Z-test for Proportions-Dependent Groups
This calculator is used to compare the proportions from two overlapping groups (i.e. groups that contain the same people) to determine if they are significantly different from one another. It allows you to enter the information for as many as 10 groups at a time. The calculator will then perform the Z-test for every possible combination of groups.
This test assumes you have a large population.
How To Interpret The Results
For example, suppose you carried out a paired preference test amongst 200 respondents and you are interested in whether there is a difference between the observed results for likelihood to purchase between the trial product (Group 1) and standard product (Group 2), at the 95% confidence level.
|Sample Size||Group 1||Group 2|
|Group 1||Group 2||Group 1|
|Observed Sample Proportion||Group 2|
‘NO’ indicates that there is not a significant difference between the two observed proportions, i.e. there is no significant difference in the proportion of respondents who are likely to purchase the trial or standard products .
Use the calculator above to calculate your results.