Análisis estadístico del experimento de fisiología vegetal: La aplicación de ácido ascórbico mitiga los efectos del estrés hídrico en plantas de rábano Raphanus sativus L., variedad Crimson Giant
Jose Gerardo Bermúdez Díaz
2023-10-31
\[ \LARGE{\text{Tratamientos:}\\\text{ }\\Rojo: 100\% \text{ de capacidad de matera (riego completo)}\\Amarillo: 50\% \text{ - Ácido ascórbico (AS)}\\Azul: 50\% \text{ + AS a 600 partes por millón (ppm)}\\Verde: 50\% \text{ + AS a 1000 partes por millón (ppm)}}\]
Librerias
library(datasets)
library(readxl)
library(tidyverse)## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.4
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ ggplot2 3.4.4 ✔ tibble 3.2.1
## ✔ lubridate 1.9.3 ✔ tidyr 1.3.0
## ✔ purrr 1.0.2
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errorslibrary(ggplot2)
library(multcompView)
library(dplyr)
library(agricolae)Lectura de datos de cada muestreo.
# Para que me indique la ubicación / dirección del documento que se requiere leer entre mis archivos, ejemplo: "/cloud/project/Datos x muestreo (corregido).xlsx".
# file.choose()# Para buscar carpetas de los archivos: excel.sheets()
# Primer muestreo --------------------------------------------
M1 <- "/cloud/project/Datos primer muestreo (corregido).xlsx"
# Carpetas
Var_M1 <- read_excel(M1,
sheet = 'Variables')
CRA_M1 <- read_excel(M1,
sheet = 'CRA')
Ind_M1 <- read_excel(M1,
sheet = 'Indices')
# Segundo muestreo --------------------------------------------
M2 <- "/cloud/project/Datos segundo muestreo (corregido).xlsx"
# Carpetas
Var_M2 <- read_excel(M2,
sheet = 'Variables')
CRA_M2 <- read_excel(M2,
sheet = 'CRA')
Ind_M2 <- read_excel(M2,
sheet = 'Indices')
# Tercer muestreo --------------------------------------------
M3 <- "/cloud/project/Datos tercer muestreo (corregido).xlsx"
# Carpetas
Var_M3 <- read_excel(M3,
sheet = 'Variables')
CRA_M3 <- read_excel(M3,
sheet = 'CRA')
PE_M3 <- read_excel(M3,
sheet = 'Pérdida de electrolitos')
Ind_M3 <- read_excel(M3,
sheet = 'Indices')
# Cuarto muestreo --------------------------------------------
M4 <- "/cloud/project/Datos cuarto muestreo (corregido).xlsx"
# Carpetas
Var_M4 <- read_excel(M4,
sheet = 'Variables')
CRA_M4 <- read_excel(M4,
sheet = 'CRA')
Ind_M4 <- read_excel(M4,
sheet = 'Indices')Temperatura de la hoja
Análisis estadístico - Muestreo 1
# Datos - Temperatura de la hoja - Muestreo 1
datos.TH.M1 <- data.frame(Var_M1$`Factor riego`, Var_M1$`Factor ácido ascórbico`, Var_M1$Tratamiento, Var_M1$`Temperatura Hoja (°C)`)
colnames(datos.TH.M1) <- c("F_riego", "F_AA", "Tratamiento", "T_Hoja")
datos.TH.M1datos.TH.M1$F_riego <- as.factor(datos.TH.M1$F_riego)
datos.TH.M1$F_AA <- as.factor(datos.TH.M1$F_AA)
str(datos.TH.M1)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ T_Hoja : num 19.2 20 19.2 18.3 20.2 ...# Anova - Temperatura de la hoja - Muestreo 1
A.TH.M1 <- aov(T_Hoja ~ F_riego*F_AA, data = datos.TH.M1)
summary(A.TH.M1)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 0.626 0.6256 1.382 0.263
## F_AA 2 2.352 1.1758 2.598 0.115
## Residuals 12 5.431 0.4526- Los dos tipos de riego y los tres tipos de concentración, como factores independientes, no tienen diferencias significativas, por lo que los resultados no se le puede atribuir a ninguno de estos dos factores, adicionalmente, no hay interacción entre los dos factores, por lo que no hay una combinación representativa de riego y ácido ascórbico para el muestreo 1.
# Prueba de normalidad de residuos
shapiro.test(A.TH.M1$residuals)##
## Shapiro-Wilk normality test
##
## data: A.TH.M1$residuals
## W = 0.93324, p-value = 0.2742- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.TH.M1)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = T_Hoja ~ F_riego * F_AA, data = datos.TH.M1)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 -0.4566667 -1.30294 0.3896065 0.2625014
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.7891667 -0.3099165 1.8882498 0.1766491
## 600 ppm-0 ppm -0.1358333 -1.2349165 0.9632498 0.9421195
## 600 ppm-1000 ppm -0.9250000 -2.1941119 0.3441119 0.1688108- Los factores no presentan diferencias significativas entre si.
# Anova - Temperatura de la hoja - Muestreo 1 - Comparación entre los tratamientos
A.TH.M1.CT <- aov(T_Hoja ~ Tratamiento, data=datos.TH.M1)
summary(A.TH.M1.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 2.977 0.9924 2.193 0.142
## Residuals 12 5.431 0.4526Tukey_A.TH.M1.CT <- TukeyHSD(A.TH.M1.CT)
Tukey_A.TH.M1.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = T_Hoja ~ Tratamiento, data = datos.TH.M1)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 0.9525 -0.4598164 2.3648164 0.2403404
## 50 + AS 600ppm-50 - AS 0.0275 -1.3848164 1.4398164 0.9999265
## Control-50 - AS -0.1300 -1.5423164 1.2823164 0.9924881
## 50 + AS 600ppm-50 + AS 1000ppm -0.9250 -2.3373164 0.4873164 0.2613531
## Control-50 + AS 1000ppm -1.0825 -2.4948164 0.3298164 0.1585399
## Control-50 + AS 600ppm -0.1575 -1.5698164 1.2548164 0.9868550- Podemos observar que ninguno de los tratamientos tiene diferencias significativas con el control para el muestreo 1.
A.TH.M1.DT <- duncan.test(A.TH.M1.CT, 'Tratamiento', console = T)##
## Study: A.TH.M1.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for T_Hoja
##
## Mean Square Error: 0.4525875
##
## Tratamiento, means
##
## T_Hoja std r se Min Max Q25 Q50 Q75
## 50 - AS 19.3125 0.7474122 4 0.3363731 18.41 20.24 19.0625 19.300 19.550
## 50 + AS 1000ppm 20.2650 0.7558439 4 0.3363731 19.34 21.06 19.8350 20.330 20.760
## 50 + AS 600ppm 19.3400 0.4298837 4 0.3363731 18.78 19.80 19.1550 19.390 19.575
## Control 19.1825 0.7040064 4 0.3363731 18.28 20.00 18.9775 19.225 19.430
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 1.036469 1.084885 1.114220
##
## Means with the same letter are not significantly different.
##
## T_Hoja groups
## 50 + AS 1000ppm 20.2650 a
## 50 + AS 600ppm 19.3400 a
## 50 - AS 19.3125 a
## Control 19.1825 aA.TH.M1.DT## $statistics
## MSerror Df Mean CV
## 0.4525875 12 19.525 3.445563
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 1.036469
## 3 3.225244 1.084885
## 4 3.312453 1.114220
##
## $means
## T_Hoja std r se Min Max Q25 Q50 Q75
## 50 - AS 19.3125 0.7474122 4 0.3363731 18.41 20.24 19.0625 19.300 19.550
## 50 + AS 1000ppm 20.2650 0.7558439 4 0.3363731 19.34 21.06 19.8350 20.330 20.760
## 50 + AS 600ppm 19.3400 0.4298837 4 0.3363731 18.78 19.80 19.1550 19.390 19.575
## Control 19.1825 0.7040064 4 0.3363731 18.28 20.00 18.9775 19.225 19.430
##
## $comparison
## NULL
##
## $groups
## T_Hoja groups
## 50 + AS 1000ppm 20.2650 a
## 50 + AS 600ppm 19.3400 a
## 50 - AS 19.3125 a
## Control 19.1825 a
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 2
# Datos - Temperatura de la hoja - Muestreo 2
datos.TH.M2 <- data.frame(Var_M2$`Factor riego`, Var_M2$`Factor ácido ascórbico`, Var_M2$Tratamiento, Var_M2$`Temperatura Hoja (°C)`)
colnames(datos.TH.M2) <- c("F_riego", "F_AA", "Tratamiento", "T_Hoja")
datos.TH.M2datos.TH.M2$F_riego <- as.factor(datos.TH.M2$F_riego)
datos.TH.M2$F_AA <- as.factor(datos.TH.M2$F_AA)
str(datos.TH.M2)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ T_Hoja : num 19.2 17.2 19.1 20 15.8 ...# Anova - Temperatura de la hoja - Muestreo 2
A.TH.M2 <- aov(T_Hoja ~ F_riego*F_AA, data = datos.TH.M2)
summary(A.TH.M2)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 39.49 39.49 14.158 0.00271 **
## F_AA 2 2.73 1.36 0.489 0.62493
## Residuals 12 33.47 2.79
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- El riego, como factor independiente, tiene diferencias signfiicativas, por lo que los resultados de la variables respuesta es estadísticamente atribuible a los tipos de riego. Por parte de la concentración de ácido ascórbico, no hay diferencias significativas, por lo que este aun no representa alguna alteración en la variable respuesta.
A.TH.M2.I <- aov(T_Hoja ~ F_riego:F_AA, data = datos.TH.M2)
summary(A.TH.M2.I)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego:F_AA 3 42.22 14.074 5.046 0.0173 *
## Residuals 12 33.47 2.789
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Adicionalmente, hay una interacción del 95% de significancia entre los dos factores, lo que significa que hay una combinación de riego y ácido ascórbico representativa para el muestreo 2.
# Prueba de normalidad de residuos
shapiro.test(A.TH.M2$residuals)##
## Shapiro-Wilk normality test
##
## data: A.TH.M2$residuals
## W = 0.91517, p-value = 0.141- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una organización normal.
TukeyHSD(A.TH.M2.I)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = T_Hoja ~ F_riego:F_AA, data = datos.TH.M2)
##
## $`F_riego:F_AA`
## diff lwr upr p adj
## 1:0 ppm-0.5:0 ppm 3.855 -0.1118403 7.82184028 0.0585697
## 0.5:1000 ppm-0.5:0 ppm 0.890 -3.0768403 4.85684028 0.9704300
## 1:1000 ppm-0.5:0 ppm NA NA NA NA
## 0.5:600 ppm-0.5:0 ppm -0.210 -4.1768403 3.75684028 0.9999678
## 1:600 ppm-0.5:0 ppm NA NA NA NA
## 0.5:1000 ppm-1:0 ppm -2.965 -6.9318403 1.00184028 0.1953487
## 1:1000 ppm-1:0 ppm NA NA NA NA
## 0.5:600 ppm-1:0 ppm -4.065 -8.0318403 -0.09815972 0.0434919
## 1:600 ppm-1:0 ppm NA NA NA NA
## 1:1000 ppm-0.5:1000 ppm NA NA NA NA
## 0.5:600 ppm-0.5:1000 ppm -1.100 -5.0668403 2.86684028 0.9305751
## 1:600 ppm-0.5:1000 ppm NA NA NA NA
## 0.5:600 ppm-1:1000 ppm NA NA NA NA
## 1:600 ppm-1:1000 ppm NA NA NA NA
## 1:600 ppm-0.5:600 ppm NA NA NA NA- El tratamiento azul (50% de riego y 600 ppm de ácido ascórbico) tiene diferencia signifcativa con el control para el muestreo 2. Adicionalmente, otra relacón que casi representa una diferencia estadística fue entre el tratamiento amarillo (50% de riego y 0 ppm de ácido ascórbico) y el control, sin embargo, a nivel agronómico, si lo podriamos tener en cuenta.
# Anova - Temperatura de la hoja - Muestreo 2 - Comparación entre los tratamientos
A.TH.M2.CT <- aov(T_Hoja ~ Tratamiento, data=datos.TH.M2)
summary(A.TH.M2.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 42.22 14.074 5.046 0.0173 *
## Residuals 12 33.47 2.789
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.TH.M2.CT <- TukeyHSD(A.TH.M2.CT)
Tukey_A.TH.M2.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = T_Hoja ~ Tratamiento, data = datos.TH.M2)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 0.890 -2.6162322 4.396232 0.8734229
## 50 + AS 600ppm-50 - AS -0.210 -3.7162322 3.296232 0.9978888
## Control-50 - AS 3.855 0.3487678 7.361232 0.0299176
## 50 + AS 600ppm-50 + AS 1000ppm -1.100 -4.6062322 2.406232 0.7889541
## Control-50 + AS 1000ppm 2.965 -0.5412322 6.471232 0.1085383
## Control-50 + AS 600ppm 4.065 0.5587678 7.571232 0.0219123- En este caso, podemos sustentar lo mencionado anteriormente, debido a que los tratamientos “50 - AS” y “50 + AS 600ppm” tienen diferencias significativas con el control en el muestreo 2.
A.TH.M2.DT <- duncan.test(A.TH.M2.CT, 'Tratamiento', console = T)##
## Study: A.TH.M2.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for T_Hoja
##
## Mean Square Error: 2.789458
##
## Tratamiento, means
##
## T_Hoja std r se Min Max Q25 Q50 Q75
## 50 - AS 14.995 1.317915 4 0.8350836 13.06 15.86 14.710 15.530 15.815
## 50 + AS 1000ppm 15.885 1.773838 4 0.8350836 13.62 17.90 15.165 16.010 16.730
## 50 + AS 600ppm 14.785 2.206740 4 0.8350836 11.66 16.40 14.015 15.540 16.310
## Control 18.850 1.185214 4 0.8350836 17.19 20.00 18.585 19.105 19.370
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 2.573149 2.693348 2.766175
##
## Means with the same letter are not significantly different.
##
## T_Hoja groups
## Control 18.850 a
## 50 + AS 1000ppm 15.885 b
## 50 - AS 14.995 b
## 50 + AS 600ppm 14.785 bA.TH.M2.DT## $statistics
## MSerror Df Mean CV
## 2.789458 12 16.12875 10.35522
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 2.573149
## 3 3.225244 2.693348
## 4 3.312453 2.766175
##
## $means
## T_Hoja std r se Min Max Q25 Q50 Q75
## 50 - AS 14.995 1.317915 4 0.8350836 13.06 15.86 14.710 15.530 15.815
## 50 + AS 1000ppm 15.885 1.773838 4 0.8350836 13.62 17.90 15.165 16.010 16.730
## 50 + AS 600ppm 14.785 2.206740 4 0.8350836 11.66 16.40 14.015 15.540 16.310
## Control 18.850 1.185214 4 0.8350836 17.19 20.00 18.585 19.105 19.370
##
## $comparison
## NULL
##
## $groups
## T_Hoja groups
## Control 18.850 a
## 50 + AS 1000ppm 15.885 b
## 50 - AS 14.995 b
## 50 + AS 600ppm 14.785 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 3
# Datos - Temperatura de la hoja - Muestreo 3
datos.TH.M3 <- data.frame(Var_M3$`Factor riego`, Var_M3$`Factor ácido ascórbico`, Var_M3$Tratamiento, Var_M3$`Temperatura Hoja (°C)`)
colnames(datos.TH.M3) <- c("F_riego", "F_AA", "Tratamiento", "T_Hoja")
datos.TH.M3datos.TH.M3$F_riego <- as.factor(datos.TH.M3$F_riego)
datos.TH.M3$F_AA <- as.factor(datos.TH.M3$F_AA)
str(datos.TH.M3)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ T_Hoja : num 16.8 18.9 18.9 14 23 ...# Anova - Temperatura de la hoja - Muestreo 3
A.TH.M3 <- aov(T_Hoja ~ F_riego*F_AA, data = datos.TH.M3)
summary(A.TH.M3)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 52.96 52.96 20.221 0.000731 ***
## F_AA 2 5.46 2.73 1.041 0.382767
## Residuals 12 31.43 2.62
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- El riego, como factor independiente, tiene diferencias signfiicativas, por lo que los resultados de la variable respuesta es estadísticamente atribuible a los tipos de riego. Por parte de la concentración de ácido ascórbico, no hay diferencias significativas, por lo que este aun no representa alguna alteración en la variable respuesta.
A.TH.M3.I <- aov(T_Hoja ~ F_riego:F_AA, data = datos.TH.M3)
summary(A.TH.M3.I)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego:F_AA 3 58.42 19.473 7.435 0.00449 **
## Residuals 12 31.43 2.619
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Adicionalmente, hay una interacción del 99% de significancia entre los dos factores, lo que significa que hay una combinación de riego y ácido ascórbico representativa para el muestreo 3.
# Prueba de normalidad de residuos
shapiro.test(A.TH.M3$residuals)##
## Shapiro-Wilk normality test
##
## data: A.TH.M3$residuals
## W = 0.94288, p-value = 0.3858- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una organización normal.
TukeyHSD(A.TH.M3.I)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = T_Hoja ~ F_riego:F_AA, data = datos.TH.M3)
##
## $`F_riego:F_AA`
## diff lwr upr p adj
## 1:0 ppm-0.5:0 ppm -5.1550 -8.9988645 -1.311136 0.0072805
## 0.5:1000 ppm-0.5:0 ppm -1.4125 -5.2563645 2.431364 0.8126219
## 1:1000 ppm-0.5:0 ppm NA NA NA NA
## 0.5:600 ppm-0.5:0 ppm -1.4475 -5.2913645 2.396364 0.7976436
## 1:600 ppm-0.5:0 ppm NA NA NA NA
## 0.5:1000 ppm-1:0 ppm 3.7425 -0.1013645 7.586364 0.0579750
## 1:1000 ppm-1:0 ppm NA NA NA NA
## 0.5:600 ppm-1:0 ppm 3.7075 -0.1363645 7.551364 0.0610036
## 1:600 ppm-1:0 ppm NA NA NA NA
## 1:1000 ppm-0.5:1000 ppm NA NA NA NA
## 0.5:600 ppm-0.5:1000 ppm -0.0350 -3.8788645 3.808864 1.0000000
## 1:600 ppm-0.5:1000 ppm NA NA NA NA
## 0.5:600 ppm-1:1000 ppm NA NA NA NA
## 1:600 ppm-1:1000 ppm NA NA NA NA
## 1:600 ppm-0.5:600 ppm NA NA NA NA- El tratamiento amarillo (50% y 0ppm de ácido ascórbico) tiene diferencia signifcativa con el control para el muestreo 3, esto es lo que se esperaría puesto a que la deficiencia de agua en la planta causa el cierre de los estomas, por lo que no hay regulación de temperatura y por ende, debe ser alta en aquel tratamiento donde no se aplique ácido ascórbico. Adicionalmente, otras relaciones que casi representan una diferencia estadísticamente significativa fue entre el tratamiento verde (50% y 1000ppm de ácido ascórbico) y el azul (50% de riego y 600 ppm de ácido ascórbico) con el control, no estuvieron por debajo del 5%, sin embargo, a nivel agronómico, si los podriamos tener en cuenta, ya que es lógico pensar que por el ácido ascórbico, estos tratamientos lograron tener una temperatura cercana a las plantas que no se encuentran estresadas.
# Anova - Temperatura de la hoja - Muestreo 3 - Comparación entre los tratamientos
A.TH.M3.CT <- aov(T_Hoja ~ Tratamiento, data=datos.TH.M3)
summary(A.TH.M3.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 58.42 19.473 7.435 0.00449 **
## Residuals 12 31.43 2.619
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.TH.M3.CT <- TukeyHSD(A.TH.M3.CT)
Tukey_A.TH.M3.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = T_Hoja ~ Tratamiento, data = datos.TH.M3)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -1.4125 -4.810036 1.9850357 0.6183228
## 50 + AS 600ppm-50 - AS -1.4475 -4.845036 1.9500357 0.6004793
## Control-50 - AS -5.1550 -8.552536 -1.7574643 0.0034728
## 50 + AS 600ppm-50 + AS 1000ppm -0.0350 -3.432536 3.3625357 0.9999891
## Control-50 + AS 1000ppm -3.7425 -7.140036 -0.3449643 0.0295990
## Control-50 + AS 600ppm -3.7075 -7.105036 -0.3099643 0.0312245- En este caso, podemos comprobar lo anterioremente mencionado. Aquí los tres tratamientos tienen diferencias significativas con respecto a su control.
A.TH.M3.DT <- duncan.test(A.TH.M3.CT, 'Tratamiento', console = T)##
## Study: A.TH.M3.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for T_Hoja
##
## Mean Square Error: 2.619187
##
## Tratamiento, means
##
## T_Hoja std r se Min Max Q25 Q50 Q75
## 50 - AS 22.2925 1.139163 4 0.8091952 20.62 23.02 22.045 22.765 23.0125
## 50 + AS 1000ppm 20.8800 1.325343 4 0.8091952 19.70 22.72 20.060 20.550 21.3700
## 50 + AS 600ppm 20.8450 1.477419 4 0.8091952 18.74 22.20 20.600 21.220 21.4650
## Control 17.1375 2.289052 4 0.8091952 14.02 18.86 16.120 17.835 18.8525
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 2.493379 2.609852 2.680421
##
## Means with the same letter are not significantly different.
##
## T_Hoja groups
## 50 - AS 22.2925 a
## 50 + AS 1000ppm 20.8800 a
## 50 + AS 600ppm 20.8450 a
## Control 17.1375 bA.TH.M3.DT## $statistics
## MSerror Df Mean CV
## 2.619187 12 20.28875 7.976787
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 2.493379
## 3 3.225244 2.609852
## 4 3.312453 2.680421
##
## $means
## T_Hoja std r se Min Max Q25 Q50 Q75
## 50 - AS 22.2925 1.139163 4 0.8091952 20.62 23.02 22.045 22.765 23.0125
## 50 + AS 1000ppm 20.8800 1.325343 4 0.8091952 19.70 22.72 20.060 20.550 21.3700
## 50 + AS 600ppm 20.8450 1.477419 4 0.8091952 18.74 22.20 20.600 21.220 21.4650
## Control 17.1375 2.289052 4 0.8091952 14.02 18.86 16.120 17.835 18.8525
##
## $comparison
## NULL
##
## $groups
## T_Hoja groups
## 50 - AS 22.2925 a
## 50 + AS 1000ppm 20.8800 a
## 50 + AS 600ppm 20.8450 a
## Control 17.1375 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 4
# Datos - Temperatura de la hoja - Muestreo 4
datos.TH.M4 <- data.frame(Var_M4$`Factor riego`, Var_M4$`Factor ácido ascórbico`, Var_M4$Tratamiento, Var_M4$`Temperatura Hoja (°C)`)
colnames(datos.TH.M4) <- c("F_riego", "F_AA", "Tratamiento", "T_Hoja")
datos.TH.M4datos.TH.M4$F_riego <- as.factor(datos.TH.M4$F_riego)
datos.TH.M4$F_AA <- as.factor(datos.TH.M4$F_AA)
str(datos.TH.M4)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ T_Hoja : num 25.2 22.7 21.2 31.5 27.2 ...# Anova - Temperatura de la hoja - Muestreo 4
A.TH.M4 <- aov(T_Hoja ~ F_riego*F_AA, data = datos.TH.M4)
summary(A.TH.M4)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 0.65 0.65 0.075 0.7884
## F_AA 2 64.18 32.09 3.741 0.0546 .
## Residuals 12 102.94 8.58
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- La concentración de ácido ascórbico (AA), como factor independiente, tiene diferencias signfiicativas, por lo que los resultados de la variable respuesta es estadísticamente atribuible a los tipos de concentración de AA. Por parte, el tipo de riego, esta vez no tuvo diferencias significativas, por lo que este no representó alguna alteración en la variable respuesta para el último muestreo.
A.TH.M4.I <- aov(T_Hoja ~ F_riego:F_AA, data = datos.TH.M4)
summary(A.TH.M4.I)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego:F_AA 3 64.83 21.609 2.519 0.107
## Residuals 12 102.94 8.579- Adicionalmente, hay una interacción del 99% de significancia entre los dos factores, lo que significa que hay una combinación de riego y ácido ascórbico representativa para el muestreo 4.
# Prueba de normalidad de residuos
shapiro.test(A.TH.M4$residuals)##
## Shapiro-Wilk normality test
##
## data: A.TH.M4$residuals
## W = 0.89968, p-value = 0.07943- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.TH.M4.I)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = T_Hoja ~ F_riego:F_AA, data = datos.TH.M4)
##
## $`F_riego:F_AA`
## diff lwr upr p adj
## 1:0 ppm-0.5:0 ppm -3.5000 -10.456501 3.456501 0.5618657
## 0.5:1000 ppm-0.5:0 ppm -5.6075 -12.564001 1.349001 0.1444941
## 1:1000 ppm-0.5:0 ppm NA NA NA NA
## 0.5:600 ppm-0.5:0 ppm -3.5000 -10.456501 3.456501 0.5618657
## 1:600 ppm-0.5:0 ppm NA NA NA NA
## 0.5:1000 ppm-1:0 ppm -2.1075 -9.064001 4.849001 0.9032596
## 1:1000 ppm-1:0 ppm NA NA NA NA
## 0.5:600 ppm-1:0 ppm 0.0000 -6.956501 6.956501 1.0000000
## 1:600 ppm-1:0 ppm NA NA NA NA
## 1:1000 ppm-0.5:1000 ppm NA NA NA NA
## 0.5:600 ppm-0.5:1000 ppm 2.1075 -4.849001 9.064001 0.9032596
## 1:600 ppm-0.5:1000 ppm NA NA NA NA
## 0.5:600 ppm-1:1000 ppm NA NA NA NA
## 1:600 ppm-1:1000 ppm NA NA NA NA
## 1:600 ppm-0.5:600 ppm NA NA NA NA# Anova - Temperatura de la hoja - Muestreo 4 - Comparación entre los tratamientos
A.TH.M4.CT <- aov(T_Hoja ~ Tratamiento, data=datos.TH.M4)
summary(A.TH.M4.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 64.83 21.609 2.519 0.107
## Residuals 12 102.94 8.579Tukey_A.TH.M4.CT <- TukeyHSD(A.TH.M4.CT)
Tukey_A.TH.M4.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = T_Hoja ~ Tratamiento, data = datos.TH.M4)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -5.6075 -11.756249 0.5412493 0.0781497
## 50 + AS 600ppm-50 - AS -3.5000 -9.648749 2.6487493 0.3700332
## Control-50 - AS -3.5000 -9.648749 2.6487493 0.3700332
## 50 + AS 600ppm-50 + AS 1000ppm 2.1075 -4.041249 8.2562493 0.7426440
## Control-50 + AS 1000ppm 2.1075 -4.041249 8.2562493 0.7426440
## Control-50 + AS 600ppm 0.0000 -6.148749 6.1487493 1.0000000- En este caso, podemos observar que los tratamientos no tuvieron diferencias significativas con el control, sin embargo, esto es lo que podemos concluir a nivel estadístico, a nivel agronómico es posible afirmar que hubo una diferencia entre el tratamiento verde (50 + AS 1000ppm) y el control, pues fue el tratamiento que menos temperatura presentó, incluso por debajo del control. De igual manera, el tratamiento azul (50 + AS 600ppm), aunque lejos de tener diferencias significativas con el control, si se puede notar un valor de temperatura media inferior al del control.
