I have a group of drivers that drove over a selected route, with their own car. Then i have requested the same group of drivers to drive one car that they never had drove, in the same route. My sample is of 26 individuals.

I want to compare the velocities of both. Can i use a paired samples T-Test, or there are other hypotheses more appropriated?


1 Answer 1


Yes, you can use a paired t test. When you do a paired t test, you are using something called blocking. In your case, you are blocking by subject.

To run you test, you need to:

  1. Find the speeds (are you sure it is speed? maybe time is a better measure here?) for each driver

  2. Calculate speed_own - speed_never, where speed_own is the speed they drove with their own car and speed_never is the speed they drove with the other car.

  3. Since your data set is kind of small, with only 26 observations, check for normality in your data. If the histogram is symmetric, proceed to step 4.

  4. Run a one-sample t test on speed_own - speed_never data.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.