Summary
Highlights
The Pearson correlation analyzes the relationship between two variables, such as a person's salary and age. If a relationship is confirmed, one variable can predict the other using regression. However, a clear causal relationship must exist. The Pearson correlation measures the linear relationship's strength and direction through the coefficient R, which ranges from -1 to 1.
The strength of correlation is read from a table: R between 0 and 0.1 indicates no correlation, while R between 0.7 and 1 indicates a very strong correlation. A positive correlation means large values of one variable correspond with large values of the other (e.g., body size and shoe size). A negative correlation means large values of one variable correspond with small values of the other (e.g., product price and sales volume).
The Pearson correlation coefficient (R) is calculated using a specific equation. It involves subtracting the mean value from each individual data point for both variables, multiplying these differences, and then summing them up. The denominator scales the coefficient between -1 and 1. The result indicates whether the relationship is positive, negative, or non-existent based on the distribution of data points around the means.
When testing a hypothesis about the population, we determine if the sample correlation coefficient is statistically significantly different from zero. The null hypothesis states no linear relationship exists (R does not differ significantly from zero), while the alternative hypothesis states there is a linear relationship (R differs significantly from zero). A t-test is used to check this, considering the correlation coefficient (R) and sample size (N).
A p-value is calculated from the test statistic T. If the p-value is less than the specified significance level (typically 5%), the null hypothesis is rejected, indicating a significant correlation. Otherwise, it is not rejected.
To calculate the Pearson correlation coefficient, only two metric variables are needed. However, to test if the coefficient is significantly different from zero, both variables must also be normally distributed. If this assumption is not met, the calculated test statistic and p-value cannot be reliably interpreted.
The video concludes by mentioning an online tool called 'Data Tab' where users can easily perform a correlation analysis by copying their data and selecting the appropriate options. The tool can automatically calculate the Pearson correlation and provide a summary of the results.