A.TH.M4.DT <- duncan.test(A.TH.M4.CT, 'Tratamiento', console = T)##
## Study: A.TH.M4.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for T_Hoja
##
## Mean Square Error: 8.578515
##
## Tratamiento, means
##
## T_Hoja std r se Min Max Q25 Q50 Q75
## 50 - AS 28.6450 2.6547253 4 1.464455 26.68 32.52 27.085 27.69 29.2500
## 50 + AS 1000ppm 23.0375 0.8162669 4 1.464455 22.28 24.09 22.460 22.89 23.4675
## 50 + AS 600ppm 25.1450 2.3942918 4 1.464455 23.38 28.68 23.965 24.26 25.4400
## Control 25.1450 4.5681032 4 1.464455 21.16 31.52 22.300 23.95 26.7950
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 4.512435 4.723224 4.850939
##
## Means with the same letter are not significantly different.
##
## T_Hoja groups
## 50 - AS 28.6450 a
## 50 + AS 600ppm 25.1450 ab
## Control 25.1450 ab
## 50 + AS 1000ppm 23.0375 bA.TH.M4.DT## $statistics
## MSerror Df Mean CV
## 8.578515 12 25.49312 11.48902
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 4.512435
## 3 3.225244 4.723224
## 4 3.312453 4.850939
##
## $means
## T_Hoja std r se Min Max Q25 Q50 Q75
## 50 - AS 28.6450 2.6547253 4 1.464455 26.68 32.52 27.085 27.69 29.2500
## 50 + AS 1000ppm 23.0375 0.8162669 4 1.464455 22.28 24.09 22.460 22.89 23.4675
## 50 + AS 600ppm 25.1450 2.3942918 4 1.464455 23.38 28.68 23.965 24.26 25.4400
## Control 25.1450 4.5681032 4 1.464455 21.16 31.52 22.300 23.95 26.7950
##
## $comparison
## NULL
##
## $groups
## T_Hoja groups
## 50 - AS 28.6450 a
## 50 + AS 600ppm 25.1450 ab
## Control 25.1450 ab
## 50 + AS 1000ppm 23.0375 b
##
## attr(,"class")
## [1] "group"Análisis descriptivo - Temperatura de la hoja
Tratamientos <- c(
rep("Control", 4),
rep("50 - AS", 4),
rep("50 + AS 600ppm", 4),
rep("50 + AS 1000ppm", 4)
)
# Datos - Gráfico - Temperatura de la hoja
Datos.G.TH <- data.frame(factor(Tratamientos, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), Var_M1$`Temperatura Hoja (°C)`, Var_M2$`Temperatura Hoja (°C)`, Var_M3$`Temperatura Hoja (°C)`, Var_M4$`Temperatura Hoja (°C)` )
colnames(Datos.G.TH) <- c("Tratamientos", "31", "38","45", "52")
Datos.G.TH# Unificar la variable respuesta en una sola columna
df.long.TH = gather(Datos.G.TH, dds, Valor_de_temperatura, 2:5)
df.long.TH# Diferencias significativas con el control
#M1 #M2 #M3 #M4
DS.TH <- c("a", "a", "b", "ab", #T1
"a", "b*", "a*", "a", #T2
"a", "b*", "a*", "ab", #T3
"a", "b", "a*", "b*") #T4
# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.TH = group_by(df.long.TH, Tratamientos, dds, ) %>% summarise(mean = mean(Valor_de_temperatura), sd = sd(Valor_de_temperatura))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.TH.ds <- data.frame(df.sumzd.TH, DS.TH)
df.sumzd.TH.dsG.TH = ggplot(df.sumzd.TH, aes(x=dds, y=mean, fill=Tratamientos))+
geom_bar(stat = "identity", position = "dodge", color = "black", alpha = 0.5)+
geom_errorbar(aes(ymin=mean-sd, ymax=mean+sd), width = 0.25, size=0.5, position = position_dodge(0.9), alpha = 0.5)+
scale_fill_manual(values=c("red", "yellow", "blue", "green")) +
theme_minimal()+
ggtitle("Temperatura promedio de las hojas \nen los diferentes muestreos") + theme(plot.title = element_text(family = "Times New Roman",
size = rel(1.5),
vjust = 0.5,
hjust = 0.5))+
labs(x = "Días después de siembra", y = "°C")+
theme(axis.title.x = element_text(family = "Times New Roman"), axis.title.y = element_text(family = "Times New Roman"))+
theme(legend.title = element_text(family = "Times New Roman"))+
geom_text(aes(x=dds, y= mean+6, label=DS.TH),
position = position_dodge(width = 1), size = 5)## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.G.TH
Contenido relativo de clorofilas
Análisis estadístico - Muestreo 1
# Datos - Contenido de clorofilas - Muestreo 1
datos.CC.M1 <- data.frame(Var_M1$`Factor riego`, Var_M1$`Factor ácido ascórbico`, Var_M1$Tratamiento, Var_M1$SPAD)
colnames(datos.CC.M1) <- c("F_riego", "F_AA", "Tratamiento", "SPAD")
datos.CC.M1datos.CC.M1$F_riego <- as.factor(datos.CC.M1$F_riego)
datos.CC.M1$F_AA <- as.factor(datos.CC.M1$F_AA)
str(datos.CC.M1)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ SPAD : num 44.2 42.3 31.2 38 37.7 44 45.1 45.4 47 43.8 ...# Anova - Contenido de clorofilas - Muestreo 1
A.CC.M1 <- aov(SPAD ~ F_riego*F_AA, data = datos.CC.M1)
summary(A.CC.M1)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 80.08 80.08 4.766 0.0496 *
## F_AA 2 8.22 4.11 0.245 0.7868
## Residuals 12 201.63 16.80
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvo diferencias con una significancia del 95%.
# Prueba de normalidad de residuos
shapiro.test(A.CC.M1$residuals)##
## Shapiro-Wilk normality test
##
## data: A.CC.M1$residuals
## W = 0.98176, p-value = 0.976- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.CC.M1)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = SPAD ~ F_riego * F_AA, data = datos.CC.M1)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 -5.166667 -10.32313 -0.01020694 0.0496153
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.5791667 -6.117699 7.276033 0.9711501
## 600 ppm-0 ppm 1.5041667 -5.192699 8.201033 0.8231596
## 600 ppm-1000 ppm 0.9250000 -6.807875 8.657875 0.9456585- Los dos riegos presentan diferencias significativas.
# Anova - Contenido de clorofilas - Muestreo 1 - Comparación entre los tratamientos
A.CC.M1.CT <- aov(SPAD ~ Tratamiento, data=datos.CC.M1)
summary(A.CC.M1.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 88.31 29.43 1.752 0.21
## Residuals 12 201.63 16.80Tukey_A.CC.M1.CT <- TukeyHSD(A.CC.M1.CT)
Tukey_A.CC.M1.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = SPAD ~ Tratamiento, data = datos.CC.M1)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 1.100 -7.50544 9.70544 0.9805139
## 50 + AS 600ppm-50 - AS 2.025 -6.58044 10.63044 0.8956467
## Control-50 - AS -4.125 -12.73044 4.48044 0.5095905
## 50 + AS 600ppm-50 + AS 1000ppm 0.925 -7.68044 9.53044 0.9881866
## Control-50 + AS 1000ppm -5.225 -13.83044 3.38044 0.3186913
## Control-50 + AS 600ppm -6.150 -14.75544 2.45544 0.2011276- Podemos observar que ninguno de los tratamientos tiene diferencias significativas con el control para el muestreo 1.
A.CC.M1.DT <- duncan.test(A.CC.M1.CT, 'Tratamiento', console = T)##
## Study: A.CC.M1.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for SPAD
##
## Mean Square Error: 16.80292
##
## Tratamiento, means
##
## SPAD std r se Min Max Q25 Q50 Q75
## 50 - AS 43.050 3.617089 4 2.049568 37.7 45.4 42.425 44.55 45.175
## 50 + AS 1000ppm 44.150 4.362339 4 2.049568 41.1 50.6 41.775 42.45 44.825
## 50 + AS 600ppm 45.075 1.359841 4 2.049568 43.8 47.0 44.475 44.75 45.350
## Control 38.925 5.766209 4 2.049568 31.2 44.2 36.300 40.15 42.775
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 6.315348 6.610356 6.789098
##
## Means with the same letter are not significantly different.
##
## SPAD groups
## 50 + AS 600ppm 45.075 a
## 50 + AS 1000ppm 44.150 a
## 50 - AS 43.050 a
## Control 38.925 aA.CC.M1.DT## $statistics
## MSerror Df Mean CV
## 16.80292 12 42.8 9.577421
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 6.315348
## 3 3.225244 6.610356
## 4 3.312453 6.789098
##
## $means
## SPAD std r se Min Max Q25 Q50 Q75
## 50 - AS 43.050 3.617089 4 2.049568 37.7 45.4 42.425 44.55 45.175
## 50 + AS 1000ppm 44.150 4.362339 4 2.049568 41.1 50.6 41.775 42.45 44.825
## 50 + AS 600ppm 45.075 1.359841 4 2.049568 43.8 47.0 44.475 44.75 45.350
## Control 38.925 5.766209 4 2.049568 31.2 44.2 36.300 40.15 42.775
##
## $comparison
## NULL
##
## $groups
## SPAD groups
## 50 + AS 600ppm 45.075 a
## 50 + AS 1000ppm 44.150 a
## 50 - AS 43.050 a
## Control 38.925 a
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 2
# Datos - Contenido de clorofilas - Muestreo 2
datos.CC.M2 <- data.frame(Var_M2$`Factor riego`, Var_M2$`Factor ácido ascórbico`, Var_M2$Tratamiento, Var_M2$SPAD)
colnames(datos.CC.M2) <- c("F_riego", "F_AA", "Tratamiento", "SPAD")
datos.CC.M2datos.CC.M2$F_riego <- as.factor(datos.CC.M2$F_riego)
datos.CC.M2$F_AA <- as.factor(datos.CC.M2$F_AA)
str(datos.CC.M2)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ SPAD : num 45.2 45.2 44.2 43.3 49.8 47.9 46.2 39.4 42 45.5 ...# Anova - Contenido de clorofilas - Muestreo 2
A.CC.M2 <- aov(SPAD ~ F_riego*F_AA, data = datos.CC.M2)
summary(A.CC.M2)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 2.90 2.901 0.276 0.609
## F_AA 2 4.19 2.093 0.200 0.822
## Residuals 12 125.91 10.493- Ninguno de los factores tuvo diferencias significativas.
# Prueba de normalidad de residuos
shapiro.test(A.CC.M2$residuals)##
## Shapiro-Wilk normality test
##
## data: A.CC.M2$residuals
## W = 0.94754, p-value = 0.4518- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.CC.M2)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = SPAD ~ F_riego * F_AA, data = datos.CC.M2)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 -0.9833333 -5.058063 3.091396 0.6086059
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -1.0166667 -6.308653 4.27532 0.8667365
## 600 ppm-0 ppm 0.2833333 -5.008653 5.57532 0.9888255
## 600 ppm-1000 ppm 1.3000000 -4.810660 7.41066 0.8395476- No se presentan diferencias significativas entre los factores.
# Anova - Contenido de clorofilas - Muestreo 2 - Comparación entre los tratamientos
A.CC.M2.CT <- aov(SPAD ~ Tratamiento, data=datos.CC.M2)
summary(A.CC.M2.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 7.09 2.362 0.225 0.877
## Residuals 12 125.91 10.492Tukey_A.CC.M2.CT <- TukeyHSD(A.CC.M2.CT)
Tukey_A.CC.M2.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = SPAD ~ Tratamiento, data = datos.CC.M2)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -1.20 -8.000177 5.600177 0.9516671
## 50 + AS 600ppm-50 - AS 0.10 -6.700177 6.900177 0.9999683
## Control-50 - AS -1.35 -8.150177 5.450177 0.9333748
## 50 + AS 600ppm-50 + AS 1000ppm 1.30 -5.500177 8.100177 0.9398385
## Control-50 + AS 1000ppm -0.15 -6.950177 6.650177 0.9998932
## Control-50 + AS 600ppm -1.45 -8.250177 5.350177 0.9193587- Podemos observar que ninguno de los tratamientos tiene diferencias significativas con el control para el muestreo 2.
A.CC.M2.DT <- duncan.test(A.CC.M2.CT, 'Tratamiento', console = T)##
## Study: A.CC.M2.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for SPAD
##
## Mean Square Error: 10.4925
##
## Tratamiento, means
##
## SPAD std r se Min Max Q25 Q50 Q75
## 50 - AS 45.825 4.5287047 4 1.619606 39.4 49.8 44.500 47.05 48.375
## 50 + AS 1000ppm 44.625 2.9044506 4 1.619606 41.5 48.5 43.300 44.25 45.575
## 50 + AS 600ppm 45.925 3.4912987 4 1.619606 42.0 50.5 44.625 45.60 46.900
## Control 44.475 0.9142392 4 1.619606 43.3 45.2 43.975 44.70 45.200
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 4.990504 5.223625 5.364870
##
## Means with the same letter are not significantly different.
##
## SPAD groups
## 50 + AS 600ppm 45.925 a
## 50 - AS 45.825 a
## 50 + AS 1000ppm 44.625 a
## Control 44.475 aA.CC.M2.DT## $statistics
## MSerror Df Mean CV
## 10.4925 12 45.2125 7.164419
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 4.990504
## 3 3.225244 5.223625
## 4 3.312453 5.364870
##
## $means
## SPAD std r se Min Max Q25 Q50 Q75
## 50 - AS 45.825 4.5287047 4 1.619606 39.4 49.8 44.500 47.05 48.375
## 50 + AS 1000ppm 44.625 2.9044506 4 1.619606 41.5 48.5 43.300 44.25 45.575
## 50 + AS 600ppm 45.925 3.4912987 4 1.619606 42.0 50.5 44.625 45.60 46.900
## Control 44.475 0.9142392 4 1.619606 43.3 45.2 43.975 44.70 45.200
##
## $comparison
## NULL
##
## $groups
## SPAD groups
## 50 + AS 600ppm 45.925 a
## 50 - AS 45.825 a
## 50 + AS 1000ppm 44.625 a
## Control 44.475 a
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 3
# Datos - Contenido de clorofilas - Muestreo 3
datos.CC.M3 <- data.frame(Var_M3$`Factor riego`, Var_M3$`Factor ácido ascórbico`, Var_M3$Tratamiento, Var_M3$SPAD)
colnames(datos.CC.M3) <- c("F_riego", "F_AA", "Tratamiento", "SPAD")
datos.CC.M3datos.CC.M3$F_riego <- as.factor(datos.CC.M3$F_riego)
datos.CC.M3$F_AA <- as.factor(datos.CC.M3$F_AA)
str(datos.CC.M3)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ SPAD : num 47.8 45.4 39.8 38.6 50.6 43.7 48.9 55.5 46.3 48 ...# Anova - Contenido de clorofilas - Muestreo 3
A.CC.M3 <- aov(SPAD ~ F_riego*F_AA, data = datos.CC.M3)
summary(A.CC.M3)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 108.90 108.90 6.885 0.0222 *
## F_AA 2 84.02 42.01 2.656 0.1109
## Residuals 12 189.80 15.82
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvo diferencias con una significancia del 95%.
# Prueba de normalidad de residuos
shapiro.test(A.CC.M3$residuals)##
## Shapiro-Wilk normality test
##
## data: A.CC.M3$residuals
## W = 0.96228, p-value = 0.7034- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.CC.M3)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = SPAD ~ F_riego * F_AA, data = datos.CC.M3)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 -6.025 -11.02787 -1.022125 0.0222219
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 2.425 -4.07240 8.922400 0.5933167
## 600 ppm-0 ppm -3.925 -10.42240 2.572400 0.2784760
## 600 ppm-1000 ppm -6.350 -13.85255 1.152551 0.1012683- Los dos riegos presentan diferencias significativas.
# Anova - Contenido de clorofilas - Muestreo 3 - Comparación entre los tratamientos
A.CC.M3.CT <- aov(SPAD ~ Tratamiento, data=datos.CC.M3)
summary(A.CC.M3.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 192.9 64.31 4.066 0.033 *
## Residuals 12 189.8 15.82
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.CC.M3.CT <- TukeyHSD(A.CC.M3.CT)
Tukey_A.CC.M3.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = SPAD ~ Tratamiento, data = datos.CC.M3)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 2.050 -6.299127 10.3991272 0.8836377
## 50 + AS 600ppm-50 - AS -4.300 -12.649127 4.0491272 0.4515449
## Control-50 - AS -6.775 -15.124127 1.5741272 0.1280824
## 50 + AS 600ppm-50 + AS 1000ppm -6.350 -14.699127 1.9991272 0.1629764
## Control-50 + AS 1000ppm -8.825 -17.174127 -0.4758728 0.0372824
## Control-50 + AS 600ppm -2.475 -10.824127 5.8741272 0.8150759- Podemos observar que solo el tratamiento verde (50 + AS 1000ppm) tuvo diferencias significativas con su control.
A.CC.M3.DT <- duncan.test(A.CC.M3.CT, 'Tratamiento', console = T)##
## Study: A.CC.M3.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for SPAD
##
## Mean Square Error: 15.81687
##
## Tratamiento, means
##
## SPAD std r se Min Max Q25 Q50 Q75
## 50 - AS 49.675 4.867837 4 1.988522 43.7 55.5 47.600 49.75 51.825
## 50 + AS 1000ppm 51.725 2.056494 4 1.988522 50.3 54.7 50.375 50.95 52.300
## 50 + AS 600ppm 45.375 3.986122 4 1.988522 39.5 48.0 44.600 47.00 47.775
## Control 42.900 4.410593 4 1.988522 38.6 47.8 39.500 42.60 46.000
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 6.127245 6.413467 6.586885
##
## Means with the same letter are not significantly different.
##
## SPAD groups
## 50 + AS 1000ppm 51.725 a
## 50 - AS 49.675 a
## 50 + AS 600ppm 45.375 ab
## Control 42.900 bA.CC.M3.DT## $statistics
## MSerror Df Mean CV
## 15.81687 12 47.41875 8.387069
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 6.127245
## 3 3.225244 6.413467
## 4 3.312453 6.586885
##
## $means
## SPAD std r se Min Max Q25 Q50 Q75
## 50 - AS 49.675 4.867837 4 1.988522 43.7 55.5 47.600 49.75 51.825
## 50 + AS 1000ppm 51.725 2.056494 4 1.988522 50.3 54.7 50.375 50.95 52.300
## 50 + AS 600ppm 45.375 3.986122 4 1.988522 39.5 48.0 44.600 47.00 47.775
## Control 42.900 4.410593 4 1.988522 38.6 47.8 39.500 42.60 46.000
##
## $comparison
## NULL
##
## $groups
## SPAD groups
## 50 + AS 1000ppm 51.725 a
## 50 - AS 49.675 a
## 50 + AS 600ppm 45.375 ab
## Control 42.900 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 4
# Datos - Contenido de clorofilas - Muestreo 4
datos.CC.M4 <- data.frame(Var_M4$`Factor riego`, Var_M4$`Factor ácido ascórbico`, Var_M4$Tratamiento, Var_M4$SPAD)
colnames(datos.CC.M4) <- c("F_riego", "F_AA", "Tratamiento", "SPAD")
datos.CC.M4datos.CC.M4$F_riego <- as.factor(datos.CC.M4$F_riego)
datos.CC.M4$F_AA <- as.factor(datos.CC.M4$F_AA)
str(datos.CC.M4)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ SPAD : num 39.7 46.8 38.9 45.4 49.6 54.5 48.5 56.2 50.9 58.7 ...# Anova - Contenido de clorofilas - Muestreo 4
A.CC.M4 <- aov(SPAD ~ F_riego*F_AA, data = datos.CC.M4)
summary(A.CC.M4)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 325.5 325.5 14.733 0.00236 **
## F_AA 2 28.8 14.4 0.653 0.53812
## Residuals 12 265.1 22.1
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvo diferencias con una significancia del 99%.
# Prueba de normalidad de residuos
shapiro.test(A.CC.M4$residuals)##
## Shapiro-Wilk normality test
##
## data: A.CC.M4$residuals
## W = 0.95218, p-value = 0.5251- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.CC.M4)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = SPAD ~ F_riego * F_AA, data = datos.CC.M4)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 -10.41667 -16.32953 -4.5038 0.0023593
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 2.6416667 -5.03757 10.320904 0.6398957
## 600 ppm-0 ppm -0.8083333 -8.48757 6.870904 0.9576104
## 600 ppm-1000 ppm -3.4500000 -12.31722 5.417219 0.5682824- Los dos riegos presentan diferencias significativas.
# Anova - Contenido de clorofilas - Muestreo 4 - Comparación entre los tratamientos
A.CC.M4.CT <- aov(SPAD ~ Tratamiento, data=datos.CC.M4)
summary(A.CC.M4.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 354.4 118.12 5.346 0.0143 *
## Residuals 12 265.1 22.09
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.CC.M4.CT <- TukeyHSD(A.CC.M4.CT)
Tukey_A.CC.M4.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = SPAD ~ Tratamiento, data = datos.CC.M4)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 3.10 -6.767782 12.9677821 0.7882944
## 50 + AS 600ppm-50 - AS -0.35 -10.217782 9.5177821 0.9995573
## Control-50 - AS -9.50 -19.367782 0.3677821 0.0604803
## 50 + AS 600ppm-50 + AS 1000ppm -3.45 -13.317782 6.4177821 0.7313168
## Control-50 + AS 1000ppm -12.60 -22.467782 -2.7322179 0.0118922
## Control-50 + AS 600ppm -9.15 -19.017782 0.7177821 0.0723725- Podemos observar que solo el tratamiento verde (50 + AS 1000ppm), presenta diferencias estadísticamente significativas, sin embargo, a pesar de que los otros dos tratamientos; el amarillo (50 - AS) y el azul (50 + AS 600ppm); se ubican por encima del 5%, los valores de cada uno nos indica que a nivel agronomico los podemos tener en cuenta.
A.CC.M4.DT <- duncan.test(A.CC.M4.CT, 'Tratamiento', console = T)##
## Study: A.CC.M4.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for SPAD
##
## Mean Square Error: 22.09417
##
## Tratamiento, means
##
## SPAD std r se Min Max Q25 Q50 Q75
## 50 - AS 52.20 3.730058 4 2.350222 48.5 56.2 49.325 52.05 54.925
## 50 + AS 1000ppm 55.30 3.995831 4 2.350222 49.6 58.4 54.025 56.60 57.875
## 50 + AS 600ppm 51.85 6.530697 4 2.350222 43.3 58.7 49.000 52.70 55.550
## Control 42.70 3.980787 4 2.350222 38.9 46.8 39.500 42.55 45.750
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 7.241753 7.580037 7.784999
##
## Means with the same letter are not significantly different.
##
## SPAD groups
## 50 + AS 1000ppm 55.30 a
## 50 - AS 52.20 a
## 50 + AS 600ppm 51.85 a
## Control 42.70 bA.CC.M4.DT## $statistics
## MSerror Df Mean CV
## 22.09417 12 50.5125 9.305505
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 7.241753
## 3 3.225244 7.580037
## 4 3.312453 7.784999
##
## $means
## SPAD std r se Min Max Q25 Q50 Q75
## 50 - AS 52.20 3.730058 4 2.350222 48.5 56.2 49.325 52.05 54.925
## 50 + AS 1000ppm 55.30 3.995831 4 2.350222 49.6 58.4 54.025 56.60 57.875
## 50 + AS 600ppm 51.85 6.530697 4 2.350222 43.3 58.7 49.000 52.70 55.550
## Control 42.70 3.980787 4 2.350222 38.9 46.8 39.500 42.55 45.750
##
## $comparison
## NULL
##
## $groups
## SPAD groups
## 50 + AS 1000ppm 55.30 a
## 50 - AS 52.20 a
## 50 + AS 600ppm 51.85 a
## Control 42.70 b
##
## attr(,"class")
## [1] "group"Análisis descriptivo - Contenido de clorofilas
# Datos - Gráfico - Contenido de clorofilas
Datos.G.CC <- data.frame(factor(Tratamientos, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), Var_M1$SPAD, Var_M2$SPAD, Var_M3$SPAD, Var_M4$SPAD)
colnames(Datos.G.CC) <- c("Tratamientos", "31", "38","45", "52")
Datos.G.CC# Unificar la variable respuesta en una sola columna
df.long.CC = gather(Datos.G.CC, dds, Contenido_de_clorofilas, 2:5)
df.long.CC# Diferencias significativas con el control
# M1 M2 M3 M4
DS.CC <- c("a", "a", "b", "b", #T1
"a", "a", "a*", "a*", #T2
"a", "a", "ab", "a*", #T3
"a", "a", "a*", "a*") #T4
# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.CC = group_by(df.long.CC, Tratamientos, dds, ) %>% summarise(mean = mean(Contenido_de_clorofilas), sd = sd(Contenido_de_clorofilas))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.CC.ds <- data.frame(df.sumzd.CC, DS.CC)
df.sumzd.CC.dsG.CC = ggplot(df.sumzd.CC, aes(x=dds, y=mean, fill=Tratamientos))+
geom_bar(stat = "identity", position = "dodge", color = "black", alpha = 0.5)+
geom_errorbar(aes(ymin=mean-sd, ymax=mean+sd), width = 0.25, size=0.5, position = position_dodge(0.9), alpha = 0.5)+
scale_fill_manual(values=c("red", "yellow", "blue", "green")) +
theme_minimal()+
ggtitle("Contenido relativo de clorofilas \nen los diferentes muestreos") + theme(plot.title = element_text(family = "Times New Roman",
size = rel(2),
vjust = 0.5,
hjust = 0.5))+
theme(axis.text.x = element_text(size=15),
axis.text.y = element_text(size=15))+
labs(x = "Días después de siembra", y = "mg de clorofila / g de tejido vegetal")+
theme(axis.title.x = element_text(family = "Times New Roman", size = rel(1.5)), axis.title.y = element_text(family = "Times New Roman", size = rel(1.5)))+
theme(legend.title = element_text(family = "Times New Roman"))+
geom_text(aes(x=dds, y= mean+1.5, label=DS.CC),
position = position_dodge(width = 1), size = 5)+
coord_cartesian(ylim = c(35, 60))
G.CC
Altura de la parte aérea
Análisis estadístico - Muestreo 1
# Datos - Altura de la parte aérea - Muestreo 1
datos.APA.M1 <- data.frame(Var_M1$`Factor riego`, Var_M1$`Factor ácido ascórbico`, Var_M1$Tratamiento, Var_M1$`Altura Aérea (cm)`)
colnames(datos.APA.M1) <- c("F_riego", "F_AA", "Tratamiento", "Altura_aérea")
datos.APA.M1datos.APA.M1$F_riego <- as.factor(datos.APA.M1$F_riego)
datos.APA.M1$F_AA <- as.factor(datos.APA.M1$F_AA)
str(datos.APA.M1)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento : chr "Control" "Control" "Control" "Control" ...
## $ Altura_aérea: num 12 15.8 11.1 11.9 10.8 9.9 9.1 13 10.8 11.5 ...# Anova - Altura de la parte aérea - Muestreo 1
A.APA.M1 <- aov(Altura_aérea ~ F_riego*F_AA, data = datos.APA.M1)
summary(A.APA.M1)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 15.19 15.187 6.959 0.0217 *
## F_AA 2 4.02 2.010 0.921 0.4245
## Residuals 12 26.19 2.183
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvo diferencias con una significancia del 95%.
# Prueba de normalidad de residuos
shapiro.test(A.APA.M1$residuals)##
## Shapiro-Wilk normality test
##
## data: A.APA.M1$residuals
## W = 0.9105, p-value = 0.1185- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.APA.M1)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Altura_aérea ~ F_riego * F_AA, data = datos.APA.M1)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 2.25 0.3916119 4.108388 0.0216557
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -0.925 -3.338550 1.488550 0.5774508
## 600 ppm-0 ppm 0.425 -1.988550 2.838550 0.8865704
## 600 ppm-1000 ppm 1.350 -1.436928 4.136928 0.4257821- Los dos riegos presentan diferencias significativas.
# Anova - Altura de la parte aérea - Muestreo 1 - Comparación entre los tratamientos
A.APA.M1.CT <- aov(Altura_aérea ~ Tratamiento, data=datos.APA.M1)
summary(A.APA.M1.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 19.21 6.402 2.934 0.0767 .
## Residuals 12 26.19 2.183
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.APA.M1.CT <- TukeyHSD(A.APA.M1.CT)
Tukey_A.APA.M1.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Altura_aérea ~ Tratamiento, data = datos.APA.M1)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -1.05 -4.15140057 2.051401 0.7495008
## 50 + AS 600ppm-50 - AS 0.30 -2.80140057 3.401401 0.9913136
## Control-50 - AS 2.00 -1.10140057 5.101401 0.2727674
## 50 + AS 600ppm-50 + AS 1000ppm 1.35 -1.75140057 4.451401 0.5845050
## Control-50 + AS 1000ppm 3.05 -0.05140057 6.151401 0.0544253
## Control-50 + AS 600ppm 1.70 -1.40140057 4.801401 0.4006626- Podemos observar que ninguno de los tratamientos tiene diferencias significativas con el control para el muestreo 1, sin embargo, el tratamiento verde (50 + AS 1000ppm), presentó un valor del 5.4%, por lo que puede ser un valor tenido en cuenta de manera agronómica.
