1. define null hypothesis (no correlation) and alternative hypothesis (correlation)
2. complete contingency table (present x present, present x absent, absent x present, absent x absent, totals)
3. complete table of expected values: \[row total x column total\] / grand total
4. calculate chi squared value: \[(observed - expected)^2\] / expected → for each box. add up chi squared for each box to get final chi squared value.
5. degrees of freedom: (no. of rows - 1) x (no. of columns - 1) = this determines after how many values it becomes a sure pattern
6. match the ideal degree of freedom with 95% accuracy on a critical values table. if the previous chi squared value is bigger than the listed table value, you can be 95% sure there is a correlation.