[R-sig-ME] lmer function 40 lakes 3 lake zone

ONKELINX, Thierry Thierry.ONKELINX at inbo.be
Mon Aug 26 11:51:58 CEST 2013

Dear Anna,

Use the original counts and add the log(Area) as an offset term. You might want to add the Zone to the fixed effects and maybe its interaction with sediment as well.

glmer(OrignalCounts  ~ offset(log(Area)) + Sediment * Zone + (1|Lake) + (1|Lake:Zone), family = poisson, data = ch2)

Best regards,

ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek / Research Institute for Nature and Forest
team Biometrie & Kwaliteitszorg / team Biometrics & Quality Assurance
Kliniekstraat 25
1070 Anderlecht
+ 32 2 525 02 51
+ 32 54 43 61 85
Thierry.Onkelinx op inbo.be

To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of.
~ Sir Ronald Aylmer Fisher

The plural of anecdote is not data.
~ Roger Brinner

The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data.
~ John Tukey

-----Oorspronkelijk bericht-----
Van: r-sig-mixed-models-bounces op r-project.org [mailto:r-sig-mixed-models-bounces op r-project.org] Namens Belyaeva Anna
Verzonden: maandag 26 augustus 2013 3:23
Aan: r-sig-mixed-models op r-project.org
Onderwerp: [R-sig-ME] lmer function 40 lakes 3 lake zone

Hello all,
I study benthos from 40 lakes. Each lake was divided into 3 different lake zones based on light penetration formula: Littoral, Sublittoral, Profundal. 5 random sediment samples were taken from the littoral, 5 random samples from sublittoral, 25 random samples profundal. Sediment type and organisms' counts and richness(count of species) was recorded for each sample.
To complicate more, 2 different types of equipment were used. Littoral and Sublittoral zones were sampled with large-area heavy sampler, but the deepest part of the lake was sampled with small area sampler. I do not know if different type of equipment should figured in the model since all organisms counts were calculated as per sq/m.
 I want to see if there is the sediment effect on total organism counts (or richness) if lake and zone are taken into account. I would consider the Lake to be the random effect. Zone seems to be meaningful; communities within lake zones are very different from each other, but comparable between lakes. Zone in one lake is comparable to lake Zone in different lake.  Zone might be random or fixed effect.  Zone by itself the very strongly explains variation. There is one sediment type for each sample. There is different number of samples per zone within the lake.
 I think that it is RCBD design. Would you agree?
 Total is the count of invertebrates - Poisson distribution. Sediment has 6 categories. Total I have ~1500 samples.
lmer(Total ~ Sediment+(1|Lake)+(1|Lake:Zone),family=poisson(link = "log"),                na.action=na.omit,data=ch2)
        [[alternative HTML version deleted]]

* * * * * * * * * * * * * D I S C L A I M E R * * * * * * * * * * * * *
Dit bericht en eventuele bijlagen geven enkel de visie van de schrijver weer en binden het INBO onder geen enkel beding, zolang dit bericht niet bevestigd is door een geldig ondertekend document.
The views expressed in this message and any annex are purely those of the writer and may not be regarded as stating an official position of INBO, as long as the message is not confirmed by a duly signed document.

More information about the R-sig-mixed-models mailing list