This seems a simple question, but I am way too dusty on statistics, so any help will be really appreciated.

Ok, I have one blood sample.

Then, I do a measurement (Aspect Ratio, vertical axis) at different stress levels (horiz. axis). However, I do the measurement under two different viscosity (red and brown in the image). Now, I need to know if the results in Red is significantly different than results in Brown.

Stress and viscosities are independent variables, while aspect ratio is the dependent variable.

Note that every single point in the plot is a single measurement. Thanks.

enter image description here

  • $\begingroup$ Perhaps an ANCOVA model $\endgroup$ – Robert Long Feb 19 '19 at 20:44

What you essentially want is to conduct a 2D t-test. You may refer to this paper (https://www.jstor.org/stable/pdf/2532625.pdf) for more details.

If you just want to do some simple tests, I would suggest you conduct two separate t-tests, one for stress, the other for aspect ratio.

| cite | improve this answer | |
  • $\begingroup$ Why do I have to do a separate test for stress. Stress and viscosities are both independent variables and aspect ratio is dependent. $\endgroup$ – Muhamad Mohaqeq Feb 19 '19 at 20:32
  • $\begingroup$ You don't need to do it for stress. My suggestion is for general cases. $\endgroup$ – X. Zhao Feb 19 '19 at 20:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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