# The Two-Sample Proportion Test Dialog Box

## Results Log Output

If this check box is checked, the result will be output to Results log.

## Input Data Form

Specify the form of the input data, summarized data or raw.

Summarized Perform the test on summarized statistics you enter in the dialog box. Number of Success of Sample1 The number of observed success data in sample 1 Sample Size of Sample1 The total number of observations in sample 1 Number of Success of Sample2 The number of observed success data in sample 2 Sample Size of Sample2 The total number of observations in sample 2 Perform the test on the raw data in the form of "failure" and " success" Input Data Form Specify the data range of sample 1 and sample 2 1st Data Range 2nd Data Range Success To specify which value to be success Perform the test on the input data contains one sample column and one grouping column. The data column is in the form of "failure" and " success" Input Data Form Specify one sample column and one grouping column Group Range Data Range Success To specify which value to be success

## Proportion Test

Use proportion to determine whether or not two proportions are different

Test Difference The hypothetical proportion difference to be tested. The null hypothesis to be tested Alternative Hypothesis Options: Proportion 1 - Proportion2<> Test Difference If chosen, two-tailed proportion test will be performed. That's to say, the procedure will test if the difference of two proportion is equal to the specified value Proportion 1 - Proportion2 > Test Difference If chosen, upper-tailed proportion test will be performed. That's to say, the procedure will test if the difference of two proportion is larger than the specified value Proportion 1 - Proportion2 < Test Difference If chosen, lower-tailed proportion will be performed. That's to say, the procedure will test if the difference of two proportion is smaller than the specified value Specify the significance level of the test. This option determines if confidence intervals for proportion difference should be computed for given confidence levels. Specifies the confidence levels for which the confidence intervals will be computed. This option is only available when Confidence Interval(s) is chosen.

## Test Methods

Normal Approximation Compute the p value and confidence interval in result with normal approximation normal distribution Specify to use a pooled estimate of p for the test when Test Difference is 0 Compute the p value and confidence interval with the Fisher's exact test, which is based on the Binomial Test .

## Power Analysis

Power Analysis check box: If checked, power will be computed for the porportion test.

Actual Power If checked, actual power will be computed Specify the significance level ( $\alpha$) for power computation. This option is only available when either Actual Power or Hypothetical Power is checked. If checked, hypothetical power will be computed. Specifies the hypothetical sample size(s), for which power will be computed. This option is only available when Hypothetical Power is checked.

## Outputs

Customize the output results of this analysis.

Output Results Specify the output worksheet for test results