Jump to content

Statistics/Testing Data/z-testsprop

From Wikibooks, open books for an open world

z test statistic: one proportion

[edit | edit source]

A test statistic that represents the difference between the hypothesized proportion and the observed proportion is used when the binomial parameter p is being tested through the use of a hypothesis-testing procedure. This value is then divided by the standard error. When n is significantly large (n > 20, np > 5, and nq > 5), the test assumptions have been satisfied, and the null hypothesis is true, then we assume that the test statistic is normally distributed.