What if the Paired T-Test results are significant, but Independent T Test results are not significant? (Done in the same sample)
Ex: Paired T-Tests done in the Experimental groups are significant. However, when the Independent T-Tests are done with both Experimental and Controlled groups, the results are not statistically significant?
Are such situations common? Provided that the sample size is taken so as to avoid Type 2 errors?
Should the findings be reported as they are?
It can happens as Paired-samples t tests compare scores on two different variables but for the same group of cases; independent-samples t tests compare scores on the same variable but for two different groups of cases.
This is common. Paired t test is like you compare yourself to yourself 10 years younger, and independent t test is like you compare yourself to someone else 10 years younger. The first one is more powerful because there are two data for the subject, you. In the second case, there are only one data point for each subject. The question is what is your experiment/study looks like? Do you have two measurements from the same subject? If yes, go for the paired t test, if not go for the independent t test
Thank you for your detailed answer.
With mine, I have a controlled group to compare with too, with which I'll have to select independent T-test. But then it would become statistically not significant. Hence, the confusion.
- Badges
- Science topic
- Similar topics
- Biological Science
- Microbiology
- Virology
- Virus