Diseño bloques completos al azar

Más información aqui

Datos

Descargar la base de datos aquí

Cargar la base de datos

# Install.pakages("readxl")
library(readxl)
datos <- read_excel("./datos/datos.xlsx", sheet = "Calculos_rendimiento_trigo")
str(datos)
## Classes 'tbl_df', 'tbl' and 'data.frame':    14 obs. of  40 variables:
##  $ Bloque      : num  1 1 1 1 1 1 1 2 2 2 ...
##  $ Parc        : num  5 7 8 10 11 12 14 16 17 19 ...
##  $ Trt         : num  1 11 7 13 5 9 3 7 1 5 ...
##  $ Wdry50S     : num  96.8 106.9 129.9 107.3 128.3 ...
##  $ Wgrain50S   : num  38.1 38.4 52.5 38.6 53.5 ...
##  $ WidthArea   : num  1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 ...
##  $ LengthArea  : num  9 9 9 9 9 9 9 9 9 9 ...
##  $ WGrain      : num  2370 2670 1732 3556 4277 ...
##  $ WHumid      : num  200 201 200 200 200 ...
##  $ WDry        : num  180 182 182 182 182 ...
##  $ W200g       : num  6.99 7.19 7.19 6.87 7.78 6.97 7.77 7.11 7.25 7.68 ...
##  $ Area        : num  13.5 13.5 13.5 13.5 13.5 13.5 13.5 13.5 13.5 13.5 ...
##  $ AreaCount   : num  0.187 0.187 0.187 0.187 0.125 0.187 0.187 0.187 0.187 0.125 ...
##  $ Count       : num  26.7 25.8 23.5 22.8 20.7 21.7 24.5 17.3 19.5 22.3 ...
##  $ Tasseling   : num  56 53 53 54 53 53 55 53 56 54 ...
##  $ Lodging     : num  0 0 0 0 0 0 0 0 0 0 ...
##  $ Flowering   : num  60 56 57 57 57 56 59 57 59 57 ...
##  $ Maturity    : num  107 106 107 108 107 106 107 106 106 107 ...
##  $ Height      : num  78.5 74.7 64.3 88.3 86.5 69.3 78 64.8 77.8 88.8 ...
##  $ HI          : num  39.4 35.9 40.5 36 41.7 ...
##  $ %Humidity   : num  0.0981 0.0937 0.0925 0.0944 0.0939 ...
##  $ H2O_Grain   : num  232 250 160 336 401 ...
##  $ Yield_Dry   : num  1583 1792 1164 2385 2871 ...
##  $ Yield_12_H2O: num  1799 2037 1323 2711 3262 ...
##  $ Biomass     : num  4019 4990 2878 6629 6888 ...
##  $ Straw       : num  2435 3198 1714 4244 4018 ...
##  $ Spikes/m2   : num  208 233 111 309 268 ...
##  $ Thou        : num  35 36 36 34.4 38.9 ...
##  $ Grains/m²   : num  4530 4986 3239 6945 7380 ...
##  $ Grains/spike: num  21.8 21.4 29.2 22.5 27.5 ...
##  $ WStem       : num  1.94 2.14 2.6 2.15 2.57 ...
##  $ WGrain/spike: num  0.763 0.768 1.051 0.772 1.069 ...
##  $ Plants/m²   : num  143 138 126 122 166 ...
##  $ DaysGrain   : num  47 50 50 51 50 50 48 49 47 50 ...
##  $ BioRate     : num  37.6 47.1 26.9 61.4 64.4 ...
##  $ GrainRateA  : num  16.8 19.2 12.4 25.1 30.5 ...
##  $ GrainRateB  : num  38.3 40.7 26.5 53.2 65.2 ...
##  $ VegRate     : num  40.6 57.1 30.1 74.5 70.5 ...
##  $ KernelRate  : num  0.744 0.719 0.719 0.674 0.778 ...
##  $ P.H.        : num  728 711 726 709 729 ...
datos <- transform(datos, Trt = factor(Trt))
datos <- transform(datos, Bloque = factor(Bloque))
str(datos)
