Biology, images, analysis, design...
Use/Abuse Principles How To Related
"It has long been an axiom of mine that the little things are infinitely the most important" (Sherlock Holmes)

Search this site




Phi coefficient

The phi coefficient, a measure of association between two binary variables (x & y), is sometimes described as a special case of Pearson's correlation coefficient. Two binary are positively associated if most of the data fall along the leading diagonal (\) cells (a&d), and negatively correlated when the converse is true (/) cells b & c, below.

In the following table, a, b, c, d, denote the cell frequencies, r1 r2 are their row totals, and c1 c2 are their column totals:
x=0ab r1 

Given which the phi coefficient is (ad-bc)/√(r1r2c1c2)

Note that the phi coefficient, unlike Pearson's correlation coefficient, cannot be assumed to lie in the range -1 to +1.