Computers
Correlation Measurements with Microsoft Excel by Stephen Nelson
[Apr 24, 2008]
Excel provides useful statistical functions for measuring correlation between two variables.As a reminder, the benefit of using a correlation coefficient to measure the relationship between two variables as opposed to using covariance is that the unit of measurement doesn't matter.
But a caution: Remember that correlation does not show causation. That is, you could easily show that as the number of ice cream cones consumed increases during a year, so does the number of drownings. But this does not mean that eating ice cream causes people to drown-more likely, these variables are both independently related to another variable-that of temperatures. Correlation is symmetrical, so you get the same coefficient if you switch the variables. Don't calculate a correlation coefficient if you manipulated one of the variables. Use linear regression instead.
CORREL
You use the CORREL function in Excel to determine whether two data sets are related, and if so, how strongly. The correlation coefficient ranges from +1, indicating a perfect positive linear relationship, to -1, indicating a perfectly negative linear relationship. To calculate a correlation coefficient for a sample, Excel uses the covariance of the samples and the standard deviations of each sample. To use the CORREL function in Excel, just select the two sets of data to use as the arguments and use the following syntax: