I want to perform One-way ANOVA in R. I do not have the raw data, but only the mean, standard error and size of each group (there are three groups). If I understand correctly, the ind.oneway.second command in the rpsychi package is relevant only when the groups are the same size. Is there a way to perform One-way ANOVA in R when the groups are not the same size?

  • $\begingroup$ ind.oneway.second(m, sd, n, unbiased = TRUE, contr = NULL, sig.level = 0.05, digits = 3) Why not specify n? $\endgroup$
    – Huub Hoofs
    Aug 29 '16 at 12:31
  • $\begingroup$ So you have 3 observations? I.e. the means in each group, and their deviations? Or did I miss interpret? $\endgroup$
    – Repmat
    Aug 29 '16 at 12:40
  • $\begingroup$ n can not get vector with different values. When I try to do it, I get the next error: Error in delta.upper + dfb : non-numeric argument to binary operator. Again, for each group I have only the following data: mean, standard error and size. I do not have the raw data (observations). $\endgroup$
    – Pini
    Aug 29 '16 at 18:54

Can use for group of different size.

(mu.ind.anova <- ind.oneway.second(mu,mu.sd,mu.n))

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.