In: Statistics and Probability
what is meant by independent and dependent variable as it pertains to correlation? Provide an example how do we quantitatively measure the strength of correlation between the two variables?
There are several ways to measure correlation between two variables. If two variables are independent of each other ,then they should have no correlation between each other. Graphically, the variables, when plotted should look like--
However, in case of dependent variables the correlation will be significant in comparison a pair of independent variables.
If we consider a pair of variables then the correlation between them is measured by correlation coeffcient, defined as--
, where sx and sy are the standard deviations of the two variables x & y.
If rx,y takes a value very close to zero, we would imply no correlation and conclude the variables to be independent of each other. On the other hand, when rx,y then we would suspect some linear correlation between the variables in question and further the value is from zero higher will be the correlation between them.