A.APA.M1.DT <- duncan.test(A.APA.M1.CT, 'Tratamiento', console = T)##
## Study: A.APA.M1.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for Altura_aérea
##
## Mean Square Error: 2.1825
##
## Tratamiento, means
##
## Altura_aérea std r se Min Max Q25 Q50
## 50 - AS 10.70 1.6832508 4 0.7386643 9.1 13.0 9.700 10.35
## 50 + AS 1000ppm 9.65 1.0344080 4 0.7386643 8.2 10.6 9.325 9.90
## 50 + AS 600ppm 11.00 0.6271629 4 0.7386643 10.2 11.5 10.650 11.15
## Control 12.70 2.1055482 4 0.7386643 11.1 15.8 11.700 11.95
## Q75
## 50 - AS 11.350
## 50 + AS 1000ppm 10.225
## 50 + AS 600ppm 11.500
## Control 12.950
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 2.276051 2.382372 2.446791
##
## Means with the same letter are not significantly different.
##
## Altura_aérea groups
## Control 12.70 a
## 50 + AS 600ppm 11.00 ab
## 50 - AS 10.70 ab
## 50 + AS 1000ppm 9.65 bA.APA.M1.DT## $statistics
## MSerror Df Mean CV
## 2.1825 12 11.0125 13.41502
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 2.276051
## 3 3.225244 2.382372
## 4 3.312453 2.446791
##
## $means
## Altura_aérea std r se Min Max Q25 Q50
## 50 - AS 10.70 1.6832508 4 0.7386643 9.1 13.0 9.700 10.35
## 50 + AS 1000ppm 9.65 1.0344080 4 0.7386643 8.2 10.6 9.325 9.90
## 50 + AS 600ppm 11.00 0.6271629 4 0.7386643 10.2 11.5 10.650 11.15
## Control 12.70 2.1055482 4 0.7386643 11.1 15.8 11.700 11.95
## Q75
## 50 - AS 11.350
## 50 + AS 1000ppm 10.225
## 50 + AS 600ppm 11.500
## Control 12.950
##
## $comparison
## NULL
##
## $groups
## Altura_aérea groups
## Control 12.70 a
## 50 + AS 600ppm 11.00 ab
## 50 - AS 10.70 ab
## 50 + AS 1000ppm 9.65 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 2
# Datos - Altura de la parte aérea - Muestreo 2
datos.APA.M2 <- data.frame(Var_M2$`Factor riego`, Var_M2$`Factor ácido ascórbico`, Var_M2$Tratamiento, Var_M2$`Altura Aérea (cm)`)
colnames(datos.APA.M2) <- c("F_riego", "F_AA", "Tratamiento", "Altura_aérea")
datos.APA.M2datos.APA.M2$F_riego <- as.factor(datos.APA.M2$F_riego)
datos.APA.M2$F_AA <- as.factor(datos.APA.M2$F_AA)
str(datos.APA.M2)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento : chr "Control" "Control" "Control" "Control" ...
## $ Altura_aérea: num 14.6 13.8 13.9 16.3 11.1 11.6 9.9 9.9 10.8 10.5 ...# Anova - Altura de la parte aérea - Muestreo 2
A.APA.M2 <- aov(Altura_aérea ~ F_riego*F_AA, data = datos.APA.M2)
summary(A.APA.M2)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 43.13 43.13 26.097 0.000258 ***
## F_AA 2 1.95 0.97 0.589 0.570187
## Residuals 12 19.83 1.65
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvo diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.APA.M2$residuals)##
## Shapiro-Wilk normality test
##
## data: A.APA.M2$residuals
## W = 0.93728, p-value = 0.317- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.APA.M2)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Altura_aérea ~ F_riego * F_AA, data = datos.APA.M2)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 3.791667 2.17449 5.408843 0.0002581
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.6833333 -1.416948 2.783614 0.6696715
## 600 ppm-0 ppm -0.2166667 -2.316948 1.883614 0.9592454
## 600 ppm-1000 ppm -0.9000000 -3.325195 1.525195 0.5966848- Los dos riegos presentan diferencias significativas.
# Anova - Altura de la parte aérea - Muestreo 2 - Comparación entre los tratamientos
A.APA.M2.CT <- aov(Altura_aérea ~ Tratamiento, data=datos.APA.M2)
summary(A.APA.M2.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 45.08 15.026 9.092 0.00205 **
## Residuals 12 19.83 1.653
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.APA.M2.CT <- TukeyHSD(A.APA.M2.CT)
Tukey_A.APA.M2.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Altura_aérea ~ Tratamiento, data = datos.APA.M2)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 0.800 -1.8988506 3.498851 0.8150990
## 50 + AS 600ppm-50 - AS -0.100 -2.7988506 2.598851 0.9994955
## Control-50 - AS 4.025 1.3261494 6.723851 0.0039574
## 50 + AS 600ppm-50 + AS 1000ppm -0.900 -3.5988506 1.798851 0.7577373
## Control-50 + AS 1000ppm 3.225 0.5261494 5.923851 0.0182067
## Control-50 + AS 600ppm 4.125 1.4261494 6.823851 0.0032835- Podemos observar que todos los tratamientos tuvieron diferencias significativas con respecto al control.
A.APA.M2.DT <- duncan.test(A.APA.M2.CT, 'Tratamiento', console = T)##
## Study: A.APA.M2.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for Altura_aérea
##
## Mean Square Error: 1.652708
##
## Tratamiento, means
##
## Altura_aérea std r se Min Max Q25 Q50
## 50 - AS 10.625 0.8616844 4 0.6427885 9.9 11.6 9.900 10.50
## 50 + AS 1000ppm 11.425 2.1140404 4 0.6427885 9.2 14.2 10.250 11.15
## 50 + AS 600ppm 10.525 0.2500000 4 0.6427885 10.2 10.8 10.425 10.55
## Control 14.650 1.1561430 4 0.6427885 13.8 16.3 13.875 14.25
## Q75
## 50 - AS 11.225
## 50 + AS 1000ppm 12.325
## 50 + AS 600ppm 10.650
## Control 15.025
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 1.980629 2.073150 2.129207
##
## Means with the same letter are not significantly different.
##
## Altura_aérea groups
## Control 14.650 a
## 50 + AS 1000ppm 11.425 b
## 50 - AS 10.625 b
## 50 + AS 600ppm 10.525 bA.APA.M2.DT## $statistics
## MSerror Df Mean CV
## 1.652708 12 11.80625 10.88895
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 1.980629
## 3 3.225244 2.073150
## 4 3.312453 2.129207
##
## $means
## Altura_aérea std r se Min Max Q25 Q50
## 50 - AS 10.625 0.8616844 4 0.6427885 9.9 11.6 9.900 10.50
## 50 + AS 1000ppm 11.425 2.1140404 4 0.6427885 9.2 14.2 10.250 11.15
## 50 + AS 600ppm 10.525 0.2500000 4 0.6427885 10.2 10.8 10.425 10.55
## Control 14.650 1.1561430 4 0.6427885 13.8 16.3 13.875 14.25
## Q75
## 50 - AS 11.225
## 50 + AS 1000ppm 12.325
## 50 + AS 600ppm 10.650
## Control 15.025
##
## $comparison
## NULL
##
## $groups
## Altura_aérea groups
## Control 14.650 a
## 50 + AS 1000ppm 11.425 b
## 50 - AS 10.625 b
## 50 + AS 600ppm 10.525 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 3
# Datos - Altura de la parte aérea - Muestreo 3
datos.APA.M3 <- data.frame(Var_M3$`Factor riego`, Var_M3$`Factor ácido ascórbico`, Var_M3$Tratamiento, Var_M3$`Altura Aérea (cm)`)
colnames(datos.APA.M3) <- c("F_riego", "F_AA", "Tratamiento", "Altura_aérea")
datos.APA.M3datos.APA.M3$F_riego <- as.factor(datos.APA.M3$F_riego)
datos.APA.M3$F_AA <- as.factor(datos.APA.M3$F_AA)
str(datos.APA.M3)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento : chr "Control" "Control" "Control" "Control" ...
## $ Altura_aérea: num 15 19.5 14.7 17.3 11.1 9.2 7.8 11.9 11.1 9.2 ...# Anova - Altura de la parte aérea - Muestreo 3
A.APA.M3 <- aov(Altura_aérea ~ F_riego*F_AA, data = datos.APA.M3)
summary(A.APA.M3)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 116.25 116.25 33.658 8.45e-05 ***
## F_AA 2 1.68 0.84 0.243 0.788
## Residuals 12 41.45 3.45
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvo diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.APA.M3$residuals)##
## Shapiro-Wilk normality test
##
## data: A.APA.M3$residuals
## W = 0.94284, p-value = 0.3852- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.APA.M3)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Altura_aérea ~ F_riego * F_AA, data = datos.APA.M3)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 6.225 3.887144 8.562856 8.45e-05
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.7 -2.336252 3.736252 0.8147701
## 600 ppm-0 ppm 0.1 -2.936252 3.136252 0.9957540
## 600 ppm-1000 ppm -0.6 -4.105962 2.905962 0.8924502- Los dos riegos presentan diferencias significativas.
# Anova - Altura de la parte aérea - Muestreo 3 - Comparación entre los tratamientos
A.APA.M3.CT <- aov(Altura_aérea ~ Tratamiento, data=datos.APA.M3)
summary(A.APA.M3.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 117.93 39.31 11.38 0.000799 ***
## Residuals 12 41.45 3.45
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.APA.M3.CT <- TukeyHSD(A.APA.M3.CT)
Tukey_A.APA.M3.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Altura_aérea ~ Tratamiento, data = datos.APA.M3)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 0.900 -3.001569 4.801569 0.9008866
## 50 + AS 600ppm-50 - AS 0.300 -3.601569 4.201569 0.9955751
## Control-50 - AS 6.625 2.723431 10.526569 0.0014203
## 50 + AS 600ppm-50 + AS 1000ppm -0.600 -4.501569 3.301569 0.9670479
## Control-50 + AS 1000ppm 5.725 1.823431 9.626569 0.0044693
## Control-50 + AS 600ppm 6.325 2.423431 10.226569 0.0020694- Podemos observar que todos los tratamientos tuvieron diferencias significativas con respecto a su control.
A.APA.M3.DT <- duncan.test(A.APA.M3.CT, 'Tratamiento', console = T)##
## Study: A.APA.M3.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for Altura_aérea
##
## Mean Square Error: 3.453958
##
## Tratamiento, means
##
## Altura_aérea std r se Min Max Q25 Q50 Q75
## 50 - AS 10.000 1.852926 4 0.9292414 7.8 11.9 8.850 10.15 11.300
## 50 + AS 1000ppm 10.900 1.023067 4 0.9292414 10.1 12.4 10.400 10.55 11.050
## 50 + AS 600ppm 10.300 2.076857 4 0.9292414 8.1 12.8 8.925 10.15 11.525
## Control 16.625 2.241093 4 0.9292414 14.7 19.5 14.925 16.15 17.850
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 2.863278 2.997030 3.078068
##
## Means with the same letter are not significantly different.
##
## Altura_aérea groups
## Control 16.625 a
## 50 + AS 1000ppm 10.900 b
## 50 + AS 600ppm 10.300 b
## 50 - AS 10.000 bA.APA.M3.DT## $statistics
## MSerror Df Mean CV
## 3.453958 12 11.95625 15.54403
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 2.863278
## 3 3.225244 2.997030
## 4 3.312453 3.078068
##
## $means
## Altura_aérea std r se Min Max Q25 Q50 Q75
## 50 - AS 10.000 1.852926 4 0.9292414 7.8 11.9 8.850 10.15 11.300
## 50 + AS 1000ppm 10.900 1.023067 4 0.9292414 10.1 12.4 10.400 10.55 11.050
## 50 + AS 600ppm 10.300 2.076857 4 0.9292414 8.1 12.8 8.925 10.15 11.525
## Control 16.625 2.241093 4 0.9292414 14.7 19.5 14.925 16.15 17.850
##
## $comparison
## NULL
##
## $groups
## Altura_aérea groups
## Control 16.625 a
## 50 + AS 1000ppm 10.900 b
## 50 + AS 600ppm 10.300 b
## 50 - AS 10.000 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 4
# Datos - Altura de la parte aérea - Muestreo 4
datos.APA.M4 <- data.frame(Var_M4$`Factor riego`, Var_M4$`Factor ácido ascórbico`, Var_M4$Tratamiento, Var_M4$`Altura Aérea (cm)`)
colnames(datos.APA.M4) <- c("F_riego", "F_AA", "Tratamiento", "Altura_aérea")
datos.APA.M4datos.APA.M4$F_riego <- as.factor(datos.APA.M4$F_riego)
datos.APA.M4$F_AA <- as.factor(datos.APA.M4$F_AA)
str(datos.APA.M4)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento : chr "Control" "Control" "Control" "Control" ...
## $ Altura_aérea: num 17 16.5 16.6 13.8 9.6 10.4 11.4 11.8 11.4 10.2 ...# Anova - Altura de la parte aérea - Muestreo 4
A.APA.M4 <- aov(Altura_aérea ~ F_riego*F_AA, data = datos.APA.M4)
summary(A.APA.M4)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 68.16 68.16 52.283 1.04e-05 ***
## F_AA 2 1.01 0.51 0.388 0.687
## Residuals 12 15.65 1.30
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvo diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.APA.M4$residuals)##
## Shapiro-Wilk normality test
##
## data: A.APA.M4$residuals
## W = 0.93679, p-value = 0.3115- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.APA.M4)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Altura_aérea ~ F_riego * F_AA, data = datos.APA.M4)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 4.766667 3.330329 6.203004 1.04e-05
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.3708333 -1.494586 2.236252 0.8581268
## 600 ppm-0 ppm 0.4458333 -1.419586 2.311252 0.8026369
## 600 ppm-1000 ppm 0.0750000 -2.079000 2.229000 0.9952559- Los dos riegos presentan diferencias significativas.
# Anova - Altura de la parte aérea - Muestreo 4 - Comparación entre los tratamientos
A.APA.M4.CT <- aov(Altura_aérea ~ Tratamiento, data=datos.APA.M4)
summary(A.APA.M4.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 69.18 23.058 17.69 0.000106 ***
## Residuals 12 15.65 1.304
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.APA.M4.CT <- TukeyHSD(A.APA.M4.CT)
Tukey_A.APA.M4.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Altura_aérea ~ Tratamiento, data = datos.APA.M4)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 0.575 -1.822054 2.972054 0.8903648
## 50 + AS 600ppm-50 - AS 0.650 -1.747054 3.047054 0.8508237
## Control-50 - AS 5.175 2.777946 7.572054 0.0001705
## 50 + AS 600ppm-50 + AS 1000ppm 0.075 -2.322054 2.472054 0.9996958
## Control-50 + AS 1000ppm 4.600 2.202946 6.997054 0.0004988
## Control-50 + AS 600ppm 4.525 2.127946 6.922054 0.0005765- Podemos observar que todos los tratamientos tienen diferencias significativas con respectos a su control
A.APA.M4.DT <- duncan.test(A.APA.M4.CT, 'Tratamiento', console = T)##
## Study: A.APA.M4.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for Altura_aérea
##
## Mean Square Error: 1.30375
##
## Tratamiento, means
##
## Altura_aérea std r se Min Max Q25 Q50
## 50 - AS 10.800 0.9933110 4 0.5709094 9.6 11.8 10.200 10.90
## 50 + AS 1000ppm 11.375 0.9215024 4 0.5709094 10.1 12.3 11.150 11.55
## 50 + AS 600ppm 11.450 1.1090537 4 0.5709094 10.2 12.9 11.025 11.35
## Control 15.975 1.4660036 4 0.5709094 13.8 17.0 15.825 16.55
## Q75
## 50 - AS 11.500
## 50 + AS 1000ppm 11.775
## 50 + AS 600ppm 11.775
## Control 16.700
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 1.759147 1.841322 1.891110
##
## Means with the same letter are not significantly different.
##
## Altura_aérea groups
## Control 15.975 a
## 50 + AS 600ppm 11.450 b
## 50 + AS 1000ppm 11.375 b
## 50 - AS 10.800 bA.APA.M4.DT## $statistics
## MSerror Df Mean CV
## 1.30375 12 12.4 9.208216
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 1.759147
## 3 3.225244 1.841322
## 4 3.312453 1.891110
##
## $means
## Altura_aérea std r se Min Max Q25 Q50
## 50 - AS 10.800 0.9933110 4 0.5709094 9.6 11.8 10.200 10.90
## 50 + AS 1000ppm 11.375 0.9215024 4 0.5709094 10.1 12.3 11.150 11.55
## 50 + AS 600ppm 11.450 1.1090537 4 0.5709094 10.2 12.9 11.025 11.35
## Control 15.975 1.4660036 4 0.5709094 13.8 17.0 15.825 16.55
## Q75
## 50 - AS 11.500
## 50 + AS 1000ppm 11.775
## 50 + AS 600ppm 11.775
## Control 16.700
##
## $comparison
## NULL
##
## $groups
## Altura_aérea groups
## Control 15.975 a
## 50 + AS 600ppm 11.450 b
## 50 + AS 1000ppm 11.375 b
## 50 - AS 10.800 b
##
## attr(,"class")
## [1] "group"Análisis descriptivo - Altura de la parte aérea
# Datos - Gráfico - Contenido de clorofilas
Datos.G.APA <- data.frame(factor(Tratamientos, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), Var_M1$`Altura Aérea (cm)`, Var_M2$`Altura Aérea (cm)`, Var_M3$`Altura Aérea (cm)`, Var_M4$`Altura Aérea (cm)`)
colnames(Datos.G.APA) <- c("Tratamientos", "31", "38","45", "52")
Datos.G.APA# Unificar la variable respuesta en una sola columna
df.long.APA = gather(Datos.G.APA, dds, Altura_aérea, 2:5)
df.long.APA# Diferencias significativas con el control
# M1 M2 M3 M4
DS.APA <- c("a", "a", "a", "a", #T1
"ab", "b*", "b*", "b*", #T2
"ab", "b*", "b*", "b*", #T3
"b*", "b*", "b*", "b*") #T4
# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.APA = group_by(df.long.APA, Tratamientos, dds, ) %>% summarise(mean = mean(Altura_aérea), sd = sd(Altura_aérea))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.APA.ds <- data.frame(df.sumzd.APA, DS.APA)
df.sumzd.APA.dsG.APA = ggplot(df.sumzd.APA, aes(x=dds, y=mean, fill=Tratamientos))+
geom_bar(stat = "identity", position = "dodge", color = "black", alpha = 0.5)+
geom_errorbar(aes(ymin=mean-sd, ymax=mean+sd), width = 0.25, size=0.5, position = position_dodge(0.9), alpha = 0.5)+
scale_fill_manual(values=c("red", "yellow", "blue", "green")) +
theme_minimal()+
ggtitle("Longitud promedio de la parte aérea \nen los diferentes muestreos") + theme(plot.title = element_text(family = "Times New Roman",
size = rel(1.5),
vjust = 0.5,
hjust = 0.5))+
labs(x = "Días después de siembra", y = "cm")+
theme(axis.title.x = element_text(family = "Times New Roman"), axis.title.y = element_text(family = "Times New Roman"))+
theme(legend.title = element_text(family = "Times New Roman"))+
geom_text(aes(x=dds, y= mean+3, label=DS.APA),
position = position_dodge(width = 1), size = 5)
G.APA
Contenido relativo de agua
Análisis estadístico - Muestreo 1
# Datos - CRA - Muestreo 1
datos.CRA.M1 <- data.frame(CRA_M1$`Factor riego`, CRA_M1$`Factor ácido ascórbico`, CRA_M1$Tratamiento, CRA_M1$`CRA (%)`)
colnames(datos.CRA.M1) <- c("F_riego", "F_AA", "Tratamiento", "CRA")
datos.CRA.M1datos.CRA.M1$F_riego <- as.factor(datos.CRA.M1$F_riego)
datos.CRA.M1$F_AA <- as.factor(datos.CRA.M1$F_AA)
str(datos.CRA.M1)## 'data.frame': 32 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 2 2 2 2 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ CRA : num 75.6 88.3 77.7 75.9 69.4 ...# Anova - CRA - Muestreo 1
A.CRA.M1 <- aov(CRA ~ F_riego*F_AA, data = datos.CRA.M1)
summary(A.CRA.M1)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 180.5 180.48 5.848 0.0224 *
## F_AA 2 134.4 67.18 2.177 0.1322
## Residuals 28 864.2 30.86
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 95%.
# Prueba de normalidad de residuos
shapiro.test(A.CRA.M1$residuals)##
## Shapiro-Wilk normality test
##
## data: A.CRA.M1$residuals
## W = 0.95536, p-value = 0.2043- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.CRA.M1)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = CRA ~ F_riego * F_AA, data = datos.CRA.M1)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 5.484586 0.8387028 10.13047 0.022352
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -0.4689017 -6.421265 5.483461 0.9792881
## 600 ppm-0 ppm 4.2895309 -1.662832 10.241894 0.1935871
## 600 ppm-1000 ppm 4.7584326 -2.114764 11.631629 0.2181225- Los dos tipos de riego presenta diferencias significativas.
# Anova - CRA - Muestreo 1 - Comparación entre los tratamientos
A.CRA.M1.CT <- aov(CRA ~ Tratamiento, data=datos.CRA.M1)
summary(A.CRA.M1.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 314.8 104.95 3.4 0.0314 *
## Residuals 28 864.2 30.86
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.CRA.M1.CT <- TukeyHSD(A.CRA.M1.CT)
Tukey_A.CRA.M1.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = CRA ~ Tratamiento, data = datos.CRA.M1)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 0.4862556 -7.0979450 8.070456 0.9980472
## 50 + AS 600ppm-50 - AS 5.2446882 -2.3395124 12.828889 0.2559147
## Control-50 - AS 7.3949003 -0.1893003 14.979101 0.0580147
## 50 + AS 600ppm-50 + AS 1000ppm 4.7584326 -2.8257681 12.342633 0.3360852
## Control-50 + AS 1000ppm 6.9086447 -0.6755560 14.492845 0.0840015
## Control-50 + AS 600ppm 2.1502121 -5.4339886 9.734413 0.8654710- Podemos observar que ninguno de los tratamientos tiene diferencias significativas con el control para el muestreo 1, sin embargo, el tratamiento amarillo (50 - AS) y verde (50 + AS 1000ppm), tuvieron valores muy cercanos a 5%, por lo tanto, lo podríamos tener en cuenta a nivel agronómico.
A.CRA.M1.DT <- duncan.test(A.CRA.M1.CT, 'Tratamiento', console = T)##
## Study: A.CRA.M1.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for CRA
##
## Mean Square Error: 30.86421
##
## Tratamiento, means
##
## CRA std r se Min Max Q25
## 50 - AS 69.61368 3.953001 8 1.964186 61.62162 74.07407 67.84085
## 50 + AS 1000ppm 70.09994 2.807400 8 1.964186 65.50079 74.35508 68.41194
## 50 + AS 600ppm 74.85837 4.979654 8 1.964186 67.95977 83.39350 71.77992
## Control 77.00858 8.669036 8 1.964186 62.71845 88.42795 74.06250
## Q50 Q75
## 50 - AS 70.52384 72.07698
## 50 + AS 1000ppm 70.13269 71.35958
## 50 + AS 600ppm 74.97420 77.46417
## Control 76.77943 80.59863
##
## Alpha: 0.05 ; DF Error: 28
##
## Critical Range
## 2 3 4
## 5.690021 5.978681 6.165302
##
## Means with the same letter are not significantly different.
##
## CRA groups
## Control 77.00858 a
## 50 + AS 600ppm 74.85837 ab
## 50 + AS 1000ppm 70.09994 b
## 50 - AS 69.61368 bA.CRA.M1.DT## $statistics
## MSerror Df Mean CV
## 30.86421 28 72.89514 7.621299
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 2.896885 5.690021
## 3 3.043847 5.978681
## 4 3.138859 6.165302
##
## $means
## CRA std r se Min Max Q25
## 50 - AS 69.61368 3.953001 8 1.964186 61.62162 74.07407 67.84085
## 50 + AS 1000ppm 70.09994 2.807400 8 1.964186 65.50079 74.35508 68.41194
## 50 + AS 600ppm 74.85837 4.979654 8 1.964186 67.95977 83.39350 71.77992
## Control 77.00858 8.669036 8 1.964186 62.71845 88.42795 74.06250
## Q50 Q75
## 50 - AS 70.52384 72.07698
## 50 + AS 1000ppm 70.13269 71.35958
## 50 + AS 600ppm 74.97420 77.46417
## Control 76.77943 80.59863
##
## $comparison
## NULL
##
## $groups
## CRA groups
## Control 77.00858 a
## 50 + AS 600ppm 74.85837 ab
## 50 + AS 1000ppm 70.09994 b
## 50 - AS 69.61368 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 2
# Datos - CRA - Muestreo 1
datos.CRA.M2 <- data.frame(CRA_M2$`Factor riego`, CRA_M2$`Factor ácido ascórbico`, CRA_M2$Tratamiento, CRA_M2$`CRA (%)`)
colnames(datos.CRA.M2) <- c("F_riego", "F_AA", "Tratamiento", "CRA")
datos.CRA.M2datos.CRA.M2$F_riego <- as.factor(datos.CRA.M2$F_riego)
datos.CRA.M2$F_AA <- as.factor(datos.CRA.M2$F_AA)
str(datos.CRA.M2)## 'data.frame': 32 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 2 2 2 2 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ CRA : num 74.1 76.3 77.4 69.8 78 ...# Anova - CRA - Muestreo 2
A.CRA.M2 <- aov(CRA ~ F_riego*F_AA, data = datos.CRA.M2)
summary(A.CRA.M2)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 300.4 300.4 5.066 0.0324 *
## F_AA 2 174.2 87.1 1.469 0.2474
## Residuals 28 1660.5 59.3
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de concentración de ácido ascórbico, como factor independiente, tuvieron diferencias con una significancia del 95%.
# Prueba de normalidad de residuos
shapiro.test(A.CRA.M2$residuals)##
## Shapiro-Wilk normality test
##
## data: A.CRA.M2$residuals
## W = 0.97117, p-value = 0.5324- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.CRA.M2)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = CRA ~ F_riego * F_AA, data = datos.CRA.M2)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 7.075756 0.6358926 13.51562 0.0324456
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 5.0401418 -3.210691 13.290975 0.3009616
## 600 ppm-0 ppm 0.6719716 -7.578861 8.922804 0.9778790
## 600 ppm-1000 ppm -4.3681702 -13.895411 5.159071 0.5014071- Los dos tipos de riego presenta diferencias significativas.
# Anova - CRA - Muestreo 2 - Comparación entre los tratamientos
A.CRA.M2.CT <- aov(CRA ~ Tratamiento, data=datos.CRA.M2)
summary(A.CRA.M2.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 474.6 158.2 2.668 0.067 .
## Residuals 28 1660.5 59.3
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.CRA.M2.CT <- TukeyHSD(A.CRA.M2.CT)
Tukey_A.CRA.M2.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = CRA ~ Tratamiento, data = datos.CRA.M2)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 6.468170 -4.0446247 16.980965 0.3528087
## 50 + AS 600ppm-50 - AS 2.100000 -8.4127949 12.612795 0.9470176
## Control-50 - AS 9.931812 -0.5809825 20.444607 0.0692571
## 50 + AS 600ppm-50 + AS 1000ppm -4.368170 -14.8809650 6.144625 0.6717789
## Control-50 + AS 1000ppm 3.463642 -7.0491527 13.976437 0.8051155
## Control-50 + AS 600ppm 7.831812 -2.6809825 18.344607 0.1999174- Podemos observar que ninguno de los tratamientos tiene diferencias significativas con el control para el muestreo 2.
