Take the 2-minute tour ×
Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It's 100% free, no registration required.

I am trying to extract variance components for selection and chance in a bifactorial design with Generation as a fixed factor and Replicate as a random term, for early fecundity.

Since I am using the individual level (because I want to make confidence intervals for Chance), the sample size is not equal between Generations so it's an unbalanced design and it's is not a repeated measures ANOVA because my individuals are different.

The problem is that when I compare the results between Statistica and R the Mean Squares(MS) for the Generation and Replicate Factors are different (but the MS for Gen*Rep and the residuals are not!)

My data consist of 2 terms:

  • Generation - 2 level fixed factor (6 and 11)
  • Replicate - 3 level random factor (Ad1, Ad2, Ad3)

The model I am using is

g <- Anova (aov(Fec~ Gen+Rep+ Gen*Rep), random=~1|Gen*Rep)

I am not sure if this is how to phrase that I want the terms with Replicate to be random (maybe that's one of my errors when comparing with the Statistica results?).

  • Might the differences be due to the unbalanced design? Do R and Statistica have different ways of handling the different number of individuals in each replicate and generation?

  • Why aren't all MS equal? Is there a way to make the results be the same?

  • How can I indicate that a factor (and all its interactions) are random?

  • How can I tell R that I want it to test the Generation factor against the Gen*Rep term or to put it in another way to make it test the significance of the Gen factor against the heterogeneity between my replicates?

share|improve this question
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.