Skip to content

nested two level mixed model with lmer

2 messages · Cristina Muschitiello, Thierry Onkelinx

Original
Cristina Muschitiello
# 
I'd need to have a suggestion for my analysis:

I have two soil treatments: CT and NT.

For each treatment, I have 11 CO2 measurements taken in 7 times, i.e. 11
"replicates" for 7 dates.
Replicates are nested in the treatments.

I'd like to evaluate if the measurements are different in time for the two
treatments.

My information are:


   - replicates(11)=random factor;
   - days=repeated measure factor (fixed factor)
   - treatment = between subject factor (fixed factor)
   - CO2 = Co2 emission measures (=dependent variable)


Is it correct to consider this as a nested two-level repeated measures
ANOVA?

Is it correct to use the following R sintax?

m1 <- lmer(CO2~days*treatment+(days|replicates),mydata)

Is this a random intercept and slope model with replicates nested in
treatment?

Is it correct to say that this model accounts for:

the main effect of treatment and time and
the interaction between the two?
What would be the difference of m1 from m2?

m2 <- lmer(CO2~days*treatment+(1|replicates),mydata)

Thank you

Cristina
Thierry Onkelinx
# 
Dear Christina,

m1 assumes a linear trend along days with different slopes among treatments
and replicates. Whereas m2 assumes that the linear trend along days only
depends on treatment. So all replicates within a treatment have the same
slope.

PS Don't post in HTML as it can make the text unreadable.

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
Belgium

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

2016-06-14 11:14 GMT+02:00 Cristina Muschitiello <muschitiello at gmail.com>: