4
$\begingroup$

We are analysing the effects of different harvesting intensities on soil nutrients. Data was collected over three different years. The first year was before treatments were applied, and the second and third years were 1 year and 6 years post treatment. There are 4 stand types (each replicated 3 times), 6 treatments (retention harvesting of all different intensities) and 6 plots within each stand type*treatment.

I am having some trouble wrapping my head around this question and am wondering which test I should be using?

Improve this question
$\endgroup$
2
  • $\begingroup$ I'm having trouble wrapping my head around your design... At what level was the randomization done? At the standtypetreatment level? How many standtypetreatment level 'units' are there? 4×3×6=72? And then 72×6=432 plots? $\endgroup$
    – onestop
    Commented Jan 13, 2011 at 20:56
  • $\begingroup$ In addition to onestops's remark - I would also be interested to know - what was the aim of the experiment, what was the response variables, and any other specifics that are important in the data analysis. I am guess this to be a split-plot-type design, but we can help you better if we get more details. $\endgroup$
    – suncoolsu
    Commented Jan 13, 2011 at 22:42

1 Answer 1

Reset to default
2
$\begingroup$

Sounds like a mixed effects ANOVA. If you have a continuous treatment variable (i.e., harvesting intensity), then an ANCOVA is warranted (or, really, just a mixed effects/hierarchical general linear model) (or generalized linear model if your response variable better fits that framework).

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.