The Chi-Square Test is intended to estimate if the differences between observed results of an experiment and the predicted results of the same experiment are significant enough to assume the differences are not due to chance. If the observed and predicted results are not significantly different, the Chi-Square test will then infer the differences are due to chance.

Example: You flip a coin 100 times and get 42 heads and 58 tails.  Those are your observations, while the hypothesis is that you will get 50 heads and 50 tails. Observations (O) = [42, 58].  Predictions (P) = [50, 50]

`chi^2` = `sum ((O - P)^2) / P

`chi^2` = `((42 - 50))^2  / 50` + `((58 - 50)^2)  / 50` = 2.56

Notes

You can apply the chi-square test to a dataset containing columns of predicted and observed data values.  To apply this to another data table, simply make a duplicate of this equation and add the UUID of the data set to the input variable enumerated list labeled "UUID of Table".

You can add additional names for the columns of predicted and observed data to input variable's enumerated lists or just create your data table with the same names: "Observed" and "Predicted".