I am analyzing a split-plot land-use experiment and I am stuck on the R code.

We used four sites and on each site the following design: We conducted a full-factorial design with two factors (land-use intensity and land-use method) with two levels, resulting in four plots per site. Then we divided each plot in three and performed a soil treatment with three levels (resulting in 12 subplots per site).

This is my first split-plot analysis and maybe I am just confused, because I read through too many examples and varying R codes. However, I can´t find a reference or code suitable for my experiment, meaning a split-plot analysis with a two-way anova on top. For example, I am having difficulties to assign an order in the error term for the land-use factors.

This is my first trial so far (heavily based on examples by Dormann 2017) and I am not happy with it:

lme(x~intensity*method*soil, random=~1|site/soil)

I am grateful for any suggestions and corrections.


Your Answer

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

Browse other questions tagged or ask your own question.