A.CRA.M2.DT <- duncan.test(A.CRA.M2.CT, 'Tratamiento', console = T)##
## Study: A.CRA.M2.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for CRA
##
## Mean Square Error: 59.30236
##
## Tratamiento, means
##
## CRA std r se Min Max Q25
## 50 - AS 67.97971 7.724836 8 2.722645 54.90411 80.00000 64.15507
## 50 + AS 1000ppm 74.44788 8.867789 8 2.722645 67.57679 93.42105 67.90344
## 50 + AS 600ppm 70.07971 7.774080 8 2.722645 58.90411 80.00000 64.90507
## Control 77.91153 6.201802 8 2.722645 69.82759 88.58075 74.00252
## Q50 Q75
## 50 - AS 68.15828 72.64969
## 50 + AS 1000ppm 71.35770 77.43724
## 50 + AS 600ppm 70.27954 75.58594
## Control 76.85915 79.83550
##
## Alpha: 0.05 ; DF Error: 28
##
## Critical Range
## 2 3 4
## 7.887189 8.287314 8.545998
##
## Means with the same letter are not significantly different.
##
## CRA groups
## Control 77.91153 a
## 50 + AS 1000ppm 74.44788 ab
## 50 + AS 600ppm 70.07971 ab
## 50 - AS 67.97971 bA.CRA.M2.DT## $statistics
## MSerror Df Mean CV
## 59.30236 28 72.60471 10.60648
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 2.896885 7.887189
## 3 3.043847 8.287314
## 4 3.138859 8.545998
##
## $means
## CRA std r se Min Max Q25
## 50 - AS 67.97971 7.724836 8 2.722645 54.90411 80.00000 64.15507
## 50 + AS 1000ppm 74.44788 8.867789 8 2.722645 67.57679 93.42105 67.90344
## 50 + AS 600ppm 70.07971 7.774080 8 2.722645 58.90411 80.00000 64.90507
## Control 77.91153 6.201802 8 2.722645 69.82759 88.58075 74.00252
## Q50 Q75
## 50 - AS 68.15828 72.64969
## 50 + AS 1000ppm 71.35770 77.43724
## 50 + AS 600ppm 70.27954 75.58594
## Control 76.85915 79.83550
##
## $comparison
## NULL
##
## $groups
## CRA groups
## Control 77.91153 a
## 50 + AS 1000ppm 74.44788 ab
## 50 + AS 600ppm 70.07971 ab
## 50 - AS 67.97971 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 3
# Datos - CRA - Muestreo 3
datos.CRA.M3 <- data.frame(CRA_M3$`Factor riego`, CRA_M3$`Factor ácido ascórbico`, CRA_M3$Tratamiento, CRA_M3$`CRA (%)`)
colnames(datos.CRA.M3) <- c("F_riego", "F_AA", "Tratamiento", "CRA")
datos.CRA.M3datos.CRA.M3$F_riego <- as.factor(datos.CRA.M3$F_riego)
datos.CRA.M3$F_AA <- as.factor(datos.CRA.M3$F_AA)
str(datos.CRA.M3)## 'data.frame': 32 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 2 2 2 2 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ CRA : num 65.8 73.7 94.5 86.4 75.8 ...# Anova - CRA - Muestreo 1
A.CRA.M3 <- aov(CRA ~ F_riego*F_AA, data = datos.CRA.M3)
summary(A.CRA.M3)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 539.1 539.1 4.980 0.0338 *
## F_AA 2 643.5 321.8 2.972 0.0675 .
## Residuals 28 3031.1 108.3
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 95%.
# Prueba de normalidad de residuos
shapiro.test(A.CRA.M3$residuals)##
## Shapiro-Wilk normality test
##
## data: A.CRA.M3$residuals
## W = 0.9383, p-value = 0.06701- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.CRA.M3)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = CRA ~ F_riego * F_AA, data = datos.CRA.M3)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 9.478762 0.7779699 18.17955 0.0338272
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 1.255720 -9.891844 12.40328 0.9581431
## 600 ppm-0 ppm 9.686756 -1.460807 20.83432 0.0979293
## 600 ppm-1000 ppm 8.431037 -4.441061 21.30313 0.2538127- Los dos tipos de riego presenta diferencias significativas.
# Anova - CRA - Muestreo 3 - Comparación entre los tratamientos
A.CRA.M3.CT <- aov(CRA ~ Tratamiento, data=datos.CRA.M3)
summary(A.CRA.M3.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 1183 394.2 3.642 0.0246 *
## Residuals 28 3031 108.3
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.CRA.M3.CT <- TukeyHSD(A.CRA.M3.CT)
Tukey_A.CRA.M3.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = CRA ~ Tratamiento, data = datos.CRA.M3)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 3.991338 -10.2123245 18.19500 0.8684702
## 50 + AS 600ppm-50 - AS 12.422375 -1.7812879 26.62604 0.1027731
## Control-50 - AS 14.950000 0.7463371 29.15366 0.0362826
## 50 + AS 600ppm-50 + AS 1000ppm 8.431037 -5.7726264 22.63470 0.3837867
## Control-50 + AS 1000ppm 10.958662 -3.2450014 25.16232 0.1757293
## Control-50 + AS 600ppm 2.527625 -11.6760379 16.73129 0.9616045- Podemos observar que ninguno de los tratamientos tiene diferencias significativas con el control para el muestreo 1.
A.CRA.M3.DT <- duncan.test(A.CRA.M3.CT, 'Tratamiento', console = T)##
## Study: A.CRA.M3.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for CRA
##
## Mean Square Error: 108.2521
##
## Tratamiento, means
##
## CRA std r se Min Max Q25
## 50 - AS 62.98490 10.877915 8 3.67852 53.74570 77.99320 55.29269
## 50 + AS 1000ppm 66.97624 13.226623 8 3.67852 51.17057 85.11628 56.06290
## 50 + AS 600ppm 75.40728 7.930765 8 3.67852 63.74570 86.37413 71.80557
## Control 77.93490 8.765771 8 3.67852 65.76000 94.49074 73.85449
## Q50 Q75
## 50 - AS 56.21180 74.76186
## 50 + AS 1000ppm 64.92068 79.58764
## 50 + AS 600ppm 75.63183 79.47509
## Control 75.66183 80.01369
##
## Alpha: 0.05 ; DF Error: 28
##
## Critical Range
## 2 3 4
## 10.65625 11.19685 11.54636
##
## Means with the same letter are not significantly different.
##
## CRA groups
## Control 77.93490 a
## 50 + AS 600ppm 75.40728 a
## 50 + AS 1000ppm 66.97624 ab
## 50 - AS 62.98490 bA.CRA.M3.DT## $statistics
## MSerror Df Mean CV
## 108.2521 28 70.82583 14.69016
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 2.896885 10.65625
## 3 3.043847 11.19685
## 4 3.138859 11.54636
##
## $means
## CRA std r se Min Max Q25
## 50 - AS 62.98490 10.877915 8 3.67852 53.74570 77.99320 55.29269
## 50 + AS 1000ppm 66.97624 13.226623 8 3.67852 51.17057 85.11628 56.06290
## 50 + AS 600ppm 75.40728 7.930765 8 3.67852 63.74570 86.37413 71.80557
## Control 77.93490 8.765771 8 3.67852 65.76000 94.49074 73.85449
## Q50 Q75
## 50 - AS 56.21180 74.76186
## 50 + AS 1000ppm 64.92068 79.58764
## 50 + AS 600ppm 75.63183 79.47509
## Control 75.66183 80.01369
##
## $comparison
## NULL
##
## $groups
## CRA groups
## Control 77.93490 a
## 50 + AS 600ppm 75.40728 a
## 50 + AS 1000ppm 66.97624 ab
## 50 - AS 62.98490 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 4
# Datos - CRA - Muestreo 4
datos.CRA.M4 <- data.frame(CRA_M4$`Factor riego`, CRA_M4$`Factor ácido ascórbico`, CRA_M4$Tratamiento, CRA_M4$`CRA (%)`)
colnames(datos.CRA.M4) <- c("F_riego", "F_AA", "Tratamiento", "CRA")
datos.CRA.M4datos.CRA.M4$F_riego <- as.factor(datos.CRA.M4$F_riego)
datos.CRA.M4$F_AA <- as.factor(datos.CRA.M4$F_AA)
str(datos.CRA.M4)## 'data.frame': 32 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 2 2 2 2 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ CRA : num 80.8 70.9 81.4 80.3 75.3 ...# Anova - CRA - Muestreo 4
A.CRA.M4 <- aov(CRA ~ F_riego*F_AA, data = datos.CRA.M4)
summary(A.CRA.M4)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 350.3 350.3 5.497 0.0264 *
## F_AA 2 139.2 69.6 1.092 0.3493
## Residuals 28 1784.5 63.7
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 95%.
# Prueba de normalidad de residuos
shapiro.test(A.CRA.M4$residuals)##
## Shapiro-Wilk normality test
##
## data: A.CRA.M4$residuals
## W = 0.97169, p-value = 0.5474- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.CRA.M4)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = CRA ~ F_riego * F_AA, data = datos.CRA.M4)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 7.641296 0.9652879 14.3173 0.0263693
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 3.6066695 -4.946716 12.160055 0.5563734
## 600 ppm-0 ppm 3.1886661 -5.364719 11.742051 0.6309138
## 600 ppm-1000 ppm -0.4180034 -10.294602 9.458595 0.9939736- Los dos tipos de riego presenta diferencias significativas.
# Anova - CRA - Muestreo 4 - Comparación entre los tratamientos
A.CRA.M4.CT <- aov(CRA ~ Tratamiento, data=datos.CRA.M4)
summary(A.CRA.M4.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 489.6 163.19 2.561 0.075 .
## Residuals 28 1784.5 63.73
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.CRA.M4.CT <- TukeyHSD(A.CRA.M4.CT)
Tukey_A.CRA.M4.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = CRA ~ Tratamiento, data = datos.CRA.M4)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 5.3055034 -5.5927885 16.20380 0.5527294
## 50 + AS 600ppm-50 - AS 4.8875000 -6.0107919 15.78579 0.6169473
## Control-50 - AS 11.0389641 0.1406722 21.93726 0.0462560
## 50 + AS 600ppm-50 + AS 1000ppm -0.4180034 -11.3162952 10.48029 0.9995785
## Control-50 + AS 1000ppm 5.7334607 -5.1648311 16.63175 0.4881159
## Control-50 + AS 600ppm 6.1514641 -4.7468278 17.04976 0.4275029- Podemos observar que ninguno de los tratamientos tiene diferencias significativas con el control para el muestreo 1.
A.CRA.M4.DT <- duncan.test(A.CRA.M4.CT, 'Tratamiento', console = T)##
## Study: A.CRA.M4.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for CRA
##
## Mean Square Error: 63.73125
##
## Tratamiento, means
##
## CRA std r se Min Max Q25
## 50 - AS 66.89244 10.463300 8 2.822482 52.00000 79.89798 60.66041
## 50 + AS 1000ppm 72.19795 10.181476 8 2.822482 58.30450 87.55869 65.55365
## 50 + AS 600ppm 71.77994 5.244714 8 2.822482 63.35740 79.79798 69.27785
## Control 77.93141 3.778212 8 2.822482 70.94972 81.38686 75.16088
## Q50 Q75
## 50 - AS 68.62023 73.76648
## 50 + AS 1000ppm 70.23930 77.04855
## 50 + AS 600ppm 72.19792 73.72169
## Control 79.93398 80.46572
##
## Alpha: 0.05 ; DF Error: 28
##
## Critical Range
## 2 3 4
## 8.176407 8.591204 8.859373
##
## Means with the same letter are not significantly different.
##
## CRA groups
## Control 77.93141 a
## 50 + AS 1000ppm 72.19795 ab
## 50 + AS 600ppm 71.77994 ab
## 50 - AS 66.89244 bA.CRA.M4.DT## $statistics
## MSerror Df Mean CV
## 63.73125 28 72.20043 11.05698
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 2.896885 8.176407
## 3 3.043847 8.591204
## 4 3.138859 8.859373
##
## $means
## CRA std r se Min Max Q25
## 50 - AS 66.89244 10.463300 8 2.822482 52.00000 79.89798 60.66041
## 50 + AS 1000ppm 72.19795 10.181476 8 2.822482 58.30450 87.55869 65.55365
## 50 + AS 600ppm 71.77994 5.244714 8 2.822482 63.35740 79.79798 69.27785
## Control 77.93141 3.778212 8 2.822482 70.94972 81.38686 75.16088
## Q50 Q75
## 50 - AS 68.62023 73.76648
## 50 + AS 1000ppm 70.23930 77.04855
## 50 + AS 600ppm 72.19792 73.72169
## Control 79.93398 80.46572
##
## $comparison
## NULL
##
## $groups
## CRA groups
## Control 77.93141 a
## 50 + AS 1000ppm 72.19795 ab
## 50 + AS 600ppm 71.77994 ab
## 50 - AS 66.89244 b
##
## attr(,"class")
## [1] "group"Análisis descriptivo - Contenido relativo de agua
# Datos - Gráfico - Contenido relativo de agua
Tratamientos_2 <- c(
rep("Control", 8),
rep("50 - AS", 8),
rep("50 + AS 600ppm", 8),
rep("50 + AS 1000ppm", 8)
)
Datos.G.CRA <- data.frame(factor(Tratamientos_2, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), CRA_M1$`CRA (%)`, CRA_M2$`CRA (%)`, CRA_M3$`CRA (%)`, CRA_M4$`CRA (%)`)
colnames(Datos.G.CRA) <- c("Tratamientos", "31", "38","45", "52")
Datos.G.CRA# Unificar la variable respuesta en una sola columna
df.long.CRA = gather(Datos.G.CRA, dds, CRA, 2:5)
df.long.CRA# Diferencias significativas con el control
# M1 M2 M3 M4
DS.CRA <- c("a", "a", "a", "a", #T1
"b*", "b*", "b*", "b*", #T2
"ab", "ab*", "a", "ab*", #T3
"b*", "ab", "ab*", "ab*") #T4
# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.CRA = group_by(df.long.CRA, Tratamientos, dds, ) %>% summarise(mean = mean(CRA), sd = sd(CRA))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.CRA.ds <- data.frame(df.sumzd.CRA, DS.CRA)
df.sumzd.CRA.dsG.CRA = ggplot(df.sumzd.CRA, aes(x=dds, y=mean, fill=Tratamientos, xmax = 4.7))+
geom_bar(stat = "identity", position = "dodge", color = "black", alpha = 0.5)+
geom_errorbar(aes(ymin=mean-sd, ymax=mean+sd), width = 0.25, size=0.5, position = position_dodge(0.9), alpha = 0.5)+
scale_fill_manual(values=c("red", "yellow", "blue", "green")) +
theme_minimal()+
theme(axis.text.x = element_text(size=15),
axis.text.y = element_text(size=15))+
ggtitle("Contenido relativo de agua \nen los diferentes muestreos") + theme(plot.title = element_text(family = "Times New Roman",
size = rel(2),
vjust = 0.5,
hjust = 0.5))+
labs(x = "Días después de siembra", y = "% promedio")+
theme(axis.title.x = element_text(family = "Times New Roman", size = rel(1.5)), axis.title.y = element_text(family = "Times New Roman", size = rel(1.5)))+
theme(legend.title = element_text(family = "Times New Roman", size = rel(1.5)))+
geom_text(aes(x=dds, y= mean+1.2, label=DS.CRA),
position = position_dodge(width = 1), size = 5)+
geom_hline(yintercept = 77.93, linetype = "dashed", color = "black", size = 1)+
geom_hline(yintercept = 74.03, linetype = "dashed", color = "black", size = 1)+
geom_hline(yintercept = 66.24, linetype = "dashed", color = "black", size = 1)+
geom_hline(yintercept = 62.34, linetype = "dashed", color = "black", size = 1)+
geom_text(aes(x = 4.65, y = 80, label="100%"), size = 5)+
geom_text(aes(x = 4.65, y = 76, label="95%"), size = 5)+
geom_text(aes(x = 4.65, y = 68, label="85%"), size = 5)+
geom_text(aes(x = 4.65, y = 64, label="70%"), size = 5)+
coord_cartesian(xlim = c(1, 4.3), ylim = c(61, 85))
G.CRA
Peso fresco de la parte aérea
Análisis estadístico - Muestreo 1
# Datos - Peso fresco de la parte aérea - Muestreo 1
datos.PFPA.M1 <- data.frame(Var_M1$`Factor riego`, Var_M1$`Factor ácido ascórbico`, Var_M1$Tratamiento, Var_M1$`Peso fresco de parte aérea completo`)
colnames(datos.PFPA.M1) <- c("F_riego", "F_AA", "Tratamiento", "PFPA")
datos.PFPA.M1datos.PFPA.M1$F_riego <- as.factor(datos.PFPA.M1$F_riego)
datos.PFPA.M1$F_AA <- as.factor(datos.PFPA.M1$F_AA)
str(datos.PFPA.M1)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PFPA : num 3.85 5.02 3.15 4.29 2.26 ...# Anova - Peso fresco de la parte aérea - Muestreo 1
A.PFPA.M1 <- aov(PFPA ~ F_riego*F_AA, data = datos.PFPA.M1)
summary(A.PFPA.M1)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 7.762 7.762 15.71 0.00188 **
## F_AA 2 0.128 0.064 0.13 0.87961
## Residuals 12 5.931 0.494
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99%.
# Prueba de normalidad de residuos
shapiro.test(A.PFPA.M1$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PFPA.M1$residuals
## W = 0.97928, p-value = 0.9577- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.PFPA.M1)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFPA ~ F_riego * F_AA, data = datos.PFPA.M1)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 1.608567 0.7242183 2.492915 0.0018828
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -0.09346667 -1.241999 1.055066 0.9744048
## 600 ppm-0 ppm -0.18086667 -1.329399 0.967666 0.9080182
## 600 ppm-1000 ppm -0.08740000 -1.413611 1.238811 0.9831268- Los dos tipos de riego presenta diferencias significativas.
# Anova - PFPA - Muestreo 1 - Comparación entre los tratamientos
A.PFPA.M1.CT <- aov(PFPA ~ Tratamiento, data=datos.PFPA.M1)
summary(A.PFPA.M1.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 7.891 2.6302 5.322 0.0145 *
## Residuals 12 5.931 0.4942
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.PFPA.M1.CT <- TukeyHSD(A.PFPA.M1.CT)
Tukey_A.PFPA.M1.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFPA ~ Tratamiento, data = datos.PFPA.M1)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -0.16205 -1.637908858 1.313809 0.9874340
## 50 + AS 600ppm-50 - AS -0.24945 -1.725308858 1.226409 0.9570973
## Control-50 - AS 1.47140 -0.004458858 2.947259 0.0507800
## 50 + AS 600ppm-50 + AS 1000ppm -0.08740 -1.563258858 1.388459 0.9979585
## Control-50 + AS 1000ppm 1.63345 0.157591142 3.109309 0.0288036
## Control-50 + AS 600ppm 1.72085 0.244991142 3.196709 0.0211685- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control, sin embargo, para el caso del tratamiento amarillo (50 - AS), este presenta un p-valor de 5.07%, que aunque este por encima, vale la pena tenerlo en cuenta.
A.PFPA.M1.DT <- duncan.test(A.PFPA.M1.CT, 'Tratamiento', console = T)##
## Study: A.PFPA.M1.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PFPA
##
## Mean Square Error: 0.494229
##
## Tratamiento, means
##
## PFPA std r se Min Max Q25 Q50
## 50 - AS 2.60620 0.4561895 4 0.3515071 2.2592 3.2764 2.371625 2.44460
## 50 + AS 1000ppm 2.44415 0.9873562 4 0.3515071 1.3531 3.4981 1.766800 2.46270
## 50 + AS 600ppm 2.35675 0.4223427 4 0.3515071 1.7529 2.6585 2.221575 2.50780
## Control 4.07760 0.7845771 4 0.3515071 3.1523 5.0226 3.673475 4.06775
## Q75
## 50 - AS 2.679175
## 50 + AS 1000ppm 3.140050
## 50 + AS 600ppm 2.642975
## Control 4.471875
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 1.083101 1.133696 1.164351
##
## Means with the same letter are not significantly different.
##
## PFPA groups
## Control 4.07760 a
## 50 - AS 2.60620 b
## 50 + AS 1000ppm 2.44415 b
## 50 + AS 600ppm 2.35675 bA.PFPA.M1.DT## $statistics
## MSerror Df Mean CV
## 0.494229 12 2.871175 24.48524
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 1.083101
## 3 3.225244 1.133696
## 4 3.312453 1.164351
##
## $means
## PFPA std r se Min Max Q25 Q50
## 50 - AS 2.60620 0.4561895 4 0.3515071 2.2592 3.2764 2.371625 2.44460
## 50 + AS 1000ppm 2.44415 0.9873562 4 0.3515071 1.3531 3.4981 1.766800 2.46270
## 50 + AS 600ppm 2.35675 0.4223427 4 0.3515071 1.7529 2.6585 2.221575 2.50780
## Control 4.07760 0.7845771 4 0.3515071 3.1523 5.0226 3.673475 4.06775
## Q75
## 50 - AS 2.679175
## 50 + AS 1000ppm 3.140050
## 50 + AS 600ppm 2.642975
## Control 4.471875
##
## $comparison
## NULL
##
## $groups
## PFPA groups
## Control 4.07760 a
## 50 - AS 2.60620 b
## 50 + AS 1000ppm 2.44415 b
## 50 + AS 600ppm 2.35675 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 2
# Datos - Peso fresco de la parte aérea - Muestreo 2
datos.PFPA.M2 <- data.frame(Var_M2$`Factor riego`, Var_M2$`Factor ácido ascórbico`, Var_M2$Tratamiento, Var_M2$`Peso fresco de parte aérea completo`)
colnames(datos.PFPA.M2) <- c("F_riego", "F_AA", "Tratamiento", "PFPA")
datos.PFPA.M2datos.PFPA.M2$F_riego <- as.factor(datos.PFPA.M2$F_riego)
datos.PFPA.M2$F_AA <- as.factor(datos.PFPA.M2$F_AA)
str(datos.PFPA.M2)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PFPA : num 5.22 6.85 7.98 8.59 4.23 ...# Anova - Peso fresco de la parte aérea - Muestreo 2
A.PFPA.M2 <- aov(PFPA ~ F_riego*F_AA, data = datos.PFPA.M2)
summary(A.PFPA.M2)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 32.13 32.13 31.268 0.000118 ***
## F_AA 2 0.17 0.09 0.083 0.921018
## Residuals 12 12.33 1.03
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.PFPA.M2$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PFPA.M2$residuals
## W = 0.93394, p-value = 0.2812- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.PFPA.M2)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFPA ~ F_riego * F_AA, data = datos.PFPA.M2)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 3.272583 1.997436 4.547731 0.0001177
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.01229167 -1.643785 1.668368 0.9997839
## 600 ppm-0 ppm -0.21845833 -1.874535 1.437618 0.9343751
## 600 ppm-1000 ppm -0.23075000 -2.143023 1.681523 0.9447330- Los dos tipos de riego presenta diferencias significativas.
# Anova - PFPA - Muestreo 2 - Comparación entre los tratamientos
A.PFPA.M2.CT <- aov(PFPA ~ Tratamiento, data=datos.PFPA.M2)
summary(A.PFPA.M2.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 32.30 10.767 10.48 0.00114 **
## Residuals 12 12.33 1.028
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.PFPA.M2.CT <- TukeyHSD(A.PFPA.M2.CT)
Tukey_A.PFPA.M2.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFPA ~ Tratamiento, data = datos.PFPA.M2)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -0.03925 -2.16730 2.08880 0.9999375
## 50 + AS 600ppm-50 - AS -0.27000 -2.39805 1.85805 0.9809263
## Control-50 - AS 3.16950 1.04145 5.29755 0.0039973
## 50 + AS 600ppm-50 + AS 1000ppm -0.23075 -2.35880 1.89730 0.9878831
## Control-50 + AS 1000ppm 3.20875 1.08070 5.33680 0.0036419
## Control-50 + AS 600ppm 3.43950 1.31145 5.56755 0.0021198- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.PFPA.M2.DT <- duncan.test(A.PFPA.M2.CT, 'Tratamiento', console = T)##
## Study: A.PFPA.M2.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PFPA
##
## Mean Square Error: 1.027548
##
## Tratamiento, means
##
## PFPA std r se Min Max Q25 Q50 Q75
## 50 - AS 3.99100 0.9177901 4 0.5068403 2.641 4.667 3.8305 4.3280 4.48850
## 50 + AS 1000ppm 3.95175 0.9023042 4 0.5068403 2.649 4.680 3.7230 4.2390 4.46775
## 50 + AS 600ppm 3.72100 0.5142963 4 0.5068403 3.201 4.390 3.4005 3.6465 3.96700
## Control 7.16050 1.4795951 4 0.5068403 5.220 8.587 6.4455 7.4175 8.13250
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 1.561730 1.634684 1.678885
##
## Means with the same letter are not significantly different.
##
## PFPA groups
## Control 7.16050 a
## 50 - AS 3.99100 b
## 50 + AS 1000ppm 3.95175 b
## 50 + AS 600ppm 3.72100 bA.PFPA.M2.DT## $statistics
## MSerror Df Mean CV
## 1.027548 12 4.706062 21.53989
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 1.561730
## 3 3.225244 1.634684
## 4 3.312453 1.678885
##
## $means
## PFPA std r se Min Max Q25 Q50 Q75
## 50 - AS 3.99100 0.9177901 4 0.5068403 2.641 4.667 3.8305 4.3280 4.48850
## 50 + AS 1000ppm 3.95175 0.9023042 4 0.5068403 2.649 4.680 3.7230 4.2390 4.46775
## 50 + AS 600ppm 3.72100 0.5142963 4 0.5068403 3.201 4.390 3.4005 3.6465 3.96700
## Control 7.16050 1.4795951 4 0.5068403 5.220 8.587 6.4455 7.4175 8.13250
##
## $comparison
## NULL
##
## $groups
## PFPA groups
## Control 7.16050 a
## 50 - AS 3.99100 b
## 50 + AS 1000ppm 3.95175 b
## 50 + AS 600ppm 3.72100 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 3
# Datos - Peso fresco de la parte aérea - Muestreo 3
datos.PFPA.M3 <- data.frame(Var_M3$`Factor riego`, Var_M3$`Factor ácido ascórbico`, Var_M3$Tratamiento, Var_M3$`Peso fresco de parte aérea completo`)
colnames(datos.PFPA.M3) <- c("F_riego", "F_AA", "Tratamiento", "PFPA")
datos.PFPA.M3datos.PFPA.M3$F_riego <- as.factor(datos.PFPA.M3$F_riego)
datos.PFPA.M3$F_AA <- as.factor(datos.PFPA.M3$F_AA)
str(datos.PFPA.M3)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PFPA : num 8.2 9.31 11.59 8.04 2.9 ...# Anova - Peso fresco de la parte aérea - Muestreo 3
A.PFPA.M3 <- aov(PFPA ~ F_riego*F_AA, data = datos.PFPA.M3)
summary(A.PFPA.M3)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 116.83 116.83 131.298 8.08e-08 ***
## F_AA 2 0.30 0.15 0.166 0.849
## Residuals 12 10.68 0.89
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.PFPA.M3$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PFPA.M3$residuals
## W = 0.86216, p-value = 0.02067- Los datos no tienen una distribución normal, en este caso hay que aplicar otra prueba que soporte la distribución de los datos de esta variable, sin embargo, en este caso no se realizará, puesto a que ya sabiamos, desde la toma de los datos, que esta varibale no iba a dar diferencias significativas.