## 'data.frame':    14 obs. of  40 variables:
##  $ Bloque      : Factor w/ 2 levels "1","2": 1 1 1 1 1 1 1 2 2 2 ...
##  $ Parc        : num  5 7 8 10 11 12 14 16 17 19 ...
##  $ Trt         : Factor w/ 7 levels "1","3","5","7",..: 1 6 4 7 3 5 2 4 1 3 ...
##  $ Wdry50S     : num  96.8 106.9 129.9 107.3 128.3 ...
##  $ Wgrain50S   : num  38.1 38.4 52.5 38.6 53.5 ...
##  $ WidthArea   : num  1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 ...
##  $ LengthArea  : num  9 9 9 9 9 9 9 9 9 9 ...
##  $ WGrain      : num  2370 2670 1732 3556 4277 ...
##  $ WHumid      : num  200 201 200 200 200 ...
##  $ WDry        : num  180 182 182 182 182 ...
##  $ W200g       : num  6.99 7.19 7.19 6.87 7.78 6.97 7.77 7.11 7.25 7.68 ...
##  $ Area        : num  13.5 13.5 13.5 13.5 13.5 13.5 13.5 13.5 13.5 13.5 ...
##  $ AreaCount   : num  0.187 0.187 0.187 0.187 0.125 0.187 0.187 0.187 0.187 0.125 ...
##  $ Count       : num  26.7 25.8 23.5 22.8 20.7 21.7 24.5 17.3 19.5 22.3 ...
##  $ Tasseling   : num  56 53 53 54 53 53 55 53 56 54 ...
##  $ Lodging     : num  0 0 0 0 0 0 0 0 0 0 ...
##  $ Flowering   : num  60 56 57 57 57 56 59 57 59 57 ...
##  $ Maturity    : num  107 106 107 108 107 106 107 106 106 107 ...
##  $ Height      : num  78.5 74.7 64.3 88.3 86.5 69.3 78 64.8 77.8 88.8 ...
##  $ HI          : num  39.4 35.9 40.5 36 41.7 ...
##  $ X.Humidity  : num  0.0981 0.0937 0.0925 0.0944 0.0939 ...
##  $ H2O_Grain   : num  232 250 160 336 401 ...
##  $ Yield_Dry   : num  1583 1792 1164 2385 2871 ...
##  $ Yield_12_H2O: num  1799 2037 1323 2711 3262 ...
##  $ Biomass     : num  4019 4990 2878 6629 6888 ...
##  $ Straw       : num  2435 3198 1714 4244 4018 ...
##  $ Spikes.m2   : num  208 233 111 309 268 ...
##  $ Thou        : num  35 36 36 34.4 38.9 ...
##  $ Grains.m²   : num  4530 4986 3239 6945 7380 ...
##  $ Grains.spike: num  21.8 21.4 29.2 22.5 27.5 ...
##  $ WStem       : num  1.94 2.14 2.6 2.15 2.57 ...
##  $ WGrain.spike: num  0.763 0.768 1.051 0.772 1.069 ...
##  $ Plants.m²   : num  143 138 126 122 166 ...
##  $ DaysGrain   : num  47 50 50 51 50 50 48 49 47 50 ...
##  $ BioRate     : num  37.6 47.1 26.9 61.4 64.4 ...
##  $ GrainRateA  : num  16.8 19.2 12.4 25.1 30.5 ...
##  $ GrainRateB  : num  38.3 40.7 26.5 53.2 65.2 ...
##  $ VegRate     : num  40.6 57.1 30.1 74.5 70.5 ...
##  $ KernelRate  : num  0.744 0.719 0.719 0.674 0.778 ...
##  $ P.H.        : num  728 711 726 709 729 ...
names(datos)
##  [1] "Bloque"       "Parc"         "Trt"          "Wdry50S"     
##  [5] "Wgrain50S"    "WidthArea"    "LengthArea"   "WGrain"      
##  [9] "WHumid"       "WDry"         "W200g"        "Area"        
## [13] "AreaCount"    "Count"        "Tasseling"    "Lodging"     
## [17] "Flowering"    "Maturity"     "Height"       "HI"          
## [21] "X.Humidity"   "H2O_Grain"    "Yield_Dry"    "Yield_12_H2O"
## [25] "Biomass"      "Straw"        "Spikes.m2"    "Thou"        
## [29] "Grains.m²"    "Grains.spike" "WStem"        "WGrain.spike"
## [33] "Plants.m²"    "DaysGrain"    "BioRate"      "GrainRateA"  
## [37] "GrainRateB"   "VegRate"      "KernelRate"   "P.H."

