[R-sig-ME] Too small a sample size for lmer?
Christine.Griffiths at bristol.ac.uk
Sat Jul 18 14:58:36 CEST 2009
Dear R users,
Many of you may be familiar with my design as I have posted a number of
queries before. Having consulted with someone in my department about
estimating bias corrected confidence intervals for small sample sizes
(rather than MCMC which Baayen et al. 2008 suggest should not be used),
they implied that I should not be using lmer for such a small sample size
as lmer was designed to deal with very large datasets. Is this still the
case? If so what is regarded as a small sample size?
Below is a description of my data. I have 5/6 enclosures (replicates) per
treatment - Aldabra/Radiata/control. Aldabra and radiata refer to two
different tortoise species, while control lacks tortoises. The enclosures
were assigned to a block: a block containing each of the 3 treatments, i.e.
6 blocks in total. Each month for ten months I collected data: a repeated
crossed design. Unfortunately, I have non-orthogonal, unbalanced data (5/6
enclosures per treatment) as I cannot use a replicate within the aldabra
and radiata treatments. These are however from different blocks so I am
reluctant to axe them to achieve balanced data as this would leave me only
4 blocks. I measured various attributes which I think that tortoises would
have an impact on, e.g. plant count, species richness. Because my data is
unbalanced and a repeated measures design I had chosen lmer to best model
For one other aspect, I calculate food web properties, for which I have no
replication, i.e. only one observation per treatment per month. Would lmer
be an acceptable way to analyse this data?
If lmer is not advised for the analyses of these data, what other analyses
techniques should I investigate?
Baayen et al. (2008)Mixed-effects modeling with crossed random effects
for subjects and items. Journal of Memory and Language, 59, 390-412.
More information about the R-sig-mixed-models