[R] - dunn.test gives strange output

[Joachim Audenaert]

Dear all,

I have a question concerning my output of the dunn.test function in R. I 
like to compare different datasets, which are not distributed normally, so 
I use the Dunn.test to do pairwise comparison. I have 2 questions 
concerning the output:

- why are my groupnames changed, the output gives 4 times swirskii, 
although my groupnames are longer (see my dataset)
- secondly when I check the p-values, I see something very odd: I see that 
p values where the same groups are compared sometimes indicate that they 
are different, why are not all those p-values = 1 ?

This is my dataset:

structure(list(controle = c(111, 88, 216, 169), chemie = c(47, 31, 35, 
30), IPM = c(0, 0, 0, 1), gallicus = c(102, 152, 102, 75), swirskii3 = 
c(1, 0, 0, 0), swirskiiA = c(0, 0, 1, 3), swirskiiP = c(0, 0, 1, 6), 
swirskii1x = c(12, 2, 75, 46)), .Names = c("controle", "chemie", "IPM", 
"gallicus", "swirskii3", "swirskiiA", "swirskiiP", "swirskii1x"), 
row.names = c(NA, 4L), class = "data.frame")

I get the following output:

 DUNN <- dunn.test(value, variable, method = "none",kw=TRUE)
 
  Kruskal-Wallis rank sum test

data: value and variable
Kruskal-Wallis chi-squared = 26.8894, df = 7, p-value = 0


                        Comparison of value by variable  
                                (No adjustment)  
Col Mean-|
Row Mean |   controle     chemie        IPM   gallicus   swirskii swirskii
---------+------------------------------------------------------------------
  chemie |  -1.341044  -3.410083  -2.069039  -0.325682   1.015361 3.084401
         |     0.0900     0.0003     0.0193     0.3723     0.1550 0.0010
         |
     IPM |  -3.410083  -2.069039  -0.325682   1.015361   3.084401 
-3.410083
         |     0.0003     0.0193     0.3723     0.1550     0.0010 0.0003
         |
gallicus |  -0.325682   1.015361   3.084401  -3.410083  -2.069039 0.000000
         |     0.3723     0.1550     0.0010     0.0003     0.0193 0.5000
         |
swirskii |  -3.410083  -2.069039   0.000000  -3.084401  -3.007770 
-1.666726
         |     0.0003     0.0193     0.5000     0.0010     0.0013 0.0478
         |
swirskii |  -3.007770  -1.666726   0.402313  -2.682088   0.402313 
-2.969454
         |     0.0013     0.0478     0.3437     0.0037     0.3437     
0.0015
         |
swirskii |  -2.969454  -1.628410   0.440628  -2.643772   0.440628 0.038315
         |     0.0015     0.0517     0.3297     0.0041     0.3297 0.4847
         |
swirskii |  -1.475148  -0.134104   1.934935  -1.149466   1.934935 1.532621
         |     0.0701     0.4467     0.0265     0.1252     0.0265 0.0627
Col Mean-|
Row Mean |   swirskii
---------+-----------
  chemie |  -3.410083
         |     0.0003
         |
     IPM |  -2.069039
         |     0.0193
         |
gallicus |  -3.084401
         |     0.0010
         |
swirskii |   0.402313
         |     0.3437
         |
swirskii |  -1.628410
         |     0.0517
         |
swirskii |  -1.475148
         |     0.0701
         |
swirskii |   1.494306
         |     0.0675 




Met vriendelijke groeten - With kind regards,

Joachim Audenaert 
onderzoeker gewasbescherming - crop protection researcher

PCS | proefcentrum voor sierteelt - ornamental plant research

Schaessestraat 18, 9070 Destelbergen, België
T: +32 (0)9 353 94 71 | F: +32 (0)9 353 94 95
E: joachim.audenaert at pcsierteelt.be | W: www.pcsierteelt.be   

Heb je je individuele begeleiding bemesting (CVBB) al aangevraagd? | Het 
PCS op LinkedIn
Disclaimer | Please consider the environment before printing. Think green, 
keep it on the screen!
	[[alternative HTML version deleted]]