 | | Paired data | | | Question | | | How do I work with paired data? | | Answer | A common experiment design is paired data, where each subject has received two different treatments. Thus there are two samples from each subject. For paired data one should not do an ordinary Two Group Comparison of the two treatments, because differences between the subjects may obscure the difference between the two treatments. Instead one should do a paired test. There are several tests to select from iin the statistics dialog.
Have a look at the following document How to handle different types of experiment designs J.pdf for more details.
You can use both the integrated test such as t-test or the Extended statistics. | | Related articles | | | Back to Search Results |
|