R help - Oct 2017 - HELP relevel INTERCEPT-COMPARISONS

Dear colleagues,


How can I do to "relevel" the intercept?

I need that the treatment "Db" be the intercept, and have p-values for
the comparisons with the others treatments.

I used the function "relevel" but it did not work out to have what I
want.

Thanks for your help,


Xavier


T1 <- read.table(file.choose(), h=T)
> head(T1)
  treatment replicate Time     Ebcg
1   CHA0+Db         1    1  9681172
2   CHA0+Db         2    1  5599000
3   CHA0+Db         3    1 10280480
4   CHA0+Db         4    1  1004526
5   CHA0+Db         5    1 12942018
6   CHA0+Db         6    1  7478773
> attach(T1)
> summary(T1)
   treatment   replicate          Time        Ebcg
 CHA0   :6   Min.   :1.000   Min.   :1   Min.   :       0
 CHA0+Db:9   1st Qu.:2.750   1st Qu.:1   1st Qu.: 1000922
 Db     :9   Median :4.500   Median :1   Median : 4044094
 Healthy:9   Mean   :4.625   Mean   :1   Mean   : 5699717
 PCL    :6   3rd Qu.:6.250   3rd Qu.:1   3rd Qu.: 7785782
 PCL+Db :9   Max.   :9.000   Max.   :1   Max.  
:24736254
> hist(Ebcg)
> treatment<-relevel(treatment,ref="Db")
> mT1<-lm(sqrt(Ebcg)~treatment,data=T1)
> summary(mT1)
Call:
lm(formula = sqrt(Ebcg) ~ treatment, data = T1)

Residuals:
     Min       1Q   Median       3Q      Max
-1748.12  -540.19     0.97   545.14  1949.22

Coefficients:
                                 Estimate         Std. Error     t value    
Pr(>|t|)
(Intercept)                     532.5          384.4       1.385        
0.173296
treatmentCHA0+Db   2048.5          496.3       4.128         0.000170 ***
treatmentDb                2491.9          496.3       5.021         9.93e-06
***
treatmentHealthy       1215.6          496.3       2.450         0.018552 *
treatmentPCL                120.4          543.6       0.221         0.825816
treatmentPCL+Db        2098.9          496.3       4.229         0.000124 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05
'.' 0.1 ' ' 1

Residual standard error: 941.6 on 42 degrees of freedom
Multiple R-squared:  0.5206,    Adjusted R-squared:  0.4635
F-statistic: 9.121 on 5 and 42 DF,  p-value: 6.227e-06



Xavier Chiriboga M.
PhD Candidate
Fundamental and Applied Research in Chemical Ecology Lab.
Institute of Biology
University of Neuchatel

Website: https://www.unine.ch/farce/home/membres/xavier-chiriboga.html

	[[alternative HTML version deleted]]

On Sat, Oct 28, 2017 at 08:11:51PM +0000, CHIRIBOGA Xavier
wrote:
>> treatment<-relevel(treatment,ref="Db")
Never used relevel myself, but shouldn't this line be instead this one?

T1$treatment <- relevel(T1$treatment, ref = "Db")

Cheers,

JMM.

Dear Xavier

You problem comes from using attach(). Do not do that, it usually ends 
in tears. Either use the data parameter to the functions which support 
it or look at with() and within().

On 28/10/2017 21:11, CHIRIBOGA Xavier wrote:
> Dear colleagues,
> 
> 
> How can I do to "relevel" the intercept?
> 
> I need that the treatment "Db" be the intercept, and have
p-values for the comparisons with the others treatments.
> 
> I used the function "relevel" but it did not work out to have
what I want.
> 
> Thanks for your help,
> 
> 
> Xavier
> 
> 
> T1 <- read.table(file.choose(), h=T)
>> head(T1)
>    treatment replicate Time     Ebcg
> 1   CHA0+Db         1    1  9681172
> 2   CHA0+Db         2    1  5599000
> 3   CHA0+Db         3    1 10280480
> 4   CHA0+Db         4    1  1004526
> 5   CHA0+Db         5    1 12942018
> 6   CHA0+Db         6    1  7478773
>> attach(T1)
>> summary(T1)
>     treatment   replicate          Time        Ebcg
>   CHA0   :6   Min.   :1.000   Min.   :1   Min.   :       0
>   CHA0+Db:9   1st Qu.:2.750   1st Qu.:1   1st Qu.: 1000922
>   Db     :9   Median :4.500   Median :1   Median : 4044094
>   Healthy:9   Mean   :4.625   Mean   :1   Mean   : 5699717
>   PCL    :6   3rd Qu.:6.250   3rd Qu.:1   3rd Qu.: 7785782
>   PCL+Db :9   Max.   :9.000   Max.   :1   Max.   :24736254
>> hist(Ebcg)
>> treatment<-relevel(treatment,ref="Db")
>> mT1<-lm(sqrt(Ebcg)~treatment,data=T1)
>> summary(mT1)
> 
> Call:
> lm(formula = sqrt(Ebcg) ~ treatment, data = T1)
> 
> Residuals:
>       Min       1Q   Median       3Q      Max
> -1748.12  -540.19     0.97   545.14  1949.22
> 
> Coefficients:
>                                   Estimate         Std. Error     t value  
Pr(>|t|)
> (Intercept)                     532.5          384.4       1.385        
0.173296
> treatmentCHA0+Db   2048.5          496.3       4.128         0.000170 ***
> treatmentDb                2491.9          496.3       5.021        
9.93e-06 ***
> treatmentHealthy       1215.6          496.3       2.450         0.018552 *
> treatmentPCL                120.4          543.6       0.221        
0.825816
> treatmentPCL+Db        2098.9          496.3       4.229         0.000124
***
> ---
> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05
'.' 0.1 ' ' 1
> 
> Residual standard error: 941.6 on 42 degrees of freedom
> Multiple R-squared:  0.5206,    Adjusted R-squared:  0.4635
> F-statistic: 9.121 on 5 and 42 DF,  p-value: 6.227e-06
> 
> 
> 
> Xavier Chiriboga M.
> PhD Candidate
> Fundamental and Applied Research in Chemical Ecology Lab.
> Institute of Biology
> University of Neuchatel
> 
> Website: https://www.unine.ch/farce/home/membres/xavier-chiriboga.html
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.
> 
-- 
Michael
http://www.dewey.myzen.co.uk/home.html