michael.eisenring at agroscope.admin.ch
2015-Dec-30 13:06 UTC
[R] "Group" argument of kruskal function (agricolae package) does not coincide with p-values
Hello, I work with the kruskla function of the agricolae package to conduct Kurskal-Wallis tests. The kurskal function has the argument "group"=T/F. If group=T, the output of the kruskal test assigns a "significance letter" to each mean of each tested treatment (means with the same letter are not significantly different). However, if I compare the "significance letters" with the p-values (obtained if group=F) sometimes the sign. letters and p-values do not coincide. E.g. in my example the treatments "2" and "2C" differ significantly from each other (p.value=0.0332, level of significance=0.05) , however both of them are assigned the letter "a" if group=T is used. Does anyone know why that's the case? First I thought is because I use a holm correction, but this should already be seen in the actual p-values. The problem is that I would like to use the "sign. letters" in one of my figures (so the reader can easily distinguish which treatments are different from each other) but also use the p-values in the text (and of course sometimes the letters and p-values do not coincide) Thank you Mike EXAMPLE: #Below is the code and the dput of my data frame (called "example") kruskal(example$H.polone,example$Position, p.adj = 'holm', console = T, group = F) #group=F shows p-values, group=T shows sign. letters> dput(example)structure(list(leaf = structure(c(2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L), .Label = c("Cot", "L1", "L2", "L4", "L7"), class = "factor"), H.polone = c(980.310618, 2474.133482, 4406.901817, 2233.141989, 1284.855658, 1202.881507, 2393.532397, 2639.775244, 2853.759572, 3009.791049, 2734.731629, 1539.309599, 9634.265878, 15627.14214, 13212.8464, 4.679175916, 5004.820592, 17045.7964, 17461.49723, 14065.03002, 24452.43564, 17267.91494, 9483.597645, 535.05702, 582.0554615, 163.9588478, 527.1015784, 621.3924963, 519.8953943, 1429.334963, 584.2620172, 600.5646899, 568.2045799, 583.0237882, 1054.557537, 576.5345625, 366.6277524, 644.2932199, 877.2948815, 343.9021574, 960.3431623, 899.6237855, 834.5025197), Position = structure(c(6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L), .Label = c("1", "1C", "1L4", "1L4C", "2", "2C", "2L4", "2L4C", "3", "3C", "3L4", "3L4C", "4L4", "4L4C", "5L4", "5L4C"), class = "factor")), class = "data.frame", row.names = c(NA, -43L), .Names = c("leaf", "H.polone", "Position")) Eisenring Michael, Msc. PhD Student Federal Department of Economic Affairs, Education and Research EAER Institute of Sustainability Sciences ISS Biosafety Reckenholzstrasse 191, CH-8046 Z?rich Tel. +41 44 37 77181 Fax +41 44 37 77201 michael.eisenring at agroscope.admin.ch<mailto:michael.eisenring at agroscope.admin.ch> www.agroscope.ch<http://www.agroscope.ch/> [[alternative HTML version deleted]]