I am trying to use multcomp to do a Tukey posthoc on growth increments among
genetic crosstypes.
#Fixed effect model
m1 <- lm(inc ~ 0 + Age+ Crosstype + Sex, data = Data.age)
summary(m1)
RESULTS of the model:
summary(m1)
Call:
lm(formula = inc ~ 0 + Age + Crosstype + Sex, data = Data.age)
Residuals:
Min 1Q Median 3Q Max
-0.87180 -0.34002 -0.02702 0.27710 2.17820
Coefficients:
Estimate Std. Error t value Pr(>|t|)
Age0 1.35164 0.03028 44.632 < 2e-16 ***
Age1 1.13451 0.03015 37.626 < 2e-16 ***
Age2 0.89568 0.03644 24.582 < 2e-16 ***
CrosstypeBGxB -0.21644 0.06198 -3.492 0.000500 ***
CrosstypeBGxG -0.22274 0.09273 -2.402 0.016484 *
CrosstypeF1_BXG 0.04361 0.03153 1.383 0.166946
CrosstypeF2_BXG -0.26120 0.07098 -3.680 0.000246 ***
CrosstypeG 0.08839 0.03683 2.400 0.016572 *
SexM 0.13177 0.02806 4.697 3.01e-06 ***
SexU -0.03137 0.04280 -0.733 0.463736
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 0.4173 on 1007 degrees of freedom
Multiple R-squared: 0.8999, Adjusted R-squared: 0.8989
F-statistic: 905.6 on 10 and 1007 DF, p-value: < 2.2e-16
When I try to use the "multcomp" I this is what happens:
tuk1 <- glht(m1, linfct = mcp(Crosstype="Tukey"))
Error in Ktotal[count:(count + nrow(h$K) - 1), h$where] <- h$K :
number of items to replace is not a multiple of replacement length
After reading through the numerous posts and vignettes, I tried this:
K <- contrMat(table(Data.age$Crosstype), type="Tukey")
and get this:
K
Multiple Comparisons of Means: Tukey Contrasts
B BGxB BGxG F1_BXG F2_BXG G
BGxB - B -1 1 0 0 0 0
BGxG - B -1 0 1 0 0 0
F1_BXG - B -1 0 0 1 0 0
F2_BXG - B -1 0 0 0 1 0
G - B -1 0 0 0 0 1
BGxG - BGxB 0 -1 1 0 0 0
F1_BXG - BGxB 0 -1 0 1 0 0
F2_BXG - BGxB 0 -1 0 0 1 0
G - BGxB 0 -1 0 0 0 1
F1_BXG - BGxG 0 0 -1 1 0 0
F2_BXG - BGxG 0 0 -1 0 1 0
G - BGxG 0 0 -1 0 0 1
F2_BXG - F1_BXG 0 0 0 -1 1 0
G - F1_BXG 0 0 0 -1 0 1
G - F2_BXG 0 0 0 0 -1 1
I get this message: Error in glht.matrix(m1, linfct = K) :
'ncol(linfct)' is not equal to 'length(coef(model))'
I get the same error when I try a variation of the model m1 as a mixed
effects model.
Can anyone tell my why I am getting this error and perhaps a direction for a
remedy. I am stuck.
--
View this message in context:
http://n4.nabble.com/error-when-using-multcomp-and-lm-tp964705p964705.html
Sent from the R help mailing list archive at Nabble.com.
______________________________________________
R-help@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.