Con la función agregate() se pueden hacer resumenes por grupo. Las opciones son: sum, mean, median, sd, se, min, max.

aggregate(Yield_12_H2O ~ Trt, data = datos, FUN = mean)
##   Trt Yield_12_H2O
## 1   1     1750.456
## 2   3     2874.998
## 3   5     3364.312
## 4   7     1037.812
## 5   9     1681.568
## 6  11     2309.546
## 7  13     2693.390

Utilizar la función boxplot para construir un diagrama

boxplot(datos$Yield_12_H2O ~ datos$Trt, xlab="Tratamiento", ylab="Rendimiento 12% H2O (kg ha-1)")

Construir un diagrama de interacción

interaction.plot(datos$Trt, datos$Bloque, datos$Yield_12_H2O, xlab="Tratamiento",ylab="Rendimiento 12% H2O (kg ha-1)")

interaction.plot(datos$Bloque, datos$Trt, datos$Yield_12_H2O, xlab="Bloque",ylab="Rendimiento 12% H2O (kg ha-1)")

Realizar un ANOVA de acuerdo al modelo para un Randomized Block design

ajuste <- lm(datos$Yield_12_H2O ~ datos$Bloque + datos$Trt)
anova(ajuste)
## Analysis of Variance Table
## 
## Response: datos$Yield_12_H2O
##              Df  Sum Sq Mean Sq F value    Pr(>F)    
## datos$Bloque  1       0       0   0.000 0.9995682    
## datos$Trt     6 7748613 1291436  22.192 0.0007507 ***
## Residuals     6  349161   58193                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Prueba de medias con el paquete agricolae

#install.packages("agricolae")
library(agricolae)
comparaciones <- HSD.test(ajuste, "datos$Trt") 
comparaciones
## $statistics
##       Mean       CV  MSerror      HSD
##   2244.583 10.74735 58193.47 1005.608
## 
## $parameters
##   Df ntr StudentizedRange alpha  test    name.t
##    6   7         5.895309  0.05 Tukey datos$Trt
## 
## $means
##    datos$Yield_12_H2O       std r       Min      Max
## 1            1750.456  69.06707 2 1701.6183 1799.294
## 11           2309.546 385.56019 2 2036.9138 2582.178
## 13           2693.390  24.61455 2 2675.9849 2710.795
## 3            2874.998  58.10076 2 2833.9146 2916.081
## 5            3364.312 144.38696 2 3262.2146 3466.409
## 7            1037.812 403.34192 2  752.6061 1323.018
## 9            1681.568  90.66437 2 1617.4590 1745.678
## 
## $comparison
## NULL
## 
## $groups
##   trt    means   M
## 1  5  3364.312   a
## 2  3  2874.998  ab
## 3  13 2693.390 abc
## 4  11 2309.546 bcd
## 5  1  1750.456 cde
## 6  9  1681.568  de
## 7  7  1037.812   e

Otras comparaciones posibles

  • The Least Significant Difference (LSD), LSD.test()
  • Duncan’s New Multiple-Range Test, duncan.test()
  • Student-Newman-Keuls, SNK.test()
  • Ryan, Einot and Gabriel and Welsch, REGW.test()
  • Tukey’s HSD (Honestly Significant Differences), HSD.test()
  • Scheffe’s Test, scheffe.test()

El ajuste en otros diseños

Diseño completamente al azar

ajuste = lm(r ~ tratamiento)
Más información aqui

Diseño factorial

ajuste = lm(r ~ tratamiento1 * tratamiento2)
Más información aqui

Diseño cuadrado latino

ajuste = lm(r ~ fila + columna + tratamiento)
Más información aqui