SPSS statistical analysis-non-parametric test

The hypothesis test of statistics can be divided into parameter test and nonparametric test, the parameter test is calculated according to some assumptions, when these assumptions can not be satisfied, the efficiency of the parameter test will be greatly discounted, even the wrong result, and the non-parameter test is usually without the hypothesis condition, so the application scope is wider than the parameter test.

Non-parametric testing in the case of no assumptions, the maximum use of sample information, statistical, mathematical methods and techniques to construct statistics and test, in some cases, non-parametric testing than parameter testing has a higher performance, however, we can not blindly use non-parametric test, after all, the parameter test more rigorous , the non-parametric test is used only when the data does not conform to the parameter test, so it is important to have a prior observation of the data.

Nonparametric test method is very many, but the majority of nonparametric test methods are based on rank and knot to construct statistics, SPSS non-parameter test is a separate process, also retained the old dialog box, the New dialog box according to the sample situation classification, according to the sample situation to choose methods, and more inclined to automated analysis, The classification of the old dialog box is not very clear, we will follow the new dialog box to introduce

Analysis-non-parametric inspection-single sample

I. Single sample 1. The two-item test, two-item test, also known as two-item distribution test, is used to test whether the sample is from a two-item distribution, that is, whether the frequency of the observed sample is consistent with the expected frequency of a given two-item distribution. Not only for the two categorical variables, for continuous variables can also be treated as two categorical variables, such as the passing of the grade, whether the product is qualified or not.

In this case, we want to test whether the passing rate of all the disciplines is above 95%.

SPSS statistical analysis-non-parametric test