TukeyHSD(A.PFPA.M3)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFPA ~ F_riego * F_AA, data = datos.PFPA.M3)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 6.24055 5.053926 7.427174 1e-07
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -0.14625 -1.687358 1.394858 0.9653828
## 600 ppm-0 ppm 0.23125 -1.309858 1.772358 0.9160560
## 600 ppm-1000 ppm 0.37750 -1.402018 2.157018 0.8403720- Los dos tipos de riego presenta diferencias significativas.
# Anova - PFPA - Muestreo 3 - Comparación entre los tratamientos
A.PFPA.M3.CT <- aov(PFPA ~ Tratamiento, data=datos.PFPA.M3)
summary(A.PFPA.M3.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 117.13 39.04 43.88 9.61e-07 ***
## Residuals 12 10.68 0.89
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.PFPA.M3.CT <- TukeyHSD(A.PFPA.M3.CT)
Tukey_A.PFPA.M3.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFPA ~ Tratamiento, data = datos.PFPA.M3)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -0.12500 -2.105316 1.855316 0.9975327
## 50 + AS 600ppm-50 - AS 0.25250 -1.727816 2.232816 0.9806541
## Control-50 - AS 6.28305 4.302734 8.263366 0.0000036
## 50 + AS 600ppm-50 + AS 1000ppm 0.37750 -1.602816 2.357816 0.9403034
## Control-50 + AS 1000ppm 6.40805 4.427734 8.388366 0.0000029
## Control-50 + AS 600ppm 6.03055 4.050234 8.010866 0.0000055- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.PFPA.M3.DT <- duncan.test(A.PFPA.M3.CT, 'Tratamiento', console = T)##
## Study: A.PFPA.M3.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PFPA
##
## Mean Square Error: 0.8898312
##
## Tratamiento, means
##
## PFPA std r se Min Max Q25 Q50
## 50 - AS 3.001075 0.2689921 4 0.4716543 2.7498 3.3791 2.86170 2.9377
## 50 + AS 1000ppm 2.876075 0.2030149 4 0.4716543 2.7498 3.1791 2.76975 2.7877
## 50 + AS 600ppm 3.253575 0.8709956 4 0.4716543 2.1129 4.2142 2.93565 3.3436
## Control 9.284125 1.6392436 4 0.4716543 8.0370 11.5909 8.15625 8.7543
## Q75
## 50 - AS 3.077075
## 50 + AS 1000ppm 2.894025
## 50 + AS 600ppm 3.661525
## Control 9.882175
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 1.453312 1.521200 1.562333
##
## Means with the same letter are not significantly different.
##
## PFPA groups
## Control 9.284125 a
## 50 + AS 600ppm 3.253575 b
## 50 - AS 3.001075 b
## 50 + AS 1000ppm 2.876075 bA.PFPA.M3.DT## $statistics
## MSerror Df Mean CV
## 0.8898312 12 4.603713 20.49017
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 1.453312
## 3 3.225244 1.521200
## 4 3.312453 1.562333
##
## $means
## PFPA std r se Min Max Q25 Q50
## 50 - AS 3.001075 0.2689921 4 0.4716543 2.7498 3.3791 2.86170 2.9377
## 50 + AS 1000ppm 2.876075 0.2030149 4 0.4716543 2.7498 3.1791 2.76975 2.7877
## 50 + AS 600ppm 3.253575 0.8709956 4 0.4716543 2.1129 4.2142 2.93565 3.3436
## Control 9.284125 1.6392436 4 0.4716543 8.0370 11.5909 8.15625 8.7543
## Q75
## 50 - AS 3.077075
## 50 + AS 1000ppm 2.894025
## 50 + AS 600ppm 3.661525
## Control 9.882175
##
## $comparison
## NULL
##
## $groups
## PFPA groups
## Control 9.284125 a
## 50 + AS 600ppm 3.253575 b
## 50 - AS 3.001075 b
## 50 + AS 1000ppm 2.876075 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 4
# Datos - Peso fresco de la parte aérea - Muestreo 4
datos.PFPA.M4 <- data.frame(Var_M4$`Factor riego`, Var_M4$`Factor ácido ascórbico`, Var_M4$Tratamiento, Var_M4$`Peso fresco de parte aérea completo`)
colnames(datos.PFPA.M4) <- c("F_riego", "F_AA", "Tratamiento", "PFPA")
datos.PFPA.M4datos.PFPA.M4$F_riego <- as.factor(datos.PFPA.M4$F_riego)
datos.PFPA.M4$F_AA <- as.factor(datos.PFPA.M4$F_AA)
str(datos.PFPA.M4)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PFPA : num 11.87 12.21 11.87 9.73 2.74 ...# Anova - Peso fresco de la parte aérea - Muestreo 4
A.PFPA.M4 <- aov(PFPA ~ F_riego*F_AA, data = datos.PFPA.M4)
summary(A.PFPA.M4)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 180.55 180.55 353.653 2.86e-10 ***
## F_AA 2 4.46 2.23 4.367 0.0376 *
## Residuals 12 6.13 0.51
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Ambos facotores tuvieron diferencias significativas entre si. Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%, por otra parte, el factor concentración de ácido ascórbico tuvo una diferencia significativa del 95%.
# Prueba de normalidad de residuos
shapiro.test(A.PFPA.M4$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PFPA.M4$residuals
## W = 0.90233, p-value = 0.08759- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.PFPA.M4)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFPA ~ F_riego * F_AA, data = datos.PFPA.M4)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 7.75775 6.858941 8.656559 0
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.6200375 -0.5472754 1.787350 0.3634057
## 600 ppm-0 ppm 1.0359125 -0.1314004 2.203225 0.0841462
## 600 ppm-1000 ppm 0.4158750 -0.9320219 1.763772 0.6963960- Los dos tipos de riego son completamente distintos.
# Anova - PFPA - Muestreo 4 - Comparación entre los tratamientos
A.PFPA.M4.CT <- aov(PFPA ~ Tratamiento, data=datos.PFPA.M4)
summary(A.PFPA.M4.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 185.01 61.67 120.8 3.14e-09 ***
## Residuals 12 6.13 0.51
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.PFPA.M4.CT <- TukeyHSD(A.PFPA.M4.CT)
Tukey_A.PFPA.M4.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFPA ~ Tratamiento, data = datos.PFPA.M4)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 1.034025 -0.46596639 2.534016 0.2251381
## 50 + AS 600ppm-50 - AS 1.449900 -0.05009139 2.949891 0.0592994
## Control-50 - AS 8.585725 7.08573361 10.085716 0.0000000
## 50 + AS 600ppm-50 + AS 1000ppm 0.415875 -1.08411639 1.915866 0.8425046
## Control-50 + AS 1000ppm 7.551700 6.05170861 9.051691 0.0000000
## Control-50 + AS 600ppm 7.135825 5.63583361 8.635816 0.0000000- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.PFPA.M4.DT <- duncan.test(A.PFPA.M4.CT, 'Tratamiento', console = T)##
## Study: A.PFPA.M4.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PFPA
##
## Mean Square Error: 0.5105239
##
## Tratamiento, means
##
## PFPA std r se Min Max Q25
## 50 - AS 2.831875 0.2135311 4 0.3572548 2.5792 3.0437 2.696500
## 50 + AS 1000ppm 3.865900 0.6309622 4 0.3572548 2.9552 4.3659 3.699425
## 50 + AS 600ppm 4.281775 0.5510396 4 0.3572548 3.7227 5.0222 3.976650
## Control 11.417600 1.1378673 4 0.3572548 9.7275 12.2055 11.331075
## Q50 Q75
## 50 - AS 2.85230 2.987675
## 50 + AS 1000ppm 4.07125 4.237725
## 50 + AS 600ppm 4.19110 4.496225
## Control 11.86870 11.955225
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 1.100811 1.152234 1.183390
##
## Means with the same letter are not significantly different.
##
## PFPA groups
## Control 11.417600 a
## 50 + AS 600ppm 4.281775 b
## 50 + AS 1000ppm 3.865900 bc
## 50 - AS 2.831875 cA.PFPA.M4.DT## $statistics
## MSerror Df Mean CV
## 0.5105239 12 5.599287 12.76072
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 1.100811
## 3 3.225244 1.152234
## 4 3.312453 1.183390
##
## $means
## PFPA std r se Min Max Q25
## 50 - AS 2.831875 0.2135311 4 0.3572548 2.5792 3.0437 2.696500
## 50 + AS 1000ppm 3.865900 0.6309622 4 0.3572548 2.9552 4.3659 3.699425
## 50 + AS 600ppm 4.281775 0.5510396 4 0.3572548 3.7227 5.0222 3.976650
## Control 11.417600 1.1378673 4 0.3572548 9.7275 12.2055 11.331075
## Q50 Q75
## 50 - AS 2.85230 2.987675
## 50 + AS 1000ppm 4.07125 4.237725
## 50 + AS 600ppm 4.19110 4.496225
## Control 11.86870 11.955225
##
## $comparison
## NULL
##
## $groups
## PFPA groups
## Control 11.417600 a
## 50 + AS 600ppm 4.281775 b
## 50 + AS 1000ppm 3.865900 bc
## 50 - AS 2.831875 c
##
## attr(,"class")
## [1] "group"Análisis descriptivo - Peso fresco de la parte aérea
# Datos - Gráfico - Peso fresco de la parte aérea
Datos.G.PFPA <- data.frame(factor(Tratamientos, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), Var_M1$`Peso fresco de parte aérea completo`, Var_M2$`Peso fresco de parte aérea completo`, Var_M3$`Peso fresco de parte aérea completo`, Var_M4$`Peso fresco de parte aérea completo`)
colnames(Datos.G.PFPA) <- c("Tratamientos", "31", "38","45", "52")
Datos.G.PFPA# Unificar la variable respuesta en una sola columna
df.long.PFPA = gather(Datos.G.PFPA, dds, PFPA, 2:5)
df.long.PFPA# Diferencias significativas con el control
# M1 M2 M3 M4
DS.PFPA <- c("a", "a", "a", "a", #T1
"b*", "b*", "b*", "c*", #T2
"b*", "b*", "b*", "b*", #T3
"b*", "b*", "b*", "bc*") #T4
# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.PFPA = group_by(df.long.PFPA, Tratamientos, dds, ) %>% summarise(mean = mean(PFPA), sd = sd(PFPA))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.PFPA.ds <- data.frame(df.sumzd.PFPA, DS.PFPA)
df.sumzd.PFPA.dsG.PFPA = ggplot(df.sumzd.PFPA, aes(x=dds, y=mean, fill=Tratamientos))+
geom_bar(stat = "identity", position = "dodge", color = "black", alpha = 0.5)+
geom_errorbar(aes(ymin=mean-sd, ymax=mean+sd), width = 0.25, size=0.5, position = position_dodge(0.9), alpha = 0.5)+
scale_fill_manual(values=c("red", "yellow", "blue", "green")) +
theme_minimal()+
ggtitle("Peso fresco promedio de la parte aérea \nen los diferentes muestreos") + theme(plot.title = element_text(family = "Times New Roman",
size = rel(1.5),
vjust = 0.5,
hjust = 0.5))+
labs(x = "Días después de siembra", y = "g")+
theme(axis.title.x = element_text(family = "Times New Roman"), axis.title.y = element_text(family = "Times New Roman"))+
theme(legend.title = element_text(family = "Times New Roman"))+
geom_text(aes(x=dds, y= mean+2.2, label=DS.PFPA),
position = position_dodge(width = 1), size = 5)
G.PFPA
Peso seco de la parte aérea
Análisis estadístico - Muestreo 1
# Datos - Peso seco de la parte aérea - Muestreo 1
datos.PSPA.M1 <- data.frame(Var_M1$`Factor riego`, Var_M1$`Factor ácido ascórbico`, Var_M1$Tratamiento, Var_M1$`Peso seco parte aérea completo`)
colnames(datos.PSPA.M1) <- c("F_riego", "F_AA", "Tratamiento", "PSPA")
datos.PSPA.M1datos.PSPA.M1$F_riego <- as.factor(datos.PSPA.M1$F_riego)
datos.PSPA.M1$F_AA <- as.factor(datos.PSPA.M1$F_AA)
str(datos.PSPA.M1)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PSPA : num 0.378 0.52 0.287 0.428 0.287 ...# Anova - Peso seco de la parte aérea - Muestreo 1
A.PSPA.M1 <- aov(PSPA ~ F_riego*F_AA, data = datos.PSPA.M1)
summary(A.PSPA.M1)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 0.04507 0.04507 7.086 0.0207 *
## F_AA 2 0.00390 0.00195 0.306 0.7417
## Residuals 12 0.07633 0.00636
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 95%.
# Prueba de normalidad de residuos
shapiro.test(A.PSPA.M1$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PSPA.M1$residuals
## W = 0.96391, p-value = 0.733- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.PSPA.M1)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSPA ~ F_riego * F_AA, data = datos.PSPA.M1)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.1225667 0.02224301 0.2228903 0.0207192
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -0.02065417 -0.1509478 0.1096395 0.9068635
## 600 ppm-0 ppm -0.02932917 -0.1596228 0.1009645 0.8224679
## 600 ppm-1000 ppm -0.00867500 -0.1591252 0.1417752 0.9870530- Los dos tipos de riego presenta diferencias significativas.
# Anova - PSPA - Muestreo 1 - Comparación entre los tratamientos
A.PSPA.M1.CT <- aov(PSPA ~ Tratamiento, data=datos.PSPA.M1)
summary(A.PSPA.M1.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 0.04897 0.01632 2.566 0.103
## Residuals 12 0.07633 0.00636Tukey_A.PSPA.M1.CT <- TukeyHSD(A.PSPA.M1.CT)
Tukey_A.PSPA.M1.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSPA ~ Tratamiento, data = datos.PSPA.M1)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -0.033150 -0.20057674 0.1342767 0.9338504
## 50 + AS 600ppm-50 - AS -0.041825 -0.20925174 0.1256017 0.8784252
## Control-50 - AS 0.097575 -0.06985174 0.2650017 0.3511000
## 50 + AS 600ppm-50 + AS 1000ppm -0.008675 -0.17610174 0.1587517 0.9986280
## Control-50 + AS 1000ppm 0.130725 -0.03670174 0.2981517 0.1482258
## Control-50 + AS 600ppm 0.139400 -0.02802674 0.3068267 0.1156512- Podemos observar que ninguno de los tratamientos tuvieron diferencias significativas con el control.
A.PSPA.M1.DT <- duncan.test(A.PSPA.M1.CT, 'Tratamiento', console = T)##
## Study: A.PSPA.M1.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PSPA
##
## Mean Square Error: 0.006360455
##
## Tratamiento, means
##
## PSPA std r se Min Max Q25 Q50
## 50 - AS 0.305525 0.04137257 4 0.03987623 0.2804 0.3674 0.285425 0.28715
## 50 + AS 1000ppm 0.272375 0.10813557 4 0.03987623 0.1469 0.3681 0.199850 0.28725
## 50 + AS 600ppm 0.263700 0.05053296 4 0.03987623 0.1999 0.3084 0.235000 0.27325
## Control 0.403100 0.09738196 4 0.03987623 0.2867 0.5200 0.355175 0.40285
## Q75
## 50 - AS 0.307250
## 50 + AS 1000ppm 0.359775
## 50 + AS 600ppm 0.301950
## Control 0.450775
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 0.1228709 0.1286106 0.1320881
##
## Means with the same letter are not significantly different.
##
## PSPA groups
## Control 0.403100 a
## 50 - AS 0.305525 ab
## 50 + AS 1000ppm 0.272375 b
## 50 + AS 600ppm 0.263700 bA.PSPA.M1.DT## $statistics
## MSerror Df Mean CV
## 0.006360455 12 0.311175 25.62946
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 0.1228709
## 3 3.225244 0.1286106
## 4 3.312453 0.1320881
##
## $means
## PSPA std r se Min Max Q25 Q50
## 50 - AS 0.305525 0.04137257 4 0.03987623 0.2804 0.3674 0.285425 0.28715
## 50 + AS 1000ppm 0.272375 0.10813557 4 0.03987623 0.1469 0.3681 0.199850 0.28725
## 50 + AS 600ppm 0.263700 0.05053296 4 0.03987623 0.1999 0.3084 0.235000 0.27325
## Control 0.403100 0.09738196 4 0.03987623 0.2867 0.5200 0.355175 0.40285
## Q75
## 50 - AS 0.307250
## 50 + AS 1000ppm 0.359775
## 50 + AS 600ppm 0.301950
## Control 0.450775
##
## $comparison
## NULL
##
## $groups
## PSPA groups
## Control 0.403100 a
## 50 - AS 0.305525 ab
## 50 + AS 1000ppm 0.272375 b
## 50 + AS 600ppm 0.263700 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 2
# Datos - Peso seco de la parte aérea - Muestreo 2
datos.PSPA.M2 <- data.frame(Var_M2$`Factor riego`, Var_M2$`Factor ácido ascórbico`, Var_M2$Tratamiento, Var_M2$`Peso seco parte aérea completo`)
colnames(datos.PSPA.M2) <- c("F_riego", "F_AA", "Tratamiento", "PSPA")
datos.PSPA.M2datos.PSPA.M2$F_riego <- as.factor(datos.PSPA.M2$F_riego)
datos.PSPA.M2$F_AA <- as.factor(datos.PSPA.M2$F_AA)
str(datos.PSPA.M2)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PSPA : num 0.508 0.661 0.75 0.77 0.481 ...# Anova - Peso seco de la parte aérea - Muestreo 2
A.PSPA.M2 <- aov(PSPA ~ F_riego*F_AA, data = datos.PSPA.M2)
summary(A.PSPA.M2)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 0.19035 0.19035 12.609 0.00399 **
## F_AA 2 0.00449 0.00225 0.149 0.86337
## Residuals 12 0.18115 0.01510
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99%.
# Prueba de normalidad de residuos
shapiro.test(A.PSPA.M2$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PSPA.M2$residuals
## W = 0.94341, p-value = 0.3929- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.PSPA.M2)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSPA ~ F_riego * F_AA, data = datos.PSPA.M2)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.2518917 0.09733484 0.4064485 0.0039896
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.02232083 -0.1784073 0.2230489 0.9528318
## 600 ppm-0 ppm -0.02500417 -0.2257323 0.1757239 0.9412307
## 600 ppm-1000 ppm -0.04732500 -0.2791058 0.1844558 0.8510457- Los dos tipos de riego presenta diferencias significativas.
# Anova - PSPA - Muestreo 2 - Comparación entre los tratamientos
A.PSPA.M2.CT <- aov(PSPA ~ Tratamiento, data=datos.PSPA.M2)
summary(A.PSPA.M2.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 0.1948 0.06495 4.302 0.0281 *
## Residuals 12 0.1812 0.01510
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.PSPA.M2.CT <- TukeyHSD(A.PSPA.M2.CT)
Tukey_A.PSPA.M2.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSPA ~ Tratamiento, data = datos.PSPA.M2)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 0.021650 -0.236284614 0.2795846 0.9942709
## 50 + AS 600ppm-50 - AS -0.025675 -0.283609614 0.2322596 0.9905556
## Control-50 - AS 0.250550 -0.007384614 0.5084846 0.0578780
## 50 + AS 600ppm-50 + AS 1000ppm -0.047325 -0.305259614 0.2106096 0.9462109
## Control-50 + AS 1000ppm 0.228900 -0.029034614 0.4868346 0.0883326
## Control-50 + AS 600ppm 0.276225 0.018290386 0.5341596 0.0346927- Podemos observar que solo el tratamiento azul (50 + AS 600ppm), presento diferencias significativas al tener un p-valor inferior al 5%. Los demás tratamientos no tuvieron diferencias significativas, sin embargo, sus p-valores son lo suficientemente bajos comos para ser considerados agronómicamente.
A.PSPA.M2.DT <- duncan.test(A.PSPA.M2.CT, 'Tratamiento', console = T)##
## Study: A.PSPA.M2.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PSPA
##
## Mean Square Error: 0.01509586
##
## Tratamiento, means
##
## PSPA std r se Min Max Q25 Q50
## 50 - AS 0.421800 0.1349560 4 0.0614326 0.2210 0.5125 0.409925 0.47685
## 50 + AS 1000ppm 0.443450 0.1294538 4 0.0614326 0.2783 0.5942 0.400625 0.45065
## 50 + AS 600ppm 0.396125 0.1058534 4 0.0614326 0.3137 0.5495 0.334100 0.36065
## Control 0.672350 0.1191934 4 0.0614326 0.5084 0.7703 0.622700 0.70535
## Q75
## 50 - AS 0.488725
## 50 + AS 1000ppm 0.493475
## 50 + AS 600ppm 0.422675
## Control 0.755000
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 0.1892927 0.1981351 0.2034926
##
## Means with the same letter are not significantly different.
##
## PSPA groups
## Control 0.672350 a
## 50 + AS 1000ppm 0.443450 b
## 50 - AS 0.421800 b
## 50 + AS 600ppm 0.396125 bA.PSPA.M2.DT## $statistics
## MSerror Df Mean CV
## 0.01509586 12 0.4834312 25.41524
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 0.1892927
## 3 3.225244 0.1981351
## 4 3.312453 0.2034926
##
## $means
## PSPA std r se Min Max Q25 Q50
## 50 - AS 0.421800 0.1349560 4 0.0614326 0.2210 0.5125 0.409925 0.47685
## 50 + AS 1000ppm 0.443450 0.1294538 4 0.0614326 0.2783 0.5942 0.400625 0.45065
## 50 + AS 600ppm 0.396125 0.1058534 4 0.0614326 0.3137 0.5495 0.334100 0.36065
## Control 0.672350 0.1191934 4 0.0614326 0.5084 0.7703 0.622700 0.70535
## Q75
## 50 - AS 0.488725
## 50 + AS 1000ppm 0.493475
## 50 + AS 600ppm 0.422675
## Control 0.755000
##
## $comparison
## NULL
##
## $groups
## PSPA groups
## Control 0.672350 a
## 50 + AS 1000ppm 0.443450 b
## 50 - AS 0.421800 b
## 50 + AS 600ppm 0.396125 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 3
# Datos - Peso seco de la parte aérea - Muestreo 3
datos.PSPA.M3 <- data.frame(Var_M3$`Factor riego`, Var_M3$`Factor ácido ascórbico`, Var_M3$Tratamiento, Var_M3$`Peso seco parte aérea completo`)
colnames(datos.PSPA.M3) <- c("F_riego", "F_AA", "Tratamiento", "PSPA")
datos.PSPA.M3datos.PSPA.M3$F_riego <- as.factor(datos.PSPA.M3$F_riego)
datos.PSPA.M3$F_AA <- as.factor(datos.PSPA.M3$F_AA)
str(datos.PSPA.M3)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PSPA : num 0.98 0.932 1.301 0.791 0.546 ...# Anova - Peso seco de la parte aérea - Muestreo 3
A.PSPA.M3 <- aov(PSPA ~ F_riego*F_AA, data = datos.PSPA.M3)
summary(A.PSPA.M3)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 0.8544 0.8544 49.03 1.43e-05 ***
## F_AA 2 0.0014 0.0007 0.04 0.96
## Residuals 12 0.2091 0.0174
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.PSPA.M3$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PSPA.M3$residuals
## W = 0.90528, p-value = 0.09768- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.PSPA.M3)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSPA ~ F_riego * F_AA, data = datos.PSPA.M3)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.5336667 0.3676147 0.6997187 1.43e-05
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.01270833 -0.2029489 0.2283656 0.9864819
## 600 ppm-0 ppm 0.01745833 -0.1981989 0.2331156 0.9746676
## 600 ppm-1000 ppm 0.00475000 -0.2442696 0.2537696 0.9985734- Los dos tipos de riego presenta diferencias significativas.
# Anova - PSPA - Muestreo 3 - Comparación entre los tratamientos
A.PSPA.M3.CT <- aov(PSPA ~ Tratamiento, data=datos.PSPA.M3)
summary(A.PSPA.M3.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 0.8558 0.28527 16.37 0.000153 ***
## Residuals 12 0.2091 0.01742
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.PSPA.M3.CT <- TukeyHSD(A.PSPA.M3.CT)
Tukey_A.PSPA.M3.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSPA ~ Tratamiento, data = datos.PSPA.M3)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 0.02025 -0.2568685 0.2973685 0.9961933
## 50 + AS 600ppm-50 - AS 0.02500 -0.2521185 0.3021185 0.9929184
## Control-50 - AS 0.54875 0.2716315 0.8258685 0.0003770
## 50 + AS 600ppm-50 + AS 1000ppm 0.00475 -0.2723685 0.2818685 0.9999498
## Control-50 + AS 1000ppm 0.52850 0.2513815 0.8056185 0.0005270
## Control-50 + AS 600ppm 0.52375 0.2466315 0.8008685 0.0005705- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.PSPA.M3.DT <- duncan.test(A.PSPA.M3.CT, 'Tratamiento', console = T)##
## Study: A.PSPA.M3.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PSPA
##
## Mean Square Error: 0.01742487
##
## Tratamiento, means
##
## PSPA std r se Min Max Q25 Q50
## 50 - AS 0.45225 0.06668521 4 0.06600166 0.396 0.546 0.41025 0.4335
## 50 + AS 1000ppm 0.47250 0.02224860 4 0.06600166 0.443 0.497 0.46625 0.4750
## 50 + AS 600ppm 0.47725 0.13536463 4 0.06600166 0.305 0.636 0.43775 0.4840
## Control 1.00100 0.21548550 4 0.06600166 0.791 1.301 0.89675 0.9560
## Q75
## 50 - AS 0.47550
## 50 + AS 1000ppm 0.48125
## 50 + AS 600ppm 0.52350
## Control 1.06025
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 0.2033713 0.2128714 0.2186274
##
## Means with the same letter are not significantly different.
##
## PSPA groups
## Control 1.00100 a
## 50 + AS 600ppm 0.47725 b
## 50 + AS 1000ppm 0.47250 b
## 50 - AS 0.45225 bA.PSPA.M3.DT## $statistics
## MSerror Df Mean CV
## 0.01742487 12 0.60075 21.97309
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 0.2033713
## 3 3.225244 0.2128714
## 4 3.312453 0.2186274
##
## $means
## PSPA std r se Min Max Q25 Q50
## 50 - AS 0.45225 0.06668521 4 0.06600166 0.396 0.546 0.41025 0.4335
## 50 + AS 1000ppm 0.47250 0.02224860 4 0.06600166 0.443 0.497 0.46625 0.4750
## 50 + AS 600ppm 0.47725 0.13536463 4 0.06600166 0.305 0.636 0.43775 0.4840
## Control 1.00100 0.21548550 4 0.06600166 0.791 1.301 0.89675 0.9560
## Q75
## 50 - AS 0.47550
## 50 + AS 1000ppm 0.48125
## 50 + AS 600ppm 0.52350
## Control 1.06025
##
## $comparison
## NULL
##
## $groups
## PSPA groups
## Control 1.00100 a
## 50 + AS 600ppm 0.47725 b
## 50 + AS 1000ppm 0.47250 b
## 50 - AS 0.45225 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 4
# Datos - Peso seco de la parte aérea - Muestreo 4
datos.PSPA.M4 <- data.frame(Var_M4$`Factor riego`, Var_M4$`Factor ácido ascórbico`, Var_M4$Tratamiento, Var_M4$`Peso seco parte aérea completo`)
colnames(datos.PSPA.M4) <- c("F_riego", "F_AA", "Tratamiento", "PSPA")
datos.PSPA.M4datos.PSPA.M4$F_riego <- as.factor(datos.PSPA.M4$F_riego)
datos.PSPA.M4$F_AA <- as.factor(datos.PSPA.M4$F_AA)
str(datos.PSPA.M4)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PSPA : num 1.358 1.784 1.786 1.171 0.401 ...# Anova - Peso seco de la parte aérea - Muestreo 4
A.PSPA.M4 <- aov(PSPA ~ F_riego*F_AA, data = datos.PSPA.M4)
summary(A.PSPA.M4)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 2.7563 2.7563 86.702 7.7e-07 ***
## F_AA 2 0.0831 0.0415 1.307 0.307
## Residuals 12 0.3815 0.0318
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.PSPA.M4$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PSPA.M4$residuals
## W = 0.96996, p-value = 0.838- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.PSPA.M4)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSPA ~ F_riego * F_AA, data = datos.PSPA.M4)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.9585167 0.7342288 1.182805 8e-07
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.08987083 -0.2014193 0.3811610 0.6964108
## 600 ppm-0 ppm 0.13864583 -0.1526443 0.4299360 0.4376326
## 600 ppm-1000 ppm 0.04877500 -0.2875779 0.3851279 0.9213411- Los dos tipos de riego son completamente distintos.
# Anova - PSPA - Muestreo 4 - Comparación entre los tratamientos
A.PSPA.M4.CT <- aov(PSPA ~ Tratamiento, data=datos.PSPA.M4)
summary(A.PSPA.M4.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 2.8394 0.9465 29.77 7.67e-06 ***
## Residuals 12 0.3815 0.0318
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.PSPA.M4.CT <- TukeyHSD(A.PSPA.M4.CT)
Tukey_A.PSPA.M4.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSPA ~ Tratamiento, data = datos.PSPA.M4)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 0.147000 -0.2273064 0.5213064 0.6581575
## 50 + AS 600ppm-50 - AS 0.195775 -0.1785314 0.5700814 0.4389475
## Control-50 - AS 1.072775 0.6984686 1.4470814 0.0000103
## 50 + AS 600ppm-50 + AS 1000ppm 0.048775 -0.3255314 0.4230814 0.9794101
## Control-50 + AS 1000ppm 0.925775 0.5514686 1.3000814 0.0000460
## Control-50 + AS 600ppm 0.877000 0.5026936 1.2513064 0.0000782- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.PSPA.M4.DT <- duncan.test(A.PSPA.M4.CT, 'Tratamiento', console = T)##
## Study: A.PSPA.M4.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PSPA
##
## Mean Square Error: 0.03179019
##
## Tratamiento, means
##
## PSPA std r se Min Max Q25 Q50
## 50 - AS 0.452200 0.06563541 4 0.08914902 0.3922 0.5242 0.398875 0.44620
## 50 + AS 1000ppm 0.599200 0.12282150 4 0.08914902 0.4308 0.7008 0.547200 0.63260
## 50 + AS 600ppm 0.647975 0.10733304 4 0.08914902 0.5440 0.7795 0.570325 0.63420
## Control 1.524975 0.31023741 4 0.08914902 1.1709 1.7863 1.311450 1.57135
## Q75
## 50 - AS 0.499525
## 50 + AS 1000ppm 0.684600
## 50 + AS 600ppm 0.711850
## Control 1.784875
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 0.2746955 0.2875273 0.2953019
