Sample size R-code

I need to conduct a sample size analysis for an experiment with the following characteristics:

  1. One intra-subject independent variable is the type of cheese with two levels, Cheese A (1) and Cheese B (2).
  2. The other intra-subject independent variable is the tasting round with three levels: blind tasting round (1), expectation tasting round (2), and informed tasting round (3).
  3. The dependent variables are five quasi-symmetric variables from 1 to 7 depending on how much the consumer liked the two cheeses over the three tasting rounds. Thus, the dependent variables are: Flavour, Smell, Texture, Aftertaste and Overall impression scoring each dependent variable according to a quasi-symmetric scale from 1 to 7.
  4. All subjects of the experiment will taste cheese A and cheese B in the three planned tasting rounds evaluating in each of them the five attributes: Flavour, Smell, Texture, Aftertaste and Overall Impression scoring each dependent variable according to a quasi-symmetric scale from 1 to 7.
  5. We expect a small medium size , the significance level is fixed at 5%, the power is set at 0.8

I need you to calculate the sample size needed to carry out such a study.
Could you help me know if I am using the correct test?

I'm using the next code in R:

wp.rmanova(ng = 2, nm = 3, f = 0.25, nscor = 1, alpha =
0.05, power =0.8, type = 2)

Hi Rossana,

I assume (hope?) you mistyped and meant to write "I need to", but in any case, this sounds like a statistical question rather than a question about code, so it may be better posted on Stack Exchange's statistics site, Cross Validated:

Folks here may be able to answer your question, too, but questions here are generally less about deciding on a course of action ("what test should I use?") and more about implementing a course of action ("how do I use this test?"), so you may get more feedback there.

This topic was automatically closed 42 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.