How to Chi Square Test

A Chi Square Test is often used to measure a goodness of fit between an observed and expected distribution of values. Knowing how to perform a Chi Square Test can be useful for testing probable to expected outcomes, fitting points to a curve, or testing a statistical hypothesis. This article will explain how the Chi Square Test is performed.

Things You'll Need

• Basic statistics knowledge
• Spreadsheet program like MS Excel or OpenOffice Calc

Instructions

1. Hypothesize the expected outcome and take actual observed data. Here you need to determine what is being tested. Initially you will have a hypothesis or hunch of the data that you will be measuring. Whether is a flat average value or a point distribution curve an idea of what is being measured needs to be understood.

2. 

   Calculate the Chi Square test statistic. Chi Square test statistic is a measurement of observed to expected outcome whose sampling distribution is very close to the Chi Square distribution. This Chi Square test statistic will be compared against a critical value to measure how good of a fit the observed is to the expected outcome. In the equation k is the number of observed values. "o" is the i-th observed value and "e" is the i-th expected value.

3. Compare the test statistic against a Chi Square critical value. Here we want to test against a Chi Square confidence level. Typically a 95% confidence is used to compare the goodness of fit and k-1 degrees of freedom where k is the number of observed values. For a 95% confidence means that only 5% error is allowed. The easiest way to measure the Chi Square critical value is to use the CHIDIST function in most spreadsheet programs. The result of the CHIDIST(test statistic, k-1) is a probability. If the probability is less than 5% than we will reject our initial hypothesis that the observed outcome meets our expected outcome.