##
## Means with the same letter are not significantly different.
##
## PSPA groups
## Control 1.524975 a
## 50 + AS 600ppm 0.647975 b
## 50 + AS 1000ppm 0.599200 b
## 50 - AS 0.452200 bA.PSPA.M4.DT## $statistics
## MSerror Df Mean CV
## 0.03179019 12 0.8060875 22.11894
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 0.2746955
## 3 3.225244 0.2875273
## 4 3.312453 0.2953019
##
## $means
## PSPA std r se Min Max Q25 Q50
## 50 - AS 0.452200 0.06563541 4 0.08914902 0.3922 0.5242 0.398875 0.44620
## 50 + AS 1000ppm 0.599200 0.12282150 4 0.08914902 0.4308 0.7008 0.547200 0.63260
## 50 + AS 600ppm 0.647975 0.10733304 4 0.08914902 0.5440 0.7795 0.570325 0.63420
## Control 1.524975 0.31023741 4 0.08914902 1.1709 1.7863 1.311450 1.57135
## Q75
## 50 - AS 0.499525
## 50 + AS 1000ppm 0.684600
## 50 + AS 600ppm 0.711850
## Control 1.784875
##
## $comparison
## NULL
##
## $groups
## PSPA groups
## Control 1.524975 a
## 50 + AS 600ppm 0.647975 b
## 50 + AS 1000ppm 0.599200 b
## 50 - AS 0.452200 b
##
## attr(,"class")
## [1] "group"Análisis descriptivo - Peso fresco de la parte aérea
# Datos - Gráfico - Peso seco de la parte aérea
Datos.G.PSPA <- data.frame(factor(Tratamientos, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), Var_M1$`Peso seco parte aérea completo`, Var_M2$`Peso seco parte aérea completo`, Var_M3$`Peso seco parte aérea completo`, Var_M4$`Peso seco parte aérea completo`)
colnames(Datos.G.PSPA) <- c("Tratamientos", "31", "38","45", "52")
Datos.G.PSPA# Unificar la variable respuesta en una sola columna
df.long.PSPA = gather(Datos.G.PSPA, dds, PSPA, 2:5)
df.long.PSPA# Diferencias significativas con el control
# M1 M2 M3 M4
DS.PSPA <- c("a", "a", "a", "a", #T1
"ab", "b*", "b*", "b*", #T2
"b", "b*", "b*", "b*", #T3
"b", "b", "b*", "b*") #T4
# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.PSPA = group_by(df.long.PSPA, Tratamientos, dds, ) %>% summarise(mean = mean(PSPA), sd = sd(PSPA))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.PSPA.ds <- data.frame(df.sumzd.PSPA, DS.PSPA)
df.sumzd.PSPA.dsG.PSPA = ggplot(df.sumzd.PSPA, aes(x=dds, y=mean, fill=Tratamientos))+
geom_bar(stat = "identity", position = "dodge", color = "black", alpha = 0.5)+
geom_errorbar(aes(ymin=mean-sd, ymax=mean+sd), width = 0.25, size=0.5, position = position_dodge(0.9), alpha = 0.5)+
scale_fill_manual(values=c("red", "yellow", "blue", "green")) +
theme_minimal()+
ggtitle("Peso seco promedio de la parte aérea \nen los diferentes muestreos") + theme(plot.title = element_text(family = "Times New Roman",
size = rel(1.5),
vjust = 0.5,
hjust = 0.5))+
labs(x = "Días después de siembra", y = "g")+
theme(axis.title.x = element_text(family = "Times New Roman"), axis.title.y = element_text(family = "Times New Roman"))+
theme(legend.title = element_text(family = "Times New Roman"))+
geom_text(aes(x=dds, y= mean+0.4, label=DS.PSPA),
position = position_dodge(width = 1), size = 5)
G.PSPA
Peso fresco de la raíz tuberosa
Análisis estadístico - Muestreo 1
# Datos - Peso fresco de la raíz tuberosa - Muestreo 1
datos.PFRT.M1 <- data.frame(Var_M1$`Factor riego`, Var_M1$`Factor ácido ascórbico`, Var_M1$Tratamiento, Var_M1$`Peso fresco raíz tuberosa`)
colnames(datos.PFRT.M1) <- c("F_riego", "F_AA", "Tratamiento", "PFRT")
datos.PFRT.M1datos.PFRT.M1$F_riego <- as.factor(datos.PFRT.M1$F_riego)
datos.PFRT.M1$F_AA <- as.factor(datos.PFRT.M1$F_AA)
str(datos.PFRT.M1)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PFRT : num 7.78 7.53 7.96 8.42 4.13 3.93 2.39 3.87 4.01 4.65 ...# Anova - Peso fresco de la raíz tuberosa - Muestreo 1
A.PFRT.M1 <- aov(PFRT ~ F_riego*F_AA, data = datos.PFRT.M1)
summary(A.PFRT.M1)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 47.28 47.28 64.644 3.57e-06 ***
## F_AA 2 2.76 1.38 1.889 0.194
## Residuals 12 8.78 0.73
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.PFRT.M1$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PFRT.M1$residuals
## W = 0.9621, p-value = 0.7001- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.PFRT.M1)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFRT ~ F_riego * F_AA, data = datos.PFRT.M1)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 3.97 2.894166 5.045834 3.6e-06
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.86375 -0.533472 2.2609720 0.2638566
## 600 ppm-0 ppm -0.11875 -1.515972 1.2784720 0.9721214
## 600 ppm-1000 ppm -0.98250 -2.595873 0.6308729 0.2733459- Los dos tipos de riego presenta diferencias significativas.
# Anova - PFRT - Muestreo 1 - Comparación entre los tratamientos
A.PFRT.M1.CT <- aov(PFRT ~ Tratamiento, data=datos.PFRT.M1)
summary(A.PFRT.M1.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 50.05 16.682 22.81 3.02e-05 ***
## Residuals 12 8.78 0.731
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.PFRT.M1.CT <- TukeyHSD(A.PFRT.M1.CT)
Tukey_A.PFRT.M1.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFRT ~ Tratamiento, data = datos.PFRT.M1)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 1.0500 -0.7454234 2.8454234 0.3483248
## 50 + AS 600ppm-50 - AS 0.0675 -1.7279234 1.8629234 0.9994731
## Control-50 - AS 4.3425 2.5470766 6.1379234 0.0000573
## 50 + AS 600ppm-50 + AS 1000ppm -0.9825 -2.7779234 0.8129234 0.4020239
## Control-50 + AS 1000ppm 3.2925 1.4970766 5.0879234 0.0007419
## Control-50 + AS 600ppm 4.2750 2.4795766 6.0704234 0.0000668- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.PFRT.M1.DT <- duncan.test(A.PFRT.M1.CT, 'Tratamiento', console = T)##
## Study: A.PFRT.M1.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PFRT
##
## Mean Square Error: 0.7314292
##
## Tratamiento, means
##
## PFRT std r se Min Max Q25 Q50 Q75
## 50 - AS 3.5800 0.8010826 4 0.4276182 2.39 4.13 3.5000 3.900 3.980
## 50 + AS 1000ppm 4.6300 1.0312129 4 0.4276182 3.59 5.98 3.9950 4.475 5.110
## 50 + AS 600ppm 3.6475 1.0389859 4 0.4276182 2.20 4.65 3.3475 3.870 4.170
## Control 7.9225 0.3756217 4 0.4276182 7.53 8.42 7.7175 7.870 8.075
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 1.317623 1.379173 1.416465
##
## Means with the same letter are not significantly different.
##
## PFRT groups
## Control 7.9225 a
## 50 + AS 1000ppm 4.6300 b
## 50 + AS 600ppm 3.6475 b
## 50 - AS 3.5800 bA.PFRT.M1.DT## $statistics
## MSerror Df Mean CV
## 0.7314292 12 4.945 17.29497
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 1.317623
## 3 3.225244 1.379173
## 4 3.312453 1.416465
##
## $means
## PFRT std r se Min Max Q25 Q50 Q75
## 50 - AS 3.5800 0.8010826 4 0.4276182 2.39 4.13 3.5000 3.900 3.980
## 50 + AS 1000ppm 4.6300 1.0312129 4 0.4276182 3.59 5.98 3.9950 4.475 5.110
## 50 + AS 600ppm 3.6475 1.0389859 4 0.4276182 2.20 4.65 3.3475 3.870 4.170
## Control 7.9225 0.3756217 4 0.4276182 7.53 8.42 7.7175 7.870 8.075
##
## $comparison
## NULL
##
## $groups
## PFRT groups
## Control 7.9225 a
## 50 + AS 1000ppm 4.6300 b
## 50 + AS 600ppm 3.6475 b
## 50 - AS 3.5800 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 2
# Datos - Peso fresco de la raíz tuberosa - Muestreo 2
datos.PFRT.M2 <- data.frame(Var_M2$`Factor riego`, Var_M2$`Factor ácido ascórbico`, Var_M2$Tratamiento, Var_M2$`Peso fresco raíz tuberosa`)
colnames(datos.PFRT.M2) <- c("F_riego", "F_AA", "Tratamiento", "PFRT")
datos.PFRT.M2datos.PFRT.M2$F_riego <- as.factor(datos.PFRT.M2$F_riego)
datos.PFRT.M2$F_AA <- as.factor(datos.PFRT.M2$F_AA)
str(datos.PFRT.M2)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PFRT : num 10.76 21.05 27.99 23.94 9.73 ...# Anova - Peso fresco de la raíz tuberosa - Muestreo 2
A.PFRT.M2 <- aov(PFRT ~ F_riego*F_AA, data = datos.PFRT.M2)
summary(A.PFRT.M2)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 450.1 450.1 23.440 0.000404 ***
## F_AA 2 3.5 1.8 0.092 0.912489
## Residuals 12 230.4 19.2
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.PFRT.M2$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PFRT.M2$residuals
## W = 0.93201, p-value = 0.2623- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.PFRT.M2)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFRT ~ F_riego * F_AA, data = datos.PFRT.M2)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 12.24917 6.736659 17.76167 0.000404
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -1.0166667 -8.175944 6.142611 0.9244204
## 600 ppm-0 ppm -0.1566667 -7.315944 7.002611 0.9981230
## 600 ppm-1000 ppm 0.8600000 -7.406822 9.126822 0.9585725- Los dos tipos de riego presenta diferencias significativas.
# Anova - PFRT - Muestreo 2 - Comparación entre los tratamientos
A.PFRT.M2.CT <- aov(PFRT ~ Tratamiento, data=datos.PFRT.M2)
summary(A.PFRT.M2.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 453.7 151.2 7.875 0.00361 **
## Residuals 12 230.4 19.2
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.PFRT.M2.CT <- TukeyHSD(A.PFRT.M2.CT)
Tukey_A.PFRT.M2.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFRT ~ Tratamiento, data = datos.PFRT.M2)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -1.3100 -10.509637 7.889637 0.9734845
## 50 + AS 600ppm-50 - AS -0.4500 -9.649637 8.749637 0.9988443
## Control-50 - AS 11.6625 2.462863 20.862137 0.0124718
## 50 + AS 600ppm-50 + AS 1000ppm 0.8600 -8.339637 10.059637 0.9921399
## Control-50 + AS 1000ppm 12.9725 3.772863 22.172137 0.0059832
## Control-50 + AS 600ppm 12.1125 2.912863 21.312137 0.0096794- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.PFRT.M2.DT <- duncan.test(A.PFRT.M2.CT, 'Tratamiento', console = T)##
## Study: A.PFRT.M2.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PFRT
##
## Mean Square Error: 19.20348
##
## Tratamiento, means
##
## PFRT std r se Min Max Q25 Q50 Q75
## 50 - AS 9.2725 1.869409 4 2.191089 6.67 11.12 8.8450 9.650 10.0775
## 50 + AS 1000ppm 7.9625 4.142402 4 2.191089 2.57 11.98 5.9075 8.650 10.7050
## 50 + AS 600ppm 8.8225 1.429717 4 2.191089 7.33 10.11 7.7350 8.925 10.0125
## Control 20.9350 7.356333 4 2.191089 10.76 27.99 18.4775 22.495 24.9525
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 6.751416 7.066794 7.257878
##
## Means with the same letter are not significantly different.
##
## PFRT groups
## Control 20.9350 a
## 50 - AS 9.2725 b
## 50 + AS 600ppm 8.8225 b
## 50 + AS 1000ppm 7.9625 bA.PFRT.M2.DT## $statistics
## MSerror Df Mean CV
## 19.20348 12 11.74812 37.30108
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 6.751416
## 3 3.225244 7.066794
## 4 3.312453 7.257878
##
## $means
## PFRT std r se Min Max Q25 Q50 Q75
## 50 - AS 9.2725 1.869409 4 2.191089 6.67 11.12 8.8450 9.650 10.0775
## 50 + AS 1000ppm 7.9625 4.142402 4 2.191089 2.57 11.98 5.9075 8.650 10.7050
## 50 + AS 600ppm 8.8225 1.429717 4 2.191089 7.33 10.11 7.7350 8.925 10.0125
## Control 20.9350 7.356333 4 2.191089 10.76 27.99 18.4775 22.495 24.9525
##
## $comparison
## NULL
##
## $groups
## PFRT groups
## Control 20.9350 a
## 50 - AS 9.2725 b
## 50 + AS 600ppm 8.8225 b
## 50 + AS 1000ppm 7.9625 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 3
# Datos - Peso fresco de la raíz tuberosa - Muestreo 3
datos.PFRT.M3 <- data.frame(Var_M3$`Factor riego`, Var_M3$`Factor ácido ascórbico`, Var_M3$Tratamiento, Var_M3$`Peso fresco raíz tuberosa`)
colnames(datos.PFRT.M3) <- c("F_riego", "F_AA", "Tratamiento", "PFRT")
datos.PFRT.M3datos.PFRT.M3$F_riego <- as.factor(datos.PFRT.M3$F_riego)
datos.PFRT.M3$F_AA <- as.factor(datos.PFRT.M3$F_AA)
str(datos.PFRT.M3)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PFRT : num 24.64 30.45 14.62 25.58 4.98 ...# Anova - Peso fresco de la raíz tuberosa - Muestreo 3
A.PFRT.M3 <- aov(PFRT ~ F_riego*F_AA, data = datos.PFRT.M3)
summary(A.PFRT.M3)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 980.1 980.1 87.288 7.43e-07 ***
## F_AA 2 3.0 1.5 0.134 0.876
## Residuals 12 134.7 11.2
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.PFRT.M3$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PFRT.M3$residuals
## W = 0.68506, p-value = 0.0001161- Los datos tienen una distribución normal, por lo tanto, se debe aplicar otra prueba que soporte la distribución de estos datos para eventualmente, analizar la residualidad, sin embargo, en esta ocasión no se realizará ya que sabemos, desde la toma de los datos de esta variable, que estos datos no iban a presentar diferencias significativas importantes.
TukeyHSD(A.PFRT.M3)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFRT ~ F_riego * F_AA, data = datos.PFRT.M3)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 18.075 13.85977 22.29023 7e-07
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.79 -4.684456 6.264456 0.9220681
## 600 ppm-0 ppm 0.61 -4.864456 6.084456 0.9526442
## 600 ppm-1000 ppm -0.18 -6.501357 6.141357 0.9968242- Los dos tipos de riego presenta diferencias significativas.
# Anova - PFRT - Muestreo 3 - Comparación entre los tratamientos
A.PFRT.M3.CT <- aov(PFRT ~ Tratamiento, data=datos.PFRT.M3)
summary(A.PFRT.M3.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 983.1 327.7 29.18 8.52e-06 ***
## Residuals 12 134.7 11.2
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.PFRT.M3.CT <- TukeyHSD(A.PFRT.M3.CT)
Tukey_A.PFRT.M3.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFRT ~ Tratamiento, data = datos.PFRT.M3)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 1.140 -5.894649 8.174649 0.9618387
## 50 + AS 600ppm-50 - AS 0.960 -6.074649 7.994649 0.9765057
## Control-50 - AS 18.775 11.740351 25.809649 0.0000215
## 50 + AS 600ppm-50 + AS 1000ppm -0.180 -7.214649 6.854649 0.9998334
## Control-50 + AS 1000ppm 17.635 10.600351 24.669649 0.0000403
## Control-50 + AS 600ppm 17.815 10.780351 24.849649 0.0000364- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.PFRT.M3.DT <- duncan.test(A.PFRT.M3.CT, 'Tratamiento', console = T)##
## Study: A.PFRT.M3.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PFRT
##
## Mean Square Error: 11.22854
##
## Tratamiento, means
##
## PFRT std r se Min Max Q25 Q50 Q75
## 50 - AS 5.0475 0.4729605 4 1.675451 4.42 5.52 4.840 5.125 5.3325
## 50 + AS 1000ppm 6.1875 0.4771705 4 1.675451 5.64 6.64 5.865 6.235 6.5575
## 50 + AS 600ppm 6.0075 0.5835166 4 1.675451 5.28 6.67 5.730 6.040 6.3175
## Control 23.8225 6.6424613 4 1.675451 14.62 30.45 22.135 25.110 26.7975
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 5.162578 5.403737 5.549852
##
## Means with the same letter are not significantly different.
##
## PFRT groups
## Control 23.8225 a
## 50 + AS 1000ppm 6.1875 b
## 50 + AS 600ppm 6.0075 b
## 50 - AS 5.0475 bA.PFRT.M3.DT## $statistics
## MSerror Df Mean CV
## 11.22854 12 10.26625 32.63998
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 5.162578
## 3 3.225244 5.403737
## 4 3.312453 5.549852
##
## $means
## PFRT std r se Min Max Q25 Q50 Q75
## 50 - AS 5.0475 0.4729605 4 1.675451 4.42 5.52 4.840 5.125 5.3325
## 50 + AS 1000ppm 6.1875 0.4771705 4 1.675451 5.64 6.64 5.865 6.235 6.5575
## 50 + AS 600ppm 6.0075 0.5835166 4 1.675451 5.28 6.67 5.730 6.040 6.3175
## Control 23.8225 6.6424613 4 1.675451 14.62 30.45 22.135 25.110 26.7975
##
## $comparison
## NULL
##
## $groups
## PFRT groups
## Control 23.8225 a
## 50 + AS 1000ppm 6.1875 b
## 50 + AS 600ppm 6.0075 b
## 50 - AS 5.0475 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 4
# Datos - Peso fresco de la raíz tuberosa - Muestreo 4
datos.PFRT.M4 <- data.frame(Var_M4$`Factor riego`, Var_M4$`Factor ácido ascórbico`, Var_M4$Tratamiento, Var_M4$`Peso fresco raíz tuberosa`)
colnames(datos.PFRT.M4) <- c("F_riego", "F_AA", "Tratamiento", "PFRT")
datos.PFRT.M4datos.PFRT.M4$F_riego <- as.factor(datos.PFRT.M4$F_riego)
datos.PFRT.M4$F_AA <- as.factor(datos.PFRT.M4$F_AA)
str(datos.PFRT.M4)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PFRT : num 24.87 25.52 22.47 27.42 8.67 ...# Anova - Peso fresco de la raíz tuberosa - Muestreo 4
A.PFRT.M4 <- aov(PFRT ~ F_riego*F_AA, data = datos.PFRT.M4)
summary(A.PFRT.M4)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 990.1 990.1 346.485 3.22e-10 ***
## F_AA 2 1.1 0.6 0.199 0.822
## Residuals 12 34.3 2.9
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.PFRT.M4$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PFRT.M4$residuals
## W = 0.96714, p-value = 0.7903- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.PFRT.M4)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFRT ~ F_riego * F_AA, data = datos.PFRT.M4)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 18.16667 16.04023 20.2931 0
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.4795833 -2.282093 3.241259 0.8894738
## 600 ppm-0 ppm -0.2479167 -3.009593 2.513759 0.9689565
## 600 ppm-1000 ppm -0.7275000 -3.916409 2.461409 0.8181924- Los dos tipos de riego son completamente distintos.
# Anova - PFRT - Muestreo 4 - Comparación entre los tratamientos
A.PFRT.M4.CT <- aov(PFRT ~ Tratamiento, data=datos.PFRT.M4)
summary(A.PFRT.M4.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 991.2 330.4 115.6 4.04e-09 ***
## Residuals 12 34.3 2.9
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.PFRT.M4.CT <- TukeyHSD(A.PFRT.M4.CT)
Tukey_A.PFRT.M4.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PFRT ~ Tratamiento, data = datos.PFRT.M4)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 0.5375 -3.01124 4.08624 0.9684291
## 50 + AS 600ppm-50 - AS -0.1900 -3.73874 3.35874 0.9984874
## Control-50 - AS 18.2825 14.73376 21.83124 0.0000000
## 50 + AS 600ppm-50 + AS 1000ppm -0.7275 -4.27624 2.82124 0.9273783
## Control-50 + AS 1000ppm 17.7450 14.19626 21.29374 0.0000000
## Control-50 + AS 600ppm 18.4725 14.92376 22.02124 0.0000000- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.PFRT.M4.DT <- duncan.test(A.PFRT.M4.CT, 'Tratamiento', console = T)##
## Study: A.PFRT.M4.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PFRT
##
## Mean Square Error: 2.857504
##
## Tratamiento, means
##
## PFRT std r se Min Max Q25 Q50 Q75
## 50 - AS 6.7875 1.522594 4 0.8452077 5.13 8.67 5.865 6.675 7.5975
## 50 + AS 1000ppm 7.3250 1.832858 4 0.8452077 5.38 9.71 6.295 7.105 8.1350
## 50 + AS 600ppm 6.5975 1.255929 4 0.8452077 5.02 7.71 5.875 6.830 7.5525
## Control 25.0700 2.043282 4 0.8452077 22.47 27.42 24.270 25.195 25.9950
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 2.604344 2.726001 2.799711
##
## Means with the same letter are not significantly different.
##
## PFRT groups
## Control 25.0700 a
## 50 + AS 1000ppm 7.3250 b
## 50 - AS 6.7875 b
## 50 + AS 600ppm 6.5975 bA.PFRT.M4.DT## $statistics
## MSerror Df Mean CV
## 2.857504 12 11.445 14.7699
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 2.604344
## 3 3.225244 2.726001
## 4 3.312453 2.799711
##
## $means
## PFRT std r se Min Max Q25 Q50 Q75
## 50 - AS 6.7875 1.522594 4 0.8452077 5.13 8.67 5.865 6.675 7.5975
## 50 + AS 1000ppm 7.3250 1.832858 4 0.8452077 5.38 9.71 6.295 7.105 8.1350
## 50 + AS 600ppm 6.5975 1.255929 4 0.8452077 5.02 7.71 5.875 6.830 7.5525
## Control 25.0700 2.043282 4 0.8452077 22.47 27.42 24.270 25.195 25.9950
##
## $comparison
## NULL
##
## $groups
## PFRT groups
## Control 25.0700 a
## 50 + AS 1000ppm 7.3250 b
## 50 - AS 6.7875 b
## 50 + AS 600ppm 6.5975 b
##
## attr(,"class")
## [1] "group"Análisis descriptivo - Peso fresco de la parte aérea
# Datos - Gráfico - Peso fresco de la parte aérea
Datos.G.PFRT <- data.frame(factor(Tratamientos, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), Var_M1$`Peso fresco raíz tuberosa`, Var_M2$`Peso fresco raíz tuberosa`, Var_M3$`Peso fresco raíz tuberosa`, Var_M4$`Peso fresco raíz tuberosa`)
colnames(Datos.G.PFRT) <- c("Tratamientos", "31", "38","45", "52")
Datos.G.PFRT# Unificar la variable respuesta en una sola columna
df.long.PFRT = gather(Datos.G.PFRT, dds, PFRT, 2:5)
df.long.PFRT# Diferencias significativas con el control
# M1 M2 M3 M4
DS.PFRT <- c("a", "a", "a", "a", #T1
"b*", "b*", "b*", "b*", #T2
"b*", "b*", "b*", "b*", #T3
"b*", "b*", "b*", "b*") #T4
# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.PFRT = group_by(df.long.PFRT, Tratamientos, dds, ) %>% summarise(mean = mean(PFRT), sd = sd(PFRT))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.PFRT.ds <- data.frame(df.sumzd.PFRT, DS.PFRT)
df.sumzd.PFRT.dsG.PFRT = ggplot(df.sumzd.PFRT, aes(x=dds, y=mean, fill=Tratamientos))+
geom_bar(stat = "identity", position = "dodge", color = "black", alpha = 0.5)+
geom_errorbar(aes(ymin=mean-sd, ymax=mean+sd), width = 0.25, size=0.5, position = position_dodge(0.9), alpha = 0.5)+
scale_fill_manual(values=c("red", "yellow", "blue", "green")) +
theme_minimal()+
theme(axis.text.x = element_text (size = 15),
axis.text.y = element_text (size = 15))+
ggtitle("Peso fresco promedio de la raíz tuberosa \nen los diferentes muestreos") + theme(plot.title = element_text(family = "Times New Roman",
size = rel(2),
vjust = 0.5,
hjust = 0.5))+
labs(x = "Días después de siembra", y = "g de peso fresco")+
theme(axis.title.x = element_text(family = "Times New Roman", size = rel(1.5)), axis.title.y = element_text(family = "Times New Roman", size = rel(1.5)))+
theme(legend.title = element_text(family = "Times New Roman"))+
geom_text(aes(x=dds, y= mean+1.5, label=DS.PFRT),
position = position_dodge(width = 1), size = 5)
G.PFRT
Peso seco de la raíz tuberosa
Análisis estadístico - Muestreo 1
# Datos - Peso seco de la raíz tuberosa - Muestreo 1
datos.PSRT.M1 <- data.frame(Var_M1$`Factor riego`, Var_M1$`Factor ácido ascórbico`, Var_M1$Tratamiento, Var_M1$`Peso seco de raíz tuberosa`)
colnames(datos.PSRT.M1) <- c("F_riego", "F_AA", "Tratamiento", "PSRT")
datos.PSRT.M1datos.PSRT.M1$F_riego <- as.factor(datos.PSRT.M1$F_riego)
datos.PSRT.M1$F_AA <- as.factor(datos.PSRT.M1$F_AA)
str(datos.PSRT.M1)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PSRT : num 0.36 0.44 0.39 0.47 0.28 0.29 0.23 0.31 0.3 0.36 ...# Anova - Peso seco de la raíz tuberosa - Muestreo 1
A.PSRT.M1 <- aov(PSRT ~ F_riego*F_AA, data = datos.PSRT.M1)
summary(A.PSRT.M1)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 0.03797 0.03797 10.920 0.00629 **
## F_AA 2 0.01220 0.00610 1.754 0.21460
## Residuals 12 0.04172 0.00348
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99%.
# Prueba de normalidad de residuos
shapiro.test(A.PSRT.M1$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PSRT.M1$residuals
## W = 0.96966, p-value = 0.833- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.PSRT.M1)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSRT ~ F_riego * F_AA, data = datos.PSRT.M1)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.1125 0.03832341 0.1866766 0.0062881
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.0575 -0.0388356 0.15383560 0.2861832
## 600 ppm-0 ppm -0.0075 -0.1038356 0.08883560 0.9765449
## 600 ppm-1000 ppm -0.0650 -0.1762388 0.04623877 0.2999469- Los dos tipos de riego presenta diferencias significativas.
# Anova - PSRT - Muestreo 1 - Comparación entre los tratamientos
A.PSRT.M1.CT <- aov(PSRT ~ Tratamiento, data=datos.PSRT.M1)
summary(A.PSRT.M1.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 0.05017 0.016723 4.809 0.0201 *
## Residuals 12 0.04172 0.003477
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.PSRT.M1.CT <- TukeyHSD(A.PSRT.M1.CT)
Tukey_A.PSRT.M1.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSRT ~ Tratamiento, data = datos.PSRT.M1)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 0.0700 -0.053790776 0.19379078 0.3753766
## 50 + AS 600ppm-50 - AS 0.0050 -0.118790776 0.12879078 0.9993473
## Control-50 - AS 0.1375 0.013709224 0.26129078 0.0282163
## 50 + AS 600ppm-50 + AS 1000ppm -0.0650 -0.188790776 0.05879078 0.4357590
## Control-50 + AS 1000ppm 0.0675 -0.056290776 0.19129078 0.4049345
## Control-50 + AS 600ppm 0.1325 0.008709224 0.25629078 0.0347928- Podemos observar que los tratamientos amarillo (50 - AS) y azul (50 + AS 600ppm), tuvieron diferencias significativas con respecto al control.
