|
Chi-Squared Tests - c2 Even though both of the chi-squared activities have similar mathematical operations in the actual execution of the statistical analysis, they are used quite differently. The chi-squared test for goodness of fit is used when there is one set of values that is predicted, and another set of data that is actual from the test. The predicted values can be obtained from probability: for example, a die can be tested to see if it is fair or not - if it is tested sixty times, then it would be predicted that each number would appear ten times. The actual data would be the results from rolling the die sixty times. The chi-squared test for independence does not deal with a predicted column, but uses more than one factor. For example, a relationship between amount of rain during a storm could be tested to see if it has any relation to the ambient temperature. The data from the experiment is then put into a contingency table, which shows the frequency for each amount of rain and temperature occurred. Original work on this document was done by Central Virginia Governor's School students Richard Barnes, Kim Tibbs, and Ryan Nash (Class of '00). This document was updated by Central Virginia Governor's School students Matthew James and Kyle Nenninger Copyright © 1999 Central Virginia Governor's School, Lynchburg, VA |