In: Statistics and Probability
In Chi-square goodness of fit and contingency table tests, it is sometimes necessary to reduce the number of classifications used in order to do what? Explain why.
Statistics
In Chi-square goodness of fit and contingency table tests, it is sometimes necessary to reduce the number of classifications used in order to draw valid conclusions from the test.
The Chi-square test is highly sensitive to sample size. When the expected frequency in a cell in the table is small, say less than 5, then if we perform the chi-square test it will lead to erroneous conclusions. Therefore sometimes we reduce the number of classifications and combine groups so as to be able to perform the chi-square test and draw valid conclusions from the same.
Let me know in the comments if anything is not clear. I will reply ASAP! Please do upvote if satisfied!