[R-sig-ME] glth function with lme returns error Error in contrMat(table(mf[[nm]]), type = types[pm]) : less than two groups

[Thierry Onkelinx]

Dear Annie,

Please don't use attach(). It's a recipe for disaster. Use the data
argument instead.

glht() can't handle interaction terms. You will have to specify the
contrasts manually. A workaround is to create a new variable that
holds the interaction an use that in the model.

spl$Interaction <- interaction(spl$treat, spl$week)
modcover <- lme(logcov ~ Interaction, random= ~1 | rep / week, data = spl)
summary(glht(modcover, linfct = mcp(Interaction = "Tukey")))

Best regards,

Thierry

Op 11-mrt.-2016 23:03 schreef "Annie-Claude Letendre"
<annieclaude.letendre op gmail.com>:
>
> Hi,
>
> I am trying to run a *posthoc* on a lme with the glht function and I get
> this error that I can't figure out "Error in contrMat(table(mf[[nm]]), type
> = types[pm]) :
>   less than two groups"
>
> My experiment had 45 indivuduals (rep) each of these had 1 treatment
> (treat) applied to it and then the variable (logcov) was measured twice for
> each individual at week 6 and 12.
>
> I've attached the csv I am using , thanks for any help you can offer !
> Is there any other posthoc test that could work for me ?
>
> Annie
>
> >
> spl<-read.table("C:/Users/Annie/Desktop/splcover.csv",header=T,sep=",",quote="")
> > attach(spl)
> > names(spl)
>  [1] "rep"     "treat"  "logcov"  "week"
>
>
> > rep<-factor(rep)
> > treat<-factor(treat)
> > week<-factor(week)
> > levels(treat)
> [1] "c" "f" "g" "p" "r"
> > levels(week)
> [1] "6"  "12"
> > levels(rep)
>  [1] "1"  "2"  "3"  "4"  "5"  "6"  "7"  "8"  "9"  "10" "11" "12" "13" "14"
> "15"
> [16] "16" "17" "18" "19" "20" "21" "22" "23" "24" "25" "26" "27" "28" "29"
> "30"
> [31] "31" "32" "33" "34" "35" "36" "37" "38" "39" "40" "41" "42" "43" "44"
> "45"
>
> > library(nlme)
> Warning message:
> package ‘nlme’ was built under R version 3.1.3
>
> > summary(modcover<-lme(logcov~treat*week, random= ~1|rep/week))
> Linear mixed-effects model fit by REML
>  Data: NULL
>         AIC       BIC   logLik
>   -33.51971 -2.553367 29.75986
>
> Random effects:
>  Formula: ~1 | rep
>         (Intercept)
> StdDev:  0.09534288
>
>  Formula: ~1 | week %in% rep
>         (Intercept)   Residual
> StdDev:   0.1095635 0.04379889
>
> Fixed effects: logcov ~ treat * week
>                    Value  Std.Error DF   t-value p-value
> (Intercept)   -2.0000000 0.05056654 40 -39.55185  0.0000
> treatf         1.5944444 0.07151188 40  22.29622  0.0000
> treatg         0.0000000 0.07151188 40   0.00000  1.0000
> treatp         1.3944444 0.07151188 40  19.49948  0.0000
> treatr         1.5744444 0.07151188 40  22.01654  0.0000
> week12         0.1400000 0.05562275 40   2.51696  0.0159
> treatf:week12 -0.0266667 0.07866245 40  -0.33900  0.7364
> treatg:week12  0.1888889 0.07866245 40   2.40126  0.0211
> treatp:week12 -0.0222222 0.07866245 40  -0.28250  0.7790
> treatr:week12  0.0077778 0.07866245 40   0.09888  0.9217
>  Correlation:
>               (Intr) treatf treatg treatp treatr week12 trtf:12 trtg:12
> trtp:12
> treatf        -0.707
>
> treatg        -0.707  0.500
>
> treatp        -0.707  0.500  0.500
>
> treatr        -0.707  0.500  0.500  0.500
>
> week12        -0.550  0.389  0.389  0.389  0.389
>
> treatf:week12  0.389 -0.550 -0.275 -0.275 -0.275 -0.707
>
> treatg:week12  0.389 -0.275 -0.550 -0.275 -0.275 -0.707  0.500
>
> treatp:week12  0.389 -0.275 -0.275 -0.550 -0.275 -0.707  0.500   0.500
>
> treatr:week12  0.389 -0.275 -0.275 -0.275 -0.550 -0.707  0.500   0.500
> 0.500
>
> Standardized Within-Group Residuals:
>          Min           Q1          Med           Q3          Max
> -0.741681806 -0.152850434 -0.008438999  0.126023769  0.836897713
>
> Number of Observations: 90
> Number of Groups:
>           rep week %in% rep
>            45            90
> > anova(modcover)
>             numDF denDF   F-value p-value
> (Intercept)     1    40 2817.8201  <.0001
> treat           4    40  366.2476  <.0001
> week            1    40   46.4611  <.0001
> treat:week      4    40    2.6323  0.0483
> > require(multcomp)
> Loading required package: multcomp
> Loading required package: mvtnorm
> Loading required package: survival
> Loading required package: splines
> Loading required package: TH.data
>
>
> > summary(glht(modcover, linfct=mcp(treat="Tukey")))
> Error in contrMat(table(mf[[nm]]), type = types[pm]) :
>   less than two groups
>
> > is.factor(treat)
> [1] TRUE
>
> > sessionInfo()
> R version 3.1.1 (2014-07-10)
> Platform: x86_64-w64-mingw32/x64 (64-bit)
>
> locale:
> [1] LC_COLLATE=English_Canada.1252  LC_CTYPE=English_Canada.1252
> [3] LC_MONETARY=English_Canada.1252 LC_NUMERIC=C
> [5] LC_TIME=English_Canada.1252
>
> attached base packages:
> [1] splines   stats     graphics  grDevices utils     datasets  methods
> [8] base
>
> other attached packages:
> [1] multcomp_1.3-6  TH.data_1.0-3   survival_2.37-7 mvtnorm_1.0-0
> [5] nlme_3.1-125
>
> loaded via a namespace (and not attached):
>  [1] colorspace_1.2-4 ggplot2_2.0.0    grid_3.1.1       gtable_0.1.2
>  [5] lattice_0.20-33  munsell_0.4.2    plyr_1.8.3       Rcpp_0.12.2
>  [9] sandwich_2.3-2   scales_0.3.0     zoo_1.7-11
> _______________________________________________
> R-sig-mixed-models op r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models