[R] Fisher's LSD problem

cjohnst cjohnst at uga.edu
Fri Dec 13 20:55:57 CET 2013


I have been trying to run a Fisher's LSD for quite some time now, however the
output I am receiving only ranks the means of each treatment in terms of
significant difference, and will not give me the actual value of "Least
Significant Difference" I am looking for.  I am using a data set that has
four replications and two factors ("Ratio" and "Species"), however for this
particular dataset "Species" was not a significant factor so I am only
interested in the LSD value for comparing the "Ratio" factor.  Note that the
response variable for two of the ratios only has three values (the fourth
rep had to be deleted due to impossibility), and I do not know if this may
be the source of my problem.  Regardless, here is my script:

f <- file.choose()
svs30dayshoot <- read.csv(f)
svs30dayshoot
str(svs30dayshoot)
svs30dayshoot$Ratio<-factor(svs30dayshoot$Ratio)

aov30dayshoot <- aov(RY ~ Species*Ratio, data=svs30dayshoot)
summary(aov30dayshoot)

install.packages('agricolae')
library(agricolae)
LSD<-LSD.test(aov30dayshoot,"Ratio", p.adj="bonferroni") 
LSD


And the resulting output is below.  Note there is no "Least Significant
Difference" value given.  In fact, this output looks nothing like the output
I'm used to seeing when running Fisher's LSD with a one-way ANOVA (in my
class):

$statistics
       Mean       CV   MSerror
  0.6804658 107.7707 0.5377916

$parameters
  Df ntr bonferroni
  28   5   3.046929

$means
           RY   std.err r        LCL       UCL   Min.   Max.
0   0.0000000 0.2592758 8 -0.5311024 0.5311024 0.0000 0.0000
100 1.0000000 0.2592758 8  0.4688976 1.5311024 1.0000 1.0000
25  0.3140571 0.2771775 7 -0.2537152 0.8818295 0.0200 0.5069
50  0.8016875 0.2592758 8  0.2705851 1.3327899 0.3800 1.5952
75  1.3208286 0.2771775 7  0.7530562 1.8886010 0.3864 4.8889

$comparison
NULL

$groups
  trt     means  M
1 75  1.3208286  a
2 100 1.0000000 ab
3 50  0.8016875 ab
4 25  0.3140571 ab
5 0   0.0000000  b

Any help would be greatly appreciated!



--
View this message in context: http://r.789695.n4.nabble.com/Fisher-s-LSD-problem-tp4682153.html
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list