Point-Biserial Correlation

Problem

Return the point-biserial correlation between continuous scores and a 0/1 group indicator: (mean_1 - mean_0)/s * sqrt(p*q), where s is the population standard deviation of scores. Return 0.0 when a group is missing, the lengths differ, or all scores are equal.

Hint

It is the Pearson correlation with a binary variable.

Test cases

case 1

in: [[1,2,3,4],[0,0,1,1]]

expected: 0.8944271909999159

case 2

in: [[5,4,3,2],[1,0,1,0]]

expected: 0.4472135954999579

case 3

in: [[1,1,2,2],[1,1,0,0]]

expected: -1

case 4

in: [[1,2],[1,1]]

expected: 0

case 5

in: [[2,2,2],[1,1,0]]

expected: 0

Starter codePython

def point_biserial(scores, groups):
    # Your code here
    pass

5 test cases · run your code against them in the editor