A.PSRT.M1.DT <- duncan.test(A.PSRT.M1.CT, 'Tratamiento', console = T)##
## Study: A.PSRT.M1.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PSRT
##
## Mean Square Error: 0.003477083
##
## Tratamiento, means
##
## PSRT std r se Min Max Q25 Q50 Q75
## 50 - AS 0.2775 0.03403430 4 0.0294834 0.23 0.31 0.2675 0.285 0.2950
## 50 + AS 1000ppm 0.3475 0.06849574 4 0.0294834 0.27 0.43 0.3075 0.345 0.3850
## 50 + AS 600ppm 0.2825 0.07500000 4 0.0294834 0.18 0.36 0.2625 0.295 0.3150
## Control 0.4150 0.04932883 4 0.0294834 0.36 0.47 0.3825 0.415 0.4475
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 0.09084739 0.09509114 0.09766238
##
## Means with the same letter are not significantly different.
##
## PSRT groups
## Control 0.4150 a
## 50 + AS 1000ppm 0.3475 ab
## 50 + AS 600ppm 0.2825 b
## 50 - AS 0.2775 bA.PSRT.M1.DT## $statistics
## MSerror Df Mean CV
## 0.003477083 12 0.330625 17.83495
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 0.09084739
## 3 3.225244 0.09509114
## 4 3.312453 0.09766238
##
## $means
## PSRT std r se Min Max Q25 Q50 Q75
## 50 - AS 0.2775 0.03403430 4 0.0294834 0.23 0.31 0.2675 0.285 0.2950
## 50 + AS 1000ppm 0.3475 0.06849574 4 0.0294834 0.27 0.43 0.3075 0.345 0.3850
## 50 + AS 600ppm 0.2825 0.07500000 4 0.0294834 0.18 0.36 0.2625 0.295 0.3150
## Control 0.4150 0.04932883 4 0.0294834 0.36 0.47 0.3825 0.415 0.4475
##
## $comparison
## NULL
##
## $groups
## PSRT groups
## Control 0.4150 a
## 50 + AS 1000ppm 0.3475 ab
## 50 + AS 600ppm 0.2825 b
## 50 - AS 0.2775 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 2
# Datos - Peso seco de la raíz tuberosa - Muestreo 2
datos.PSRT.M2 <- data.frame(Var_M2$`Factor riego`, Var_M2$`Factor ácido ascórbico`, Var_M2$Tratamiento, Var_M2$`Peso seco de raíz tuberosa`)
colnames(datos.PSRT.M2) <- c("F_riego", "F_AA", "Tratamiento", "PSRT")
datos.PSRT.M2datos.PSRT.M2$F_riego <- as.factor(datos.PSRT.M2$F_riego)
datos.PSRT.M2$F_AA <- as.factor(datos.PSRT.M2$F_AA)
str(datos.PSRT.M2)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PSRT : num 0.33 0.71 0.87 0.97 0.51 0.41 0.61 0.4 0.52 0.52 ...# Anova - Peso seco de la raíz tuberosa - Muestreo 2
A.PSRT.M2 <- aov(PSRT ~ F_riego*F_AA, data = datos.PSRT.M2)
summary(A.PSRT.M2)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 0.1633 0.16333 4.958 0.0459 *
## F_AA 2 0.0037 0.00186 0.056 0.9454
## Residuals 12 0.3954 0.03295
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 95%.
# Prueba de normalidad de residuos
shapiro.test(A.PSRT.M2$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PSRT.M2$residuals
## W = 0.94092, p-value = 0.3605- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.PSRT.M2)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSRT ~ F_riego * F_AA, data = datos.PSRT.M2)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.2333333 0.005005139 0.4616615 0.0458927
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -0.01708333 -0.3136208 0.2794541 0.9870757
## 600 ppm-0 ppm 0.02541667 -0.2711208 0.3219541 0.9716541
## 600 ppm-1000 ppm 0.04250000 -0.2999119 0.3849119 0.9416378- Los dos tipos de riego presenta diferencias significativas.
# Anova - PSRT - Muestreo 2 - Comparación entre los tratamientos
A.PSRT.M2.CT <- aov(PSRT ~ Tratamiento, data=datos.PSRT.M2)
summary(A.PSRT.M2.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 0.1670 0.05568 1.69 0.222
## Residuals 12 0.3953 0.03295Tukey_A.PSRT.M2.CT <- TukeyHSD(A.PSRT.M2.CT)
Tukey_A.PSRT.M2.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSRT ~ Tratamiento, data = datos.PSRT.M2)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -0.0150 -0.3960491 0.3660491 0.9993955
## 50 + AS 600ppm-50 - AS 0.0275 -0.3535491 0.4085491 0.9963309
## Control-50 - AS 0.2375 -0.1435491 0.6185491 0.2984441
## 50 + AS 600ppm-50 + AS 1000ppm 0.0425 -0.3385491 0.4235491 0.9868498
## Control-50 + AS 1000ppm 0.2525 -0.1285491 0.6335491 0.2528907
## Control-50 + AS 600ppm 0.2100 -0.1710491 0.5910491 0.3962598- Podemos observar que ninguna de los tratamientos tienen diferencias signifciativas con respecto a su control.
A.PSRT.M2.DT <- duncan.test(A.PSRT.M2.CT, 'Tratamiento', console = T)##
## Study: A.PSRT.M2.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PSRT
##
## Mean Square Error: 0.03294583
##
## Tratamiento, means
##
## PSRT std r se Min Max Q25 Q50 Q75
## 50 - AS 0.4825 0.09844626 4 0.09075494 0.40 0.61 0.4075 0.46 0.5350
## 50 + AS 1000ppm 0.4675 0.20451161 4 0.09075494 0.22 0.65 0.3400 0.50 0.6275
## 50 + AS 600ppm 0.5100 0.03464102 4 0.09075494 0.46 0.54 0.5050 0.52 0.5250
## Control 0.7200 0.28118796 4 0.09075494 0.33 0.97 0.6150 0.79 0.8950
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 0.2796438 0.2927068 0.3006215
##
## Means with the same letter are not significantly different.
##
## PSRT groups
## Control 0.7200 a
## 50 + AS 600ppm 0.5100 a
## 50 - AS 0.4825 a
## 50 + AS 1000ppm 0.4675 aA.PSRT.M2.DT## $statistics
## MSerror Df Mean CV
## 0.03294583 12 0.545 33.30456
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 0.2796438
## 3 3.225244 0.2927068
## 4 3.312453 0.3006215
##
## $means
## PSRT std r se Min Max Q25 Q50 Q75
## 50 - AS 0.4825 0.09844626 4 0.09075494 0.40 0.61 0.4075 0.46 0.5350
## 50 + AS 1000ppm 0.4675 0.20451161 4 0.09075494 0.22 0.65 0.3400 0.50 0.6275
## 50 + AS 600ppm 0.5100 0.03464102 4 0.09075494 0.46 0.54 0.5050 0.52 0.5250
## Control 0.7200 0.28118796 4 0.09075494 0.33 0.97 0.6150 0.79 0.8950
##
## $comparison
## NULL
##
## $groups
## PSRT groups
## Control 0.7200 a
## 50 + AS 600ppm 0.5100 a
## 50 - AS 0.4825 a
## 50 + AS 1000ppm 0.4675 a
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 3
# Datos - Peso seco de la raíz tuberosa - Muestreo 3
datos.PSRT.M3 <- data.frame(Var_M3$`Factor riego`, Var_M3$`Factor ácido ascórbico`, Var_M3$Tratamiento, Var_M3$`Peso seco de raíz tuberosa`)
colnames(datos.PSRT.M3) <- c("F_riego", "F_AA", "Tratamiento", "PSRT")
datos.PSRT.M3datos.PSRT.M3$F_riego <- as.factor(datos.PSRT.M3$F_riego)
datos.PSRT.M3$F_AA <- as.factor(datos.PSRT.M3$F_AA)
str(datos.PSRT.M3)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PSRT : num 1.381 1.488 0.813 1.193 0.682 ...# Anova - Peso seco de la raíz tuberosa - Muestreo 3
A.PSRT.M3 <- aov(PSRT ~ F_riego*F_AA, data = datos.PSRT.M3)
summary(A.PSRT.M3)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 0.8990 0.8990 33.466 8.67e-05 ***
## F_AA 2 0.0448 0.0224 0.834 0.458
## Residuals 12 0.3224 0.0269
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.PSRT.M3$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PSRT.M3$residuals
## W = 0.90457, p-value = 0.09516- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.PSRT.M3)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSRT ~ F_riego * F_AA, data = datos.PSRT.M3)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.5474167 0.3412416 0.7535917 8.67e-05
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.10345833 -0.164308 0.3712247 0.5725280
## 600 ppm-0 ppm -0.03329167 -0.301058 0.2344747 0.9414457
## 600 ppm-1000 ppm -0.13675000 -0.445940 0.1724400 0.4866046- Los dos tipos de riego presenta diferencias significativas.
# Anova - PSRT - Muestreo 3 - Comparación entre los tratamientos
A.PSRT.M3.CT <- aov(PSRT ~ Tratamiento, data=datos.PSRT.M3)
summary(A.PSRT.M3.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 0.9438 0.31459 11.71 0.000706 ***
## Residuals 12 0.3224 0.02686
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.PSRT.M3.CT <- TukeyHSD(A.PSRT.M3.CT)
Tukey_A.PSRT.M3.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSRT ~ Tratamiento, data = datos.PSRT.M3)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 0.12100 -0.2230785 0.4650785 0.7279285
## 50 + AS 600ppm-50 - AS -0.01575 -0.3598285 0.3283285 0.9990517
## Control-50 - AS 0.58250 0.2384215 0.9265785 0.0014559
## 50 + AS 600ppm-50 + AS 1000ppm -0.13675 -0.4808285 0.2073285 0.6500203
## Control-50 + AS 1000ppm 0.46150 0.1174215 0.8055785 0.0085228
## Control-50 + AS 600ppm 0.59825 0.2541715 0.9423285 0.0011668- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.PSRT.M3.DT <- duncan.test(A.PSRT.M3.CT, 'Tratamiento', console = T)##
## Study: A.PSRT.M3.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PSRT
##
## Mean Square Error: 0.02686294
##
## Tratamiento, means
##
## PSRT std r se Min Max Q25 Q50
## 50 - AS 0.63625 0.10607977 4 0.08194958 0.479 0.711 0.6245 0.6775
## 50 + AS 1000ppm 0.75725 0.08269371 4 0.08194958 0.674 0.862 0.7025 0.7465
## 50 + AS 600ppm 0.62050 0.03635473 4 0.08194958 0.577 0.664 0.6025 0.6205
## Control 1.21875 0.29671353 4 0.08194958 0.813 1.488 1.0980 1.2870
## Q75
## 50 - AS 0.68925
## 50 + AS 1000ppm 0.80125
## 50 + AS 600ppm 0.63850
## Control 1.40775
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 0.2525118 0.2643074 0.2714541
##
## Means with the same letter are not significantly different.
##
## PSRT groups
## Control 1.21875 a
## 50 + AS 1000ppm 0.75725 b
## 50 - AS 0.63625 b
## 50 + AS 600ppm 0.62050 bA.PSRT.M3.DT## $statistics
## MSerror Df Mean CV
## 0.02686294 12 0.8081875 20.27984
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 0.2525118
## 3 3.225244 0.2643074
## 4 3.312453 0.2714541
##
## $means
## PSRT std r se Min Max Q25 Q50
## 50 - AS 0.63625 0.10607977 4 0.08194958 0.479 0.711 0.6245 0.6775
## 50 + AS 1000ppm 0.75725 0.08269371 4 0.08194958 0.674 0.862 0.7025 0.7465
## 50 + AS 600ppm 0.62050 0.03635473 4 0.08194958 0.577 0.664 0.6025 0.6205
## Control 1.21875 0.29671353 4 0.08194958 0.813 1.488 1.0980 1.2870
## Q75
## 50 - AS 0.68925
## 50 + AS 1000ppm 0.80125
## 50 + AS 600ppm 0.63850
## Control 1.40775
##
## $comparison
## NULL
##
## $groups
## PSRT groups
## Control 1.21875 a
## 50 + AS 1000ppm 0.75725 b
## 50 - AS 0.63625 b
## 50 + AS 600ppm 0.62050 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 4
# Datos - Peso seco de la raíz tuberosa - Muestreo 4
datos.PSRT.M4 <- data.frame(Var_M4$`Factor riego`, Var_M4$`Factor ácido ascórbico`, Var_M4$Tratamiento, Var_M4$`Peso seco de raíz tuberosa`)
colnames(datos.PSRT.M4) <- c("F_riego", "F_AA", "Tratamiento", "PSRT")
datos.PSRT.M4datos.PSRT.M4$F_riego <- as.factor(datos.PSRT.M4$F_riego)
datos.PSRT.M4$F_AA <- as.factor(datos.PSRT.M4$F_AA)
str(datos.PSRT.M4)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PSRT : num 1.81 1.69 1.57 1.55 1.02 ...# Anova - Peso seco de la raíz tuberosa - Muestreo 4
A.PSRT.M4 <- aov(PSRT ~ F_riego*F_AA, data = datos.PSRT.M4)
summary(A.PSRT.M4)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 1.9341 1.9341 61.192 4.73e-06 ***
## F_AA 2 0.0334 0.0167 0.529 0.603
## Residuals 12 0.3793 0.0316
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%..
# Prueba de normalidad de residuos
shapiro.test(A.PSRT.M4$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PSRT.M4$residuals
## W = 0.94074, p-value = 0.3582- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.PSRT.M4)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSRT ~ F_riego * F_AA, data = datos.PSRT.M4)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.8029417 0.5792989 1.026584 4.7e-06
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.004395833 -0.2860565 0.2948481 0.9991017
## 600 ppm-0 ppm -0.096979167 -0.3874315 0.1934731 0.6560492
## 600 ppm-1000 ppm -0.101375000 -0.4367604 0.2340104 0.7063045- Los dos tipos de riego son completamente distintos.
# Anova - PSRT - Muestreo 4 - Comparación entre los tratamientos
A.PSRT.M4.CT <- aov(PSRT ~ Tratamiento, data=datos.PSRT.M4)
summary(A.PSRT.M4.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 1.9676 0.6559 20.75 4.85e-05 ***
## Residuals 12 0.3793 0.0316
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.PSRT.M4.CT <- TukeyHSD(A.PSRT.M4.CT)
Tukey_A.PSRT.M4.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PSRT ~ Tratamiento, data = datos.PSRT.M4)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -0.018750 -0.3919798 0.3544798 0.9987487
## 50 + AS 600ppm-50 - AS -0.120125 -0.4933548 0.2531048 0.7762726
## Control-50 - AS 0.756650 0.3834202 1.1298798 0.0003048
## 50 + AS 600ppm-50 + AS 1000ppm -0.101375 -0.4746048 0.2718548 0.8502160
## Control-50 + AS 1000ppm 0.775400 0.4021702 1.1486298 0.0002436
## Control-50 + AS 600ppm 0.876775 0.5035452 1.2500048 0.0000762- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.PSRT.M4.DT <- duncan.test(A.PSRT.M4.CT, 'Tratamiento', console = T)##
## Study: A.PSRT.M4.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PSRT
##
## Mean Square Error: 0.03160757
##
## Tratamiento, means
##
## PSRT std r se Min Max Q25 Q50
## 50 - AS 0.898900 0.2244063 4 0.08889259 0.6096 1.1230 0.786000 0.93150
## 50 + AS 1000ppm 0.880150 0.2141000 4 0.08889259 0.5665 1.0475 0.845575 0.95330
## 50 + AS 600ppm 0.778775 0.1256403 4 0.08889259 0.6533 0.9441 0.701150 0.75885
## Control 1.655550 0.1201990 4 0.08889259 1.5491 1.8087 1.565600 1.63220
## Q75
## 50 - AS 1.044400
## 50 + AS 1000ppm 0.987875
## 50 + AS 600ppm 0.836475
## Control 1.722150
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 0.2739053 0.2867002 0.2944525
##
## Means with the same letter are not significantly different.
##
## PSRT groups
## Control 1.655550 a
## 50 - AS 0.898900 b
## 50 + AS 1000ppm 0.880150 b
## 50 + AS 600ppm 0.778775 bA.PSRT.M4.DT## $statistics
## MSerror Df Mean CV
## 0.03160757 12 1.053344 16.87817
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 0.2739053
## 3 3.225244 0.2867002
## 4 3.312453 0.2944525
##
## $means
## PSRT std r se Min Max Q25 Q50
## 50 - AS 0.898900 0.2244063 4 0.08889259 0.6096 1.1230 0.786000 0.93150
## 50 + AS 1000ppm 0.880150 0.2141000 4 0.08889259 0.5665 1.0475 0.845575 0.95330
## 50 + AS 600ppm 0.778775 0.1256403 4 0.08889259 0.6533 0.9441 0.701150 0.75885
## Control 1.655550 0.1201990 4 0.08889259 1.5491 1.8087 1.565600 1.63220
## Q75
## 50 - AS 1.044400
## 50 + AS 1000ppm 0.987875
## 50 + AS 600ppm 0.836475
## Control 1.722150
##
## $comparison
## NULL
##
## $groups
## PSRT groups
## Control 1.655550 a
## 50 - AS 0.898900 b
## 50 + AS 1000ppm 0.880150 b
## 50 + AS 600ppm 0.778775 b
##
## attr(,"class")
## [1] "group"Análisis descriptivo - Peso seco de la raíz tuberosa
# Datos - Gráfico - Peso seco de la raíz tuberosa
Datos.G.PSRT <- data.frame(factor(Tratamientos, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), Var_M1$`Peso seco de raíz tuberosa`, Var_M2$`Peso seco de raíz tuberosa`, Var_M3$`Peso seco de raíz tuberosa`, Var_M4$`Peso seco de raíz tuberosa`)
colnames(Datos.G.PSRT) <- c("Tratamientos", "31", "38","45", "52")
Datos.G.PSRT# Unificar la variable respuesta en una sola columna
df.long.PSRT = gather(Datos.G.PSRT, dds, PSRT, 2:5)
df.long.PSRT# Diferencias significativas con el control
# M1 M2 M3 M4
DS.PSRT <- c("a", "a", "a", "a", #T1
"b*", "a", "b*", "b*", #T2
"b*", "a", "b*", "b*", #T3
"ab", "a", "b*", "b*") #T4
# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.PSRT = group_by(df.long.PSRT, Tratamientos, dds, ) %>% summarise(mean = mean(PSRT), sd = sd(PSRT))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.PSRT.ds <- data.frame(df.sumzd.PSRT, DS.PSRT)
df.sumzd.PSRT.dsG.PSRT = ggplot(df.sumzd.PSRT, aes(x=dds, y=mean, fill=Tratamientos))+
geom_bar(stat = "identity", position = "dodge", color = "black", alpha = 0.5)+
geom_errorbar(aes(ymin=mean-sd, ymax=mean+sd), width = 0.25, size=0.5, position = position_dodge(0.9), alpha = 0.5)+
scale_fill_manual(values=c("red", "yellow", "blue", "green")) +
theme_minimal()+
ggtitle("Peso seco promedio de la raíz tuberosa \nen los diferentes muestreos") + theme(plot.title = element_text(family = "Times New Roman",
size = rel(1.5),
vjust = 0.5,
hjust = 0.5))+
labs(x = "Días después de siembra", y = "g")+
theme(axis.title.x = element_text(family = "Times New Roman"), axis.title.y = element_text(family = "Times New Roman"))+
theme(legend.title = element_text(family = "Times New Roman"))+
geom_text(aes(x=dds, y= mean+0.4, label=DS.PSRT),
position = position_dodge(width = 1), size = 5)
G.PSRT
Área bajo el dosel
Análisis estadístico - Muestreo 1
# Datos - Área bajo el dosel - Muestreo 1
datos.ABD.M1 <- data.frame(Var_M1$`Factor riego`, Var_M1$`Factor ácido ascórbico`, Var_M1$Tratamiento, Var_M1$`Área bajo el dosel (cm2)`)
colnames(datos.ABD.M1) <- c("F_riego", "F_AA", "Tratamiento", "ABD")
datos.ABD.M1datos.ABD.M1$F_riego <- as.factor(datos.ABD.M1$F_riego)
datos.ABD.M1$F_AA <- as.factor(datos.ABD.M1$F_AA)
str(datos.ABD.M1)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ ABD : num 89.5 146.8 103.6 94.8 35 ...# Anova - Área bajo el dosel - Muestreo 1
A.ABD.M1 <- aov(ABD ~ F_riego*F_AA, data = datos.ABD.M1)
summary(A.ABD.M1)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 9512 9512 21.52 0.000572 ***
## F_AA 2 17 9 0.02 0.980609
## Residuals 12 5305 442
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.ABD.M1$residuals)##
## Shapiro-Wilk normality test
##
## data: A.ABD.M1$residuals
## W = 0.92859, p-value = 0.2315- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.ABD.M1)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = ABD ~ F_riego * F_AA, data = datos.ABD.M1)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 56.3075 29.85853 82.75647 0.0005717
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -1.6875 -36.03766 32.66266 0.9905821
## 600 ppm-0 ppm -1.7125 -36.06266 32.63766 0.9903026
## 600 ppm-1000 ppm -0.0250 -39.68914 39.63914 0.9999984- Los dos tipos de riego presenta diferencias significativas.
# Anova - ABD - Muestreo 1 - Comparación entre los tratamientos
A.ABD.M1.CT <- aov(ABD ~ Tratamiento, data=datos.ABD.M1)
summary(A.ABD.M1.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 9529 3176 7.185 0.00511 **
## Residuals 12 5305 442
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.ABD.M1.CT <- TukeyHSD(A.ABD.M1.CT)
Tukey_A.ABD.M1.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = ABD ~ Tratamiento, data = datos.ABD.M1)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -2.5375 -46.67728 41.60228 0.9981309
## 50 + AS 600ppm-50 - AS -2.5625 -46.70228 41.57728 0.9980756
## Control-50 - AS 54.6075 10.46772 98.74728 0.0146180
## 50 + AS 600ppm-50 + AS 1000ppm -0.0250 -44.16478 44.11478 1.0000000
## Control-50 + AS 1000ppm 57.1450 13.00522 101.28478 0.0108464
## Control-50 + AS 600ppm 57.1700 13.03022 101.30978 0.0108146- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.ABD.M1.DT <- duncan.test(A.ABD.M1.CT, 'Tratamiento', console = T)##
## Study: A.ABD.M1.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for ABD
##
## Mean Square Error: 442.078
##
## Tratamiento, means
##
## ABD std r se Min Max Q25 Q50
## 50 - AS 54.0775 16.75190 4 10.51283 34.96 71.77 43.0750 54.790
## 50 + AS 1000ppm 51.5400 23.40818 4 10.51283 26.86 79.02 35.5150 50.140
## 50 + AS 600ppm 51.5150 16.14043 4 10.51283 32.80 68.49 41.1775 52.385
## Control 108.6850 26.06204 4 10.51283 89.53 146.80 93.5050 99.205
## Q75
## 50 - AS 65.7925
## 50 + AS 1000ppm 66.1650
## 50 + AS 600ppm 62.7225
## Control 114.3850
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 32.39324 33.90642 34.82324
##
## Means with the same letter are not significantly different.
##
## ABD groups
## Control 108.6850 a
## 50 - AS 54.0775 b
## 50 + AS 1000ppm 51.5400 b
## 50 + AS 600ppm 51.5150 bA.ABD.M1.DT## $statistics
## MSerror Df Mean CV
## 442.078 12 66.45437 31.63923
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 32.39324
## 3 3.225244 33.90642
## 4 3.312453 34.82324
##
## $means
## ABD std r se Min Max Q25 Q50
## 50 - AS 54.0775 16.75190 4 10.51283 34.96 71.77 43.0750 54.790
## 50 + AS 1000ppm 51.5400 23.40818 4 10.51283 26.86 79.02 35.5150 50.140
## 50 + AS 600ppm 51.5150 16.14043 4 10.51283 32.80 68.49 41.1775 52.385
## Control 108.6850 26.06204 4 10.51283 89.53 146.80 93.5050 99.205
## Q75
## 50 - AS 65.7925
## 50 + AS 1000ppm 66.1650
## 50 + AS 600ppm 62.7225
## Control 114.3850
##
## $comparison
## NULL
##
## $groups
## ABD groups
## Control 108.6850 a
## 50 - AS 54.0775 b
## 50 + AS 1000ppm 51.5400 b
## 50 + AS 600ppm 51.5150 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 2
# Datos - Área bajo el dosel - Muestreo 1
datos.ABD.M2 <- data.frame(Var_M2$`Factor riego`, Var_M2$`Factor ácido ascórbico`, Var_M2$Tratamiento, Var_M2$`Área bajo el dosel (cm2)`)
colnames(datos.ABD.M2) <- c("F_riego", "F_AA", "Tratamiento", "ABD")
datos.ABD.M2datos.ABD.M2$F_riego <- as.factor(datos.ABD.M2$F_riego)
datos.ABD.M2$F_AA <- as.factor(datos.ABD.M2$F_AA)
str(datos.ABD.M2)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ ABD : num 119 177 215 194 128 ...# Anova - Área bajo el dosel - Muestreo 2
A.ABD.M2 <- aov(ABD ~ F_riego*F_AA, data = datos.ABD.M2)
summary(A.ABD.M2)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 16111 16111 19.808 0.000792 ***
## F_AA 2 1121 560 0.689 0.520956
## Residuals 12 9761 813
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.ABD.M2$residuals)##
## Shapiro-Wilk normality test
##
## data: A.ABD.M2$residuals
## W = 0.92191, p-value = 0.1811- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.ABD.M2)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = ABD ~ F_riego * F_AA, data = datos.ABD.M2)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 73.28333 37.40714 109.1595 0.000792
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 11.74667 -34.84694 58.34028 0.7834004
## 600 ppm-0 ppm -11.92333 -58.51694 34.67028 0.7777510
## 600 ppm-1000 ppm -23.67000 -77.47166 30.13166 0.4900891- Los dos tipos de riego presenta diferencias significativas.
# Anova - ABD - Muestreo 2 - Comparación entre los tratamientos
A.ABD.M2.CT <- aov(ABD ~ Tratamiento, data=datos.ABD.M2)
summary(A.ABD.M2.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 17232 5744 7.062 0.00544 **
## Residuals 12 9761 813
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.ABD.M2.CT <- TukeyHSD(A.ABD.M2.CT)
Tukey_A.ABD.M2.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = ABD ~ Tratamiento, data = datos.ABD.M2)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 11.7025 -48.170059 71.57506 0.9361157
## 50 + AS 600ppm-50 - AS -11.9675 -71.840059 47.90506 0.9321405
## Control-50 - AS 73.1950 13.322441 133.06756 0.0157742
## 50 + AS 600ppm-50 + AS 1000ppm -23.6700 -83.542559 36.20256 0.6536464
## Control-50 + AS 1000ppm 61.4925 1.619941 121.36506 0.0435111
## Control-50 + AS 600ppm 85.1625 25.289941 145.03506 0.0056189- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.ABD.M2.DT <- duncan.test(A.ABD.M2.CT, 'Tratamiento', console = T)##
## Study: A.ABD.M2.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for ABD
##
## Mean Square Error: 813.3813
##
## Tratamiento, means
##
## ABD std r se Min Max Q25 Q50
## 50 - AS 103.3375 31.19703 4 14.25992 59.45 128.38 92.1425 112.760
## 50 + AS 1000ppm 115.0400 16.41308 4 14.25992 93.30 133.18 110.9550 116.840
## 50 + AS 600ppm 91.3700 17.09741 4 14.25992 70.14 111.58 84.2400 91.880
## Control 176.5325 41.45552 4 14.25992 118.92 215.44 162.6975 185.885
## Q75
## 50 - AS 123.955
## 50 + AS 1000ppm 120.925
## 50 + AS 600ppm 99.010
## Control 199.720
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 43.93919 45.99171 47.23531
##
## Means with the same letter are not significantly different.
##
## ABD groups
## Control 176.5325 a
## 50 + AS 1000ppm 115.0400 b
## 50 - AS 103.3375 b
## 50 + AS 600ppm 91.3700 bA.ABD.M2.DT## $statistics
## MSerror Df Mean CV
## 813.3813 12 121.57 23.4596
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 43.93919
## 3 3.225244 45.99171
## 4 3.312453 47.23531
##
## $means
## ABD std r se Min Max Q25 Q50
## 50 - AS 103.3375 31.19703 4 14.25992 59.45 128.38 92.1425 112.760
## 50 + AS 1000ppm 115.0400 16.41308 4 14.25992 93.30 133.18 110.9550 116.840
## 50 + AS 600ppm 91.3700 17.09741 4 14.25992 70.14 111.58 84.2400 91.880
## Control 176.5325 41.45552 4 14.25992 118.92 215.44 162.6975 185.885
## Q75
## 50 - AS 123.955
## 50 + AS 1000ppm 120.925
## 50 + AS 600ppm 99.010
## Control 199.720
##
## $comparison
## NULL
##
## $groups
## ABD groups
## Control 176.5325 a
## 50 + AS 1000ppm 115.0400 b
## 50 - AS 103.3375 b
## 50 + AS 600ppm 91.3700 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 3
# Datos - Área bajo el dosel - Muestreo 3
datos.ABD.M3 <- data.frame(Var_M3$`Factor riego`, Var_M3$`Factor ácido ascórbico`, Var_M3$Tratamiento, Var_M3$`Área bajo el dosel (cm2)`)
colnames(datos.ABD.M3) <- c("F_riego", "F_AA", "Tratamiento", "ABD")
datos.ABD.M3datos.ABD.M3$F_riego <- as.factor(datos.ABD.M3$F_riego)
datos.ABD.M3$F_AA <- as.factor(datos.ABD.M3$F_AA)
str(datos.ABD.M3)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ ABD : num 243.1 155.6 305.6 210.5 46.1 ...# Anova - Área bajo el dosel - Muestreo 3
A.ABD.M3 <- aov(ABD ~ F_riego*F_AA, data = datos.ABD.M3)
summary(A.ABD.M3)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 110139 110139 97.836 4.03e-07 ***
## F_AA 2 190 95 0.084 0.92
## Residuals 12 13509 1126
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.ABD.M3$residuals)##
## Shapiro-Wilk normality test
##
## data: A.ABD.M3$residuals
## W = 0.87094, p-value = 0.02811- Los datos no tienen una distribución normal, por lo tanto, se debe aplicar otra prueba que soporte la distribución de estos datos.
