Search

One Proportion Test Help

This test is performed with binary data when you want to examine the absence or presence of a specific attribute. This type of test is very common in elections when polling people to find out if they support person A or person B.

The test gives two options based on the size of the sample: large or small sample. A different approach is used based on which case is selected. You should select the large sample case when npbar > 10 where n = sample size and pbar is the average proportion of events.

You have the option for a hypothesis test based on a hypothesized proportion. You also have the option for a two-sided, lower one-sided or an upper one-sided test.

The One Sample Proportion Test input form is shown above. The data are entered directly into the input form. For example, consider the following example. A magazine conducted a telephone survey of 800 adults and asked if they had guns in the home. 45% of the respondents said yes. We want to develop a 95% confidence interval for the proportion of homes with guns. We also want a hypothesis test to see if 50% of the homes have guns. In this example, n = 800 and pbar = .45. Thus, n(pbar) =800(.45) = 360. We will use the large sample option.

One Proportion Test Output

The output from the One Sample Proportion Test is shown below. The entries are explained below the output.

The output tells you the conclusion from the test. The null hypothesis (H0) and the alternate hypothesis (H1) are printed below the title.

A plot of the confidence interval is also given which shows the relationship of the confidence interval to the hypothesized proportion.

Scroll to Top