Scientific experiments often consist of comparing two or more sets of data. This data is described as unpaired or independent when the data set appears from separate or paired individuals when it appears from the same individual at various points in time. For example one clinical trial may involve measurement of blood pressure from one group of patients given medication and blood pressure from other groups not given. This will be unpaired data. Other clinical trials may record blood pressure in the same group of patients before and after giving the drug. In this case the data "in pairs" because of the possibility of blood pressure after drug administration will be related to the patient's blood pressure before the drug is given.
Statistical tests used to compare data sets have been designed for paired or unpaid data and it is important to use the correct test to prevent wrong results.
Exam test for unpaired data:
- Chi-square Pearson test
- Exact Fisher test
Examples of tests suitable for paired data:
- Test McNemar
Video Paired data
Note
Source of the article : Wikipedia
0 Comments