linear relationship (not suitable for non-linear data).
❌ Sensitive to outliers, which can distort results.
❌ Does not imply causation—a high correlation does not mean one variable causes changes in the other.
5. Comparison of Rank Correlation and Pearson’s Correlation
6. Conclusion
Both Rank Correlation (Spearman’s) and Pearson’s Correlation are essential for analyzing relationships between variables. Spearman’s Rank is used for ordinal data, while Pearson’s Correlation is best for numerical data with a linear relationship.