gomez.stripsplitplot.Rd
A strip-split-plot experiment with three reps, genotype as the horizontal strip, nitrogen fertilizer as the vertical strip, and planting method as the subplot factor.
yield
grain yield in kg/ha
planting
planting factor, P1=broadcast, P2=transplanted
rep
rep, 3 levels
nitro
nitrogen fertilizer, kg/ha
gen
genotype, G1 to G6
col
column
row
row
Note, this is a superset of the the 'gomez.stripplot' data.
Used with permission of Kwanchai Gomez.
Gomez, K.A. and Gomez, A.A.. 1984, Statistical Procedures for Agricultural Research. Wiley-Interscience. Page 155.
# \dontrun{ library(agridat) data(gomez.stripsplitplot) dat <- gomez.stripsplitplot # Layout libs(desplot) desplot(dat, gen ~ col*row, out1=rep, col=nitro, text=planting, cex=1, main="gomez.stripsplitplot")# Gomez table 4.19, ANOVA of strip-split-plot design dat <- transform(dat, nf=factor(nitro)) m1 <- aov(yield ~ nf * gen * planting + Error(rep + rep:nf + rep:gen + rep:nf:gen), data=dat) summary(m1)#> #> Error: rep #> Df Sum Sq Mean Sq F value Pr(>F) #> Residuals 2 15289498 7644749 #> #> Error: rep:nf #> Df Sum Sq Mean Sq F value Pr(>F) #> nf 2 116489166 58244583 36.62 0.00268 ** #> Residuals 4 6361491 1590373 #> --- #> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 #> #> Error: rep:gen #> Df Sum Sq Mean Sq F value Pr(>F) #> gen 5 49119270 9823854 3.676 0.0379 * #> Residuals 10 26721828 2672183 #> --- #> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 #> #> Error: rep:nf:gen #> Df Sum Sq Mean Sq F value Pr(>F) #> nf:gen 10 24595731 2459573 2.575 0.0344 * #> Residuals 20 19106733 955337 #> --- #> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 #> #> Error: Within #> Df Sum Sq Mean Sq F value Pr(>F) #> planting 1 723079 723079 1.715 0.1986 #> nf:planting 2 2468132 1234066 2.927 0.0664 . #> gen:planting 5 23761441 4752288 11.271 1.37e-06 *** #> nf:gen:planting 10 7512072 751207 1.782 0.1000 . #> Residuals 36 15179354 421649 #> --- #> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# There is a noticeable linear trend along the y coordinate which may be # an artifact that blocking will remove, or may need to be modeled. # Note the outside values in the high-nitro boxplot. libs("HH") interaction2wt(yield ~ nitro + gen + planting + row, dat, x.between=0, y.between=0, x.relation="free")# }