Statistically Analyzing Data

[Ocalaortho]

My project is comparing Knee Extension, after total knee replacement, measuring with a goniometer and also using the computer to compare knee extension.

The experiment was performed on (30) thirty different people that had total knee arthroplasty. All thirty knees were done under a computer assisted guidence system, and at the end, three measurements of knee extension were made using the computer; and also three measurements, using the goniometer.

In looking at the data, the goniometer has a wide range of numbers, often times measuring between 0-5 degrees of knee extension, at the completion of the case. This is compared with the computer, which almost consistently has numbers such as 1.5, 1.0, with the final results in a very narrow range.

My question is: How do I structure the data to compare (30) thirty different knees with measurements showing extension with the goinometer, and extension with the computer, and how do I compare the statistical significance?

If you need more information, please feel free to e-mail me at (e-mail deleted by moderator)

Thank you for your help.

[hhemken]

Hi Ocalaortho,

You should do a Student's t-Test on the results, testing one method against the other for each individual. You can do the same over the entire patient population as well. This will give you a good idea of whether you can claim that the methods are equivalent or not.
See:
http://udel.edu/~mcdonald/statttest.html
http://studentsttest.com/
http://www.usablestats.com/calcs/2samplet

Nevertheless, it could be tricky. Is one more precise than the other? More accurate? Precision means that the actual measurement is reproducible with very little variation between measurements. Accuracy means that on average it gives you a more correct measurement, even if it doesn't do it with great precision.
See http://en.wikipedia.org/wiki/Accuracy_and_precision

Good luck!