So im doing some joint position sense study, using a digital goniometer for one measure and a device we made using a distant sensor (for that we already tested reliability)
So my question it, what data should i use for my sample size?
i can take sd from other studies using digital app goniometer (same as we used)
Or should i take the SD from the reliability testing of the distant sensor- we did those without patients , only testing the sensor itself (SD was really good for that)
Another option is to use SD for the patients we already have with the sensor, but those got a really high SD since people are probably just not good in replicating their results (one of our conclusions)
in any of these cases, what is the correct way to calculate? how my report on the sample size method should be reported? should i include reference for the method i chose?
Thanks!
Hi Tomer,
Please follow,
Article How to Calculate Sample Size for Different Study Designs in ...
Good luck!
I don't think it is necessary to estimate the standard deviation. Most sample size estimates require you to input an effect size (incorporates variability in some manner). If you are trying to relate the 2 measures you would just need to identify the r value that you think is relevant.
G*Power software can be useful for sample size estimation and is free to download. It also has papers to reference.
http://www.gpower.hhu.de/
You could also cite Cohen's work to support your decision about a meaningful effect size, but it probably depends on the nature of your question.
Hope this helps. Let me know if I misunderstood your question.
Gus
May you can use SD of this article.
Article The Effect of Exergaming on Knee Proprioception in Older Men...
Use gpower for sample size calculations . Select the appropriate distribution such as t, z ,chi square or correlation . As yours is reliability study correlation is the right choice.Between two measurements times you give your assumed r value such as 0.6,0.7 etc.
If you are comparing means then select t or z test and keep medium effect size between o.3 to 0.5 ,power 0.80
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size