R help - Jul 2008 - Mixed effects model where nested factor is not the repeated across treatments lme???

Hi,

 

I have searched the archives and can't quite confirm the answer to this.
I appreciate your time...

 

I have 4 treatments (fixed) and I would like to know if there is a
significant difference in metal volume (metal) between the treatments.
The experiment has 5 blocks (random) in each treatment and no block is
repeated across treatments. Within each plot there are varying numbers
of replicates (random) (some plots have 4 individuals in them some have
14 and a range in between). NOTE the plots in one treatment are not
replicated in the others.

 

So I end up with a data.frame with 4 treatments repeated down one column
(treatment=A, B, C, D), 20 plots repeated down the next (block= 1 to 20)
and records for metal volume (metal- 124 of these)

I have made treatment and block a factor. But haven't grouped them (do I
need to and how if so)

 

The main question is in 3 parts:

 

1.	is this the correct formula to use for this situation:
lme1<-lme(metal~treatment,data=data,random=~1|block) (or is lme even the
right thing to use here?)

 

I get:
> summary(lme1)
Linear mixed-effects model fit by REML

 Data: data 

       AIC      BIC    logLik

  365.8327 382.5576 -176.9163

 

Random effects:

 Formula: ~1 | block

        (Intercept)  Residual

StdDev:   0.4306096 0.9450976

 

Fixed effects: Cu ~ Treatment 

                 Value Std.Error  DF   t-value p-value

(Intercept)   5.587839 0.2632831 104 21.223688  0.0000 ***

TreatmentB -0.970384 0.3729675  16 -2.601792  0.0193 ***

TreatmentC  -1.449250 0.3656351  16 -3.963651  0.0011 ***

TreatmentD  -1.319564 0.3633837  16 -3.631323  0.0022 ***

 Correlation: 

             (Intr) TrtmAN TrtmCH

TreatmentB -0.706              

TreatmentC  -0.720  0.508       

TreatmentD  -0.725  0.511  0.522

 

Standardized Within-Group Residuals:

        Min          Q1         Med          Q3         Max 

-2.85762206 -0.68568460 -0.09004478  0.56237152  3.20650288 

 

Number of Observations: 124

Number of Groups: 20

 

2.	if so how can I get p values for comparisons between every
group... ie is A different from B, is A different from C, is A different
from D, is B different from C, is B different from D etc... is there a
way to get all of these instead of just "is A different from B, is A
different from C, is A different from D" which summary seems to give?
3.	last of all what is the best way to print out all the residuals
for lme... I can get qqplot(lme1) is there a pre-programmed call for
multiple diagnostic plots like in some other functions...

 

 

Thankyou so Much for your time....

 

It is much appreciated

;-)

 

Miki

 


	[[alternative HTML version deleted]]

Hi Miki,
I just got the same problem with you couple hours ago.
Rusers (Anna, and Mark {thank you guys}) provide me a vary valuable  
information.
link to following address.

http://www.nabble.com/Tukey-HSD-(or-other-post-hoc-tests)-following-repeated-measures-ANOVA-td17508294.html#a17559307
for the A vs. B, A vs. C....
You could install and download the multcomp package and perform the  
post hoc test
such as
summary(glht(lmel,linfct=mcp(treatment="Tukey")))

hopefully it helps
Chunhao


Quoting M Ensbey <m.ensbey at unimelb.edu.au>:
> Hi,
>
>
>
> I have searched the archives and can't quite confirm the answer to
this.
> I appreciate your time...
>
>
>
> I have 4 treatments (fixed) and I would like to know if there is a
> significant difference in metal volume (metal) between the treatments.
> The experiment has 5 blocks (random) in each treatment and no block is
> repeated across treatments. Within each plot there are varying numbers
> of replicates (random) (some plots have 4 individuals in them some have
> 14 and a range in between). NOTE the plots in one treatment are not
> replicated in the others.
>
>
>
> So I end up with a data.frame with 4 treatments repeated down one column
> (treatment=A, B, C, D), 20 plots repeated down the next (block= 1 to 20)
> and records for metal volume (metal- 124 of these)
>
> I have made treatment and block a factor. But haven't grouped them (do
I
> need to and how if so)
>
>
>
> The main question is in 3 parts:
>
>
>
> 1.	is this the correct formula to use for this situation:
> lme1<-lme(metal~treatment,data=data,random=~1|block) (or is lme even the
> right thing to use here?)
>
>
>
> I get:
>
>> summary(lme1)
>
> Linear mixed-effects model fit by REML
>
>  Data: data
>
>        AIC      BIC    logLik
>
>   365.8327 382.5576 -176.9163
>
>
>
> Random effects:
>
>  Formula: ~1 | block
>
>         (Intercept)  Residual
>
> StdDev:   0.4306096 0.9450976
>
>
>
> Fixed effects: Cu ~ Treatment
>
>                  Value Std.Error  DF   t-value p-value
>
> (Intercept)   5.587839 0.2632831 104 21.223688  0.0000 ***
>
> TreatmentB -0.970384 0.3729675  16 -2.601792  0.0193 ***
>
> TreatmentC  -1.449250 0.3656351  16 -3.963651  0.0011 ***
>
> TreatmentD  -1.319564 0.3633837  16 -3.631323  0.0022 ***
>
>  Correlation:
>
>              (Intr) TrtmAN TrtmCH
>
> TreatmentB -0.706
>
> TreatmentC  -0.720  0.508
>
> TreatmentD  -0.725  0.511  0.522
>
>
>
> Standardized Within-Group Residuals:
>
>         Min          Q1         Med          Q3         Max
>
> -2.85762206 -0.68568460 -0.09004478  0.56237152  3.20650288
>
>
>
> Number of Observations: 124
>
> Number of Groups: 20
>
>
>
> 2.	if so how can I get p values for comparisons between every
> group... ie is A different from B, is A different from C, is A different
> from D, is B different from C, is B different from D etc... is there a
> way to get all of these instead of just "is A different from B, is A
> different from C, is A different from D" which summary seems to give?
> 3.	last of all what is the best way to print out all the residuals
> for lme... I can get qqplot(lme1) is there a pre-programmed call for
> multiple diagnostic plots like in some other functions...
>
>
>
>
>
> Thankyou so Much for your time....
>
>
>
> It is much appreciated
>
> ;-)
>
>
>
> Miki
>
>
>
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>