Chi Square Formula

Chi square is a method used in statistics that calculates the difference between observed and
expected data values. It is used to determine how closely actual data fit expected data. The value of
chi square helps us to answer the question 'is the difference in expected and observed data
statistically significant?' A small chi square value tells us that any differences in actual and expected
data are due to chance, so the data is not statistically significant. A large value tells us the data is
statistically significant and there is something causing the differences in data. From there a
statistician may explore factors that may be responsible for the differences.

Chi square formula

Chi is a Greek symbol that looks like the letter x as you can see in the 'Chi square formula' image.
To calculate chi square, we take the square of the difference between the observed (o) and
expected (e) values and divide it by the expected value. Depending on the number of categories of
data, we may end up with two or more values. Chi square is the sum of those values. It is a common
mistake for students to take the square root, as you would if you were solving for chi. However, the
value that we are looking for is chi square so we do not need to take the square root.
Once we have the value of chi square, we would use a Chi Square table to determine whether the
data is statistically significant.

Calculating Chi Square

A poll was taken of 100 college freshman to determine whether chocolate or vanilla was the
preferred ice cream flavor. It was expected that 25% would prefer chocolate and 75% would prefer
vanilla. The results of the survey are shown below.

Chocolate Vanilla
Expected 25 75
Observed 39 61
Because we have two categories, chocolate and vanilla, we will need to calculate two values and
take their sum. This will give us the value of chi square. The expected value gets substituted for e in
the equation, and the observed value gets substituted for o.
Notice that in the numerator of each fraction, the expected value is subtracted from the observed
value. It is then squared to make sure that the result is a positive value, as chi square must be
positive.