Two observers measured the same thing. Should I compare them with a paired t test?
The question. Another student and I want to know whether our measurements of joint swelling in rats differ. Today I measured 10 rats once, and the other student measured the same 10 rats once. To compare the two of us, should I use a paired t test?
Studies of whether different observers agree when they observe or measure the same thing, and by how much they differ, have a special name: inter-observer variability. For the statistics, please use a Bland-Altman plot. The method and explanation are here (link as of 2023):
https://www.medcalc.org/manual/bland-altman-plot.php
If the same person measures today and again tomorrow, and you want to know whether the results differ, that is intra-observer variability. I happen to have done such a study; you can refer to its statistical methods.
- Tsai IC, et al. Left ventricular myocardial volumes measured during arterial and delayed phases of multidetector row computed tomography: a study on intra- and interobserver variability. https://pubmed.ncbi.nlm.nih.gov/19132545/
這些回答是經驗與建議，不是規定。你的稿件以目標期刊的規定為準，你的研究以所屬機構的倫理審查為準。