Statistics NonParametricTests Essay Example | Topics and Well Written Essays - 750 words

Statistics NonParametricTests - Essay Example A chi square test to determine whether there is a difference in three or more samples proportions has a number of conditions, the main difference for the Chi square when comparing two independent sample proportion and more than three independent samples proportions is the value requirement in each cell. For the two samples test cells must have a value equal or greater than five while for the three independent samples the value should be greater than one. One of the conditions is that the observations must be independent. The other condition is that the data should be numeric and not percentages or proportions. Another condition is that the cells should have an expected value greater than one. Another condition is that the data should be categorical. A chi square test can also be undertaken to determine whether two samples are drawn from the same population, there are a number of conditions for this tests and one of them is that the samples should be random, the sample sizes should be relatively large because small sample sizes will yield inaccurate conclusions. The other condition is that all cells should have a value greater than zero and that all cells should have an expected value greater than 5, the other condition is that the observations should be independent. The McNemar test is a non parametric test that is used to compare two sample proportions when the two samples are correlated or related. When comparing two sample proportions the Chi square test can be used when the samples are independent, however if the samples are dependent the McNemar test is used. One of the conditions that need to met when using the McNemar test is that the test is comparing proportions, the two samples are dependent or related, when comparing nominal or categorical data and if the samples are matched pair, matched pair means that data contains before and after