January 2017
Beginner
882 pages
203h 41m
English
The chi-squared test of the null hypothesis of independence compares observed frequencies in a contingency table to those expected if the underlying random variables are independent. As a special case, we can use this test to compare the equality of several proportions, extending the two-sample comparison in Chapter 18. When used to search for patterns in data mining, we need to assign p-values to the test statistic to adjust for the effects of sample sizes and the size of the contingency table.