Chapter Notes: Correlation
1. Definition of Correlation
Correlation is a statistical measure that describes the extent to which two variables change together. It indicates the relationship between the two variables but does not imply causation. For example, increased temperature correlates with increased ice-cream sales, but one does not cause the other.
Key Concepts to Understand:
- Correlation vs Causation: Just because two variables are correlated does not mean that one causes the other. For instance, the correlation between ice-cream sales and drowning incidents may be influenced by a third variable: temperature.
- Positive and Negative Correlation: Positive correlation occurs when both variables move in the same direction (e.g., as income increases, consumption increases). Negative correlation occurs when one variable increases while the other decreases (e.g., as price decreases, demand increases).
2. Measuring Correlation
Correlation is measured quantitatively to analyze the strength and direction of the relationship between two variables.
Common Techniques:
- Scatter Diagram: A graphical representation where individual data points for two variables are plotted on a graph. The visual presentation helps assess the type of correlation (positive, negative, none).
- Karl Pearson’s Coefficient of Correlation ( ): This is the most widely used method to calculate correlation, providing a numerical value that ranges from -1 to +1.
- Spearman’s Rank Correlation: Applicable when the data is not continuous or where ranks are more meaningful than actual values. It involves ranking values and then calculating the correlation.
3. Types of Relationships
Correlation can be defined in various terms:
- Positive Correlation: Both variables increase together. For example, higher education leads to higher income.
- Negative Correlation: One variable increases while the other decreases. For example, increased supply of a product may lead to a decrease in its price.
- No Correlation: The two variables do not show any relationship at all.
4. Properties of Correlation Coefficient
- The correlation coefficient ( ) varies from -1 to +1.
- 1 or -1 indicates a perfect linear relationship (positive or negative).
- 0 indicates no linear correlation.
- The sign of the correlation coefficient indicates the direction of the relationship.
- Correlation coefficient is unitless and does not depend on the scale of measurement, making comparative analysis easier.
- A high correlation does not imply that the changes in one variable will lead to predictable changes in another (i.e., correlation does not imply causation).
5. Example Calculations
To calculate the correlation one must:
-
Gather data sets of two related variables,
-
Create a scatter plot to visualize data points,
-
Use the formulas provided for Karl Pearson’s or Spearman’s correlation.
-
Pearson correlation formula:
[ r = \frac{\sum (XY) - n(\bar{X})(\bar{Y})}{\sqrt{(\sum (X^2) - n(\bar{X})^2)(\sum (Y^2) - n(\bar{Y})^2}} ]
-
Spearman’s rank correlation formula:
[ r_s = 1 - \frac{6 \sum D^2}{n(n^2 - 1)} ]
where D is the difference in the ranks of each observation.
6. Limitations of Correlation
Correlation analysis has limitations due to the various factors that influence data series:
- In cases of non-linear relationships, the Pearson correlation can be misleading.
- Correlation does not consider the underlying cause of fluctuations in the relationship.
7. Conclusion
Overall, correlation provides valuable insights into the characteristics of relationships among variables. However, it is critical to apply reasoning and domain knowledge to interpret these relationships correctly. The presentation of data is as significant as the calculations to ensure accurate insights.