Contingency Coefficient

Problem

Return the Pearson contingency coefficient C = sqrt(chi2/(chi2 + n)) for a 2D table of observed counts, where chi2 uses expected counts row_total*col_total/n. Return 0.0 for an empty table, a zero total, or any zero expected count.

Hint

C measures association but has an upper bound below 1 that depends on table size.

Test cases

case 1

in: [[[10,20],[30,40]]]

expected: 0.08873565094161137

case 2

in: [[[5,5],[5,5]]]

expected: 0

case 3

in: [[[10,0],[0,10]]]

expected: 0.7071067811865476

case 4

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

expected: 0.3779644730092272

case 5

in: [[[10,20,30]]]

expected: 0

Starter codePython

def contingency_coefficient(observed):
    # Your code here
    pass

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