TukeyHSD(A.ABD.M3)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = ABD ~ F_riego * F_AA, data = datos.ABD.M3)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 191.6067 149.4 233.8133 4e-07
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -5.074917 -59.89012 49.74028 0.9670198
## 600 ppm-0 ppm -6.115417 -60.93062 48.69978 0.9525305
## 600 ppm-1000 ppm -1.040500 -64.33564 62.25464 0.9989402- Los dos tipos de riego presenta diferencias significativas.
# Anova - ABD - Muestreo 3 - Comparación entre los tratamientos
A.ABD.M3.CT <- aov(ABD ~ Tratamiento, data=datos.ABD.M3)
summary(A.ABD.M3.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 110329 36776 32.67 4.7e-06 ***
## Residuals 12 13509 1126
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.ABD.M3.CT <- TukeyHSD(A.ABD.M3.CT)
Tukey_A.ABD.M3.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = ABD ~ Tratamiento, data = datos.ABD.M3)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -7.8725 -78.30976 62.56476 0.9867703
## 50 + AS 600ppm-50 - AS -8.9130 -79.35026 61.52426 0.9810721
## Control-50 - AS 186.0115 115.57424 256.44876 0.0000239
## 50 + AS 600ppm-50 + AS 1000ppm -1.0405 -71.47776 69.39676 0.9999679
## Control-50 + AS 1000ppm 193.8840 123.44674 264.32126 0.0000157
## Control-50 + AS 600ppm 194.9245 124.48724 265.36176 0.0000148- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
Análisis estadístico - Muestreo 4
# Datos - Área bajo el dosel - Muestreo 4
datos.ABD.M4 <- data.frame(Var_M4$`Factor riego`, Var_M4$`Factor ácido ascórbico`, Var_M4$Tratamiento, Var_M4$`Área bajo el dosel (cm2)`)
colnames(datos.ABD.M4) <- c("F_riego", "F_AA", "Tratamiento", "ABD")
datos.ABD.M4datos.ABD.M4$F_riego <- as.factor(datos.ABD.M4$F_riego)
datos.ABD.M4$F_AA <- as.factor(datos.ABD.M4$F_AA)
str(datos.ABD.M4)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ ABD : num 180.6 286.1 337.3 270.8 62.7 ...# Anova - Área bajo el dosel - Muestreo 4
A.ABD.M4 <- aov(ABD ~ F_riego*F_AA, data = datos.ABD.M4)
summary(A.ABD.M4)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 128425 128425 102.104 3.2e-07 ***
## F_AA 2 2078 1039 0.826 0.461
## Residuals 12 15093 1258
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.ABD.M4$residuals)##
## Shapiro-Wilk normality test
##
## data: A.ABD.M4$residuals
## W = 0.86669, p-value = 0.02421- Los datos no tienen una distribución normal, por lo tanto, se debe aplicar otra prueba que soporte la distribución de estos datos.
TukeyHSD(A.ABD.M4)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = ABD ~ F_riego * F_AA, data = datos.ABD.M4)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 206.9018 162.2885 251.515 3e-07
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 19.00875 -38.93192 76.94942 0.6653349
## 600 ppm-0 ppm 18.20125 -39.73942 76.14192 0.6875191
## 600 ppm-1000 ppm -0.80750 -67.71162 66.09662 0.9994286- Los dos tipos de riego presenta diferencias significativas.
# Anova - ABD - Muestreo 4 - Comparación entre los tratamientos
A.ABD.M4.CT <- aov(ABD ~ Tratamiento, data=datos.ABD.M4)
summary(A.ABD.M4.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 130503 43501 34.59 3.47e-06 ***
## Residuals 12 15093 1258
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.ABD.M4.CT <- TukeyHSD(A.ABD.M4.CT)
Tukey_A.ABD.M4.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = ABD ~ Tratamiento, data = datos.ABD.M4)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 28.31125 -46.14222 102.76472 0.6796043
## 50 + AS 600ppm-50 - AS 27.50375 -46.94972 101.95722 0.6981024
## Control-50 - AS 225.50675 151.05328 299.96022 0.0000058
## 50 + AS 600ppm-50 + AS 1000ppm -0.80750 -75.26097 73.64597 0.9999873
## Control-50 + AS 1000ppm 197.19550 122.74203 271.64897 0.0000232
## Control-50 + AS 600ppm 198.00300 123.54953 272.45647 0.0000223- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
Análisis descriptivo - Área bajo el dosel
# Datos - Gráfico - Área bajo el dosel
Datos.G.ABD <- data.frame(factor(Tratamientos, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), Var_M1$`Área bajo el dosel (cm2)`, Var_M2$`Área bajo el dosel (cm2)`, Var_M3$`Área bajo el dosel (cm2)`, Var_M4$`Área bajo el dosel (cm2)`)
colnames(Datos.G.ABD) <- c("Tratamientos", "31", "38","45", "52")
Datos.G.ABD# Unificar la variable respuesta en una sola columna
df.long.ABD = gather(Datos.G.ABD, dds, ABD, 2:5)
df.long.ABD# Diferencias significativas con el control
# M1 M2 M3 M4
DS.ABD <- c("", "", "", "", #T1
"*", "*", "*", "*", #T2
"*", "*", "*", "*", #T3
"*", "*", "*", "*") #T4
# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.ABD = group_by(df.long.ABD, Tratamientos, dds, ) %>% summarise(mean = mean(ABD), sd = sd(ABD))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.ABD.ds <- data.frame(df.sumzd.ABD, DS.ABD)
df.sumzd.ABD.dsG.ABD <- ggplot(df.sumzd.ABD, aes(x = dds, y = mean, group = Tratamientos, colour = Tratamientos)) +
scale_colour_manual(values=c("red", "yellow", "blue", "green"))+
geom_line() +
geom_point( size=2, shape = 21, fill="white") +
theme_test()+
ggtitle("Área promedio bajo el dosel \nen los diferentes muestreos")+ theme(plot.title = element_text(family = "Times New Roman",
size = rel(1.5),
vjust = 0.5,
hjust = 0.5))+
labs(x = "Días después de siembra", y = "cm2")+
theme(axis.title.x = element_text(family = "Times New Roman"), axis.title.y = element_text(family = "Times New Roman"))+
theme(legend.title = element_text(family = "Times New Roman"))
G.ABD
G2.ABD <- ggplot(df.sumzd.ABD, aes(x=dds, y=mean, fill=Tratamientos))+
geom_bar(stat = "identity", position = "dodge", color = "black", alpha = 0.5)+
geom_errorbar(aes(ymin=mean-sd, ymax=mean+sd), width = 0.25, size=0.5, position = position_dodge(0.9), alpha = 0.5)+
scale_fill_manual(values=c("red", "yellow", "blue", "green")) +
theme_minimal()+
ggtitle("Área promedio bajo el dosel \nen los diferentes muestreos") + theme(plot.title = element_text(family = "Times New Roman",
size = rel(1.5),
vjust = 0.5,
hjust = 0.5))+
labs(x = "Días después de siembra", y = "g")+
theme(axis.title.x = element_text(family = "Times New Roman"), axis.title.y = element_text(family = "Times New Roman"))+
theme(legend.title = element_text(family = "Times New Roman"))+
geom_text(aes(x=dds, y= mean+30, label=DS.ABD),
position = position_dodge(width = 1), size = 7)
G2.ABD
Tasa de asimilación neta
Análisis estadístico - Muestreo 1
# Datos - Tasa de asimilación neta - Muestreo 1
datos.TAN.M1 <- data.frame(Ind_M1$`Factor riego`, Ind_M1$`Factor ácido ascórbico`, Ind_M1$Tratamiento, Ind_M1$`Tasa de asimilación neta (gr/cm2*día)`)
colnames(datos.TAN.M1) <- c("F_riego", "F_AA", "Tratamiento", "TAN")
datos.TAN.M1datos.TAN.M1$F_riego <- as.factor(datos.TAN.M1$F_riego)
datos.TAN.M1$F_AA <- as.factor(datos.TAN.M1$F_AA)
str(datos.TAN.M1)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ TAN : num 0.00127 0.00127 0.00127 0.00127 0.00114 ...# Anova - Tasa de asimilación neta - Muestreo 1
A.TAN.M1 <- aov(TAN ~ F_riego*F_AA, data = datos.TAN.M1)
summary(A.TAN.M1)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 9.800e-10 9.800e-10 1.790e+29 <2e-16 ***
## F_AA 2 3.055e-07 1.527e-07 2.803e+31 <2e-16 ***
## Residuals 12 0.000e+00 0.000e+00
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Tanto las concentraciones de ácido ascórbico como los dos tipos de riego, como factores independientes, tuvieron diferencias entre si con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.TAN.M1$residuals)##
## Shapiro-Wilk normality test
##
## data: A.TAN.M1$residuals
## W = 0.59712, p-value = 1.521e-05- Los datos no tienen una distribución normal, por lo tanto, se debe aplicar una prueba que soporte la distribución de estos datos.
TukeyHSD(A.TAN.M1)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAN ~ F_riego * F_AA, data = datos.TAN.M1)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 -1.803209e-05 -1.803209e-05 -1.803209e-05 0
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 2.957265e-04 2.957265e-04 2.957265e-04 0
## 600 ppm-0 ppm 2.122449e-06 2.122449e-06 2.122449e-06 0
## 600 ppm-1000 ppm -2.936041e-04 -2.936041e-04 -2.936041e-04 0- Las concentraciones de ácido áscórbico y los dos tipos de riego presenta diferencias significativas.
# Anova - TAN - Muestreo 1 - Comparación entre los tratamientos
A.TAN.M1.CT <- aov(TAN ~ Tratamiento, data=datos.TAN.M1)
summary(A.TAN.M1.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 3.064e-07 1.021e-07 9.934e+30 <2e-16 ***
## Residuals 12 0.000e+00 0.000e+00
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.TAN.M1.CT <- TukeyHSD(A.TAN.M1.CT)
Tukey_A.TAN.M1.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAN ~ Tratamiento, data = datos.TAN.M1)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 3.701887e-04 3.701887e-04 3.701887e-04 0
## 50 + AS 600ppm-50 - AS 7.658469e-05 7.658469e-05 7.658469e-05 0
## Control-50 - AS 1.308924e-04 1.308924e-04 1.308924e-04 0
## 50 + AS 600ppm-50 + AS 1000ppm -2.936041e-04 -2.936041e-04 -2.936041e-04 0
## Control-50 + AS 1000ppm -2.392964e-04 -2.392964e-04 -2.392964e-04 0
## Control-50 + AS 600ppm 5.430770e-05 5.430770e-05 5.430770e-05 0- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
Análisis estadístico - Muestreo 2
# Datos - Tasa de asimilación neta - Muestreo 2
datos.TAN.M2 <- data.frame(Ind_M2$`Factor riego`, Ind_M2$`Factor ácido ascórbico`, Ind_M2$Tratamiento, Ind_M2$`Tasa de asimilación neta (gr/cm2*día)`)
colnames(datos.TAN.M2) <- c("F_riego", "F_AA", "Tratamiento", "TAN")
datos.TAN.M2datos.TAN.M2$F_riego <- as.factor(datos.TAN.M2$F_riego)
datos.TAN.M2$F_AA <- as.factor(datos.TAN.M2$F_AA)
str(datos.TAN.M2)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ TAN : num 0.000612 0.000612 0.000612 0.000612 0.000464 ...# Anova - Tasa de asimilación neta - Muestreo 2
A.TAN.M2 <- aov(TAN ~ F_riego*F_AA, data = datos.TAN.M2)
summary(A.TAN.M2)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 2.800e-10 2.800e-10 2.112e+28 <2e-16 ***
## F_AA 2 1.562e-07 7.809e-08 5.878e+30 <2e-16 ***
## Residuals 12 0.000e+00 0.000e+00
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.TAN.M2$residuals)##
## Shapiro-Wilk normality test
##
## data: A.TAN.M2$residuals
## W = 0.58314, p-value = 1.127e-05- Los datos no tienen una distribución normal, por lo tanto, se debe aplicar una prueba que soporte la distribución de estos datos.
TukeyHSD(A.TAN.M2)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAN ~ F_riego * F_AA, data = datos.TAN.M2)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 -9.671788e-06 -9.671788e-06 -9.671788e-06 0
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 1.265384e-04 1.265384e-04 1.265384e-04 0
## 600 ppm-0 ppm 1.881863e-04 1.881863e-04 1.881863e-04 0
## 600 ppm-1000 ppm 6.164797e-05 6.164797e-05 6.164797e-05 0- Ambos factores presenta diferencias significativas entre si.
# Anova - TAN - Muestreo 2 - Comparación entre los tratamientos
A.TAN.M2.CT <- aov(TAN ~ Tratamiento, data=datos.TAN.M2)
summary(A.TAN.M2.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 1.565e-07 5.215e-08 8.295e+30 <2e-16 ***
## Residuals 12 0.000e+00 0.000e+00
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.TAN.M2.CT <- TukeyHSD(A.TAN.M2.CT)
Tukey_A.TAN.M2.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAN ~ Tratamiento, data = datos.TAN.M2)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 2.052195e-04 2.052195e-04 2.052195e-04 0
## 50 + AS 600ppm-50 - AS 2.668675e-04 2.668675e-04 2.668675e-04 0
## Control-50 - AS 1.476906e-04 1.476906e-04 1.476906e-04 0
## 50 + AS 600ppm-50 + AS 1000ppm 6.164797e-05 6.164797e-05 6.164797e-05 0
## Control-50 + AS 1000ppm -5.752898e-05 -5.752898e-05 -5.752898e-05 0
## Control-50 + AS 600ppm -1.191770e-04 -1.191770e-04 -1.191770e-04 0- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
Análisis estadístico - Muestreo 3
# Datos - Tasa de asimilación neta - Muestreo 3
datos.TAN.M3 <- data.frame(Ind_M3$`Factor riego`, Ind_M3$`Factor ácido ascórbico`, Ind_M3$Tratamiento, Ind_M3$`Tasa de asimilación neta (gr/cm2*día)`)
colnames(datos.TAN.M3) <- c("F_riego", "F_AA", "Tratamiento", "TAN")
datos.TAN.M3datos.TAN.M3$F_riego <- as.factor(datos.TAN.M3$F_riego)
datos.TAN.M3$F_AA <- as.factor(datos.TAN.M3$F_AA)
str(datos.TAN.M3)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ TAN : num 0.000623 0.000623 0.000623 0.000623 0.000318 ...# Anova - Tasa de asimilación neta - Muestreo 3
A.TAN.M3 <- aov(TAN ~ F_riego*F_AA, data = datos.TAN.M3)
summary(A.TAN.M3)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 5.860e-08 5.860e-08 9.643e+30 <2e-16 ***
## F_AA 2 7.108e-07 3.554e-07 5.847e+31 <2e-16 ***
## Residuals 12 0.000e+00 0.000e+00
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.TAN.M3$residuals)##
## Shapiro-Wilk normality test
##
## data: A.TAN.M3$residuals
## W = 0.59247, p-value = 1.376e-05- Los datos no tienen una distribución normal, por lo tanto, se debe aplicar una prueba que soporte la distribución de estos datos.
TukeyHSD(A.TAN.M3)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAN ~ F_riego * F_AA, data = datos.TAN.M3)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.0001397814 0.0001397814 0.0001397814 0
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 4.267373e-04 4.267373e-04 4.267373e-04 0
## 600 ppm-0 ppm -9.623044e-05 -9.623044e-05 -9.623044e-05 0
## 600 ppm-1000 ppm -5.229677e-04 -5.229677e-04 -5.229677e-04 0- Los dos facotres presentan diferencias significativas entre si.
# Anova - TAN - Muestreo 3 - Comparación entre los tratamientos
A.TAN.M3.CT <- aov(TAN ~ Tratamiento, data=datos.TAN.M3)
summary(A.TAN.M3.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 7.695e-07 2.565e-07 1.414e+31 <2e-16 ***
## Residuals 12 0.000e+00 0.000e+00
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.TAN.M3.CT <- TukeyHSD(A.TAN.M3.CT)
Tukey_A.TAN.M3.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAN ~ Tratamiento, data = datos.TAN.M3)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 5.093640e-04 5.093640e-04 5.093640e-04 0
## 50 + AS 600ppm-50 - AS -1.360374e-05 -1.360374e-05 -1.360374e-05 0
## Control-50 - AS 3.050348e-04 3.050348e-04 3.050348e-04 0
## 50 + AS 600ppm-50 + AS 1000ppm -5.229677e-04 -5.229677e-04 -5.229677e-04 0
## Control-50 + AS 1000ppm -2.043292e-04 -2.043292e-04 -2.043292e-04 0
## Control-50 + AS 600ppm 3.186385e-04 3.186385e-04 3.186385e-04 0- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
Análisis estadístico - Muestreo 4
# Datos - Tasa de asimilación neta - Muestreo 4
datos.TAN.M4 <- data.frame(Ind_M4$`Factor riego`, Ind_M4$`Factor ácido ascórbico`, Ind_M4$Tratamiento, Ind_M4$`Tasa de asimilación neta (gr/cm2*día)`)
colnames(datos.TAN.M4) <- c("F_riego", "F_AA", "Tratamiento", "TAN")
datos.TAN.M4datos.TAN.M4$F_riego <- as.factor(datos.TAN.M4$F_riego)
datos.TAN.M4$F_AA <- as.factor(datos.TAN.M4$F_AA)
str(datos.TAN.M4)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ TAN : num 0.000662 0.000662 0.000662 0.000662 0.000289 ...# Anova - Tasa de asimilación neta - Muestreo 4
A.TAN.M4 <- aov(TAN ~ F_riego*F_AA, data = datos.TAN.M4)
summary(A.TAN.M4)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 1.947e-07 1.947e-07 3.172e+31 <2e-16 ***
## F_AA 2 1.653e-07 8.265e-08 1.347e+31 <2e-16 ***
## Residuals 12 0.000e+00 0.000e+00
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factor independiente, tuvieron diferencias con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.TAN.M4$residuals)##
## Shapiro-Wilk normality test
##
## data: A.TAN.M4$residuals
## W = 0.59123, p-value = 1.339e-05- Los datos no tienen una distribución normal, por lo tanto, se debe aplicar una prueba que soporte la distribución de estos datos.
TukeyHSD(A.TAN.M4)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAN ~ F_riego * F_AA, data = datos.TAN.M4)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.0002547282 0.0002547282 0.0002547282 0
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 1.740115e-05 1.740115e-05 1.740115e-05 0
## 600 ppm-0 ppm 2.191283e-04 2.191283e-04 2.191283e-04 0
## 600 ppm-1000 ppm 2.017271e-04 2.017271e-04 2.017271e-04 0- Los dos factores presentan diferencias significativas entre si.
# Anova - TAN - Muestreo 4 - Comparación entre los tratamientos
A.TAN.M4.CT <- aov(TAN ~ Tratamiento, data=datos.TAN.M4)
summary(A.TAN.M4.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 3.6e-07 1.2e-07 2.975e+31 <2e-16 ***
## Residuals 12 0.0e+00 0.0e+00
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.TAN.M4.CT <- TukeyHSD(A.TAN.M4.CT)
Tukey_A.TAN.M4.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAN ~ Tratamiento, data = datos.TAN.M4)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 7.653350e-05 7.653350e-05 7.653350e-05 0
## 50 + AS 600ppm-50 - AS 2.782606e-04 2.782606e-04 2.782606e-04 0
## Control-50 - AS 3.729930e-04 3.729930e-04 3.729930e-04 0
## 50 + AS 600ppm-50 + AS 1000ppm 2.017271e-04 2.017271e-04 2.017271e-04 0
## Control-50 + AS 1000ppm 2.964595e-04 2.964595e-04 2.964595e-04 0
## Control-50 + AS 600ppm 9.473233e-05 9.473233e-05 9.473233e-05 0- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
Análisis descriptivo - Tasa de asimilación neta
# Datos - Gráfico - Tasa de asimilación neta
Datos.G.TAN <- data.frame(factor(Tratamientos, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), Ind_M1$`Tasa de asimilación neta (gr/cm2*día)`, Ind_M2$`Tasa de asimilación neta (gr/cm2*día)`, Ind_M3$`Tasa de asimilación neta (gr/cm2*día)`, Ind_M4$`Tasa de asimilación neta (gr/cm2*día)`)
colnames(Datos.G.TAN) <- c("Tratamientos", "31", "38","45", "52")
Datos.G.TAN# Unificar la variable respuesta en una sola columna
df.long.TAN = gather(Datos.G.TAN, dds, TAN, 2:5)
df.long.TAN# Diferencias significativas con el control
# M1 M2 M3 M4
DS.TAN <- c("", "", "", "", #T1
"*", "*", "*", "*", #T2
"*", "*", "*", "*", #T3
"*", "*", "*", "*") #T4
# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.TAN = group_by(df.long.TAN, Tratamientos, dds, ) %>% summarise(mean = mean(TAN), sd = sd(TAN))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.TAN.ds <- data.frame(df.sumzd.TAN, DS.TAN)
df.sumzd.TAN.dsG.TAN <- ggplot(df.sumzd.TAN, aes(x = dds, y = mean, group = Tratamientos, colour = Tratamientos)) +
scale_colour_manual(values=c("red", "#EFA30C", "blue", "green"))+
geom_line() +
geom_point( size=2, shape = 21, fill="white") +
theme_test()+
ggtitle("Tasa de asimilación neta \nen los diferentes muestreos")+ theme(plot.title = element_text(family = "Times New Roman",
size = rel(2),
vjust = 0.5,
hjust = 0.5))+
theme(axis.text.x = element_text(size=15),
axis.text.y = element_text(size=15))+
labs(x = "Días después de siembra", y = "gr / cm2 * día")+
theme(axis.title.x = element_text(family = "Times New Roman", size = rel(1.5)), axis.title.y = element_text(family = "Times New Roman", size = rel(1.5)))+
theme(legend.title = element_text(family = "Times New Roman"))
G.TAN
Tasa de asimilación neta 2
- Se realizó haciendo los calculos en un orden diferente y terminó siendo un ajuste beneficioso para el analisis y la explicación de los resultados de esta variable.
# Datos - Gráfico - Tasa de asimilación neta
Datos.G.TAN.2 <- data.frame(factor(Tratamientos, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), Ind_M1$`TAN 2`, Ind_M2$`TAN 2`, Ind_M3$`TAN 2`, Ind_M4$`TAN 2`)
colnames(Datos.G.TAN.2) <- c("Tratamientos", "31", "38","45", "52")
Datos.G.TAN.2# Unificar la variable respuesta en una sola columna
df.long.TAN.2 = gather(Datos.G.TAN.2, dds, TAN2, 2:5)
df.long.TAN.2# Diferencias significativas con el control
# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.TAN.2 = group_by(df.long.TAN.2, Tratamientos, dds, ) %>% summarise(mean = mean(TAN2), sd = sd(TAN2))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.TAN.2.ds <- data.frame(df.sumzd.TAN.2)
df.sumzd.TAN.2.dsG.TAN.2 <- ggplot(df.sumzd.TAN.2, aes(x = dds, y = mean, group = Tratamientos, colour = Tratamientos)) +
scale_colour_manual(values=c("red", "#EFA30C", "blue", "green"))+
geom_line() +
geom_point( size=2, shape = 21, fill="white") +
theme_test()+
theme(axis.text.x = element_text(size=15),
axis.text.y = element_text(size=15))+
ggtitle("Tasa de asimilación neta \nen los diferentes muestreos")+ theme(plot.title = element_text(family = "Times New Roman",
size = rel(2),
vjust = 0.5,
hjust = 0.5))+
labs(x = "Días después de siembra", y = "gr / cm2 * día")+
theme(axis.title.x = element_text(family = "Times New Roman", size = rel(1.5)), axis.title.y = element_text(family = "Times New Roman", size = rel(1.5)))+
theme(legend.title = element_text(family = "Times New Roman"))
G.TAN.2
Tasa absoluta de crecimiento
Análisis estadístico - Muestreo 1
# Datos - Tasa absoluta de crecimiento - Muestreo 1
datos.TAC.M1 <- data.frame(Ind_M1$`Factor riego`, Ind_M1$`Factor ácido ascórbico`, Ind_M1$Tratamiento, Ind_M1$`Tasa absoluta de crecimiento (gr/dia)`)
colnames(datos.TAC.M1) <- c("F_riego", "F_AA", "Tratamiento", "TAC")
datos.TAC.M1datos.TAC.M1$F_riego <- as.factor(datos.TAC.M1$F_riego)
datos.TAC.M1$F_AA <- as.factor(datos.TAC.M1$F_AA)
str(datos.TAC.M1)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ TAC : num 0.0264 0.0264 0.0264 0.0264 0.0188 ...# Anova - Tasa absoluta de crecimiento - Muestreo 1
A.TAC.M1 <- aov(TAC ~ F_riego*F_AA, data = datos.TAC.M1)
summary(A.TAC.M1)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 0.0001725 1.725e-04 1.761e+32 <2e-16 ***
## F_AA 2 0.0000113 5.650e-06 5.766e+30 <2e-16 ***
## Residuals 12 0.0000000 0.000e+00
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Tanto las concentraciones de ácido ascórbico como los dos tipos de riego, como factores independientes, tuvieron diferencias entre si con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.TAC.M1$residuals)##
## Shapiro-Wilk normality test
##
## data: A.TAC.M1$residuals
## W = 0.67722, p-value = 9.572e-05- Los datos no tienen una distribución normal, por lo tanto, se debe aplicar una prueba que soporte la distribución de estos datos.
TukeyHSD(A.TAC.M1)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAC ~ F_riego * F_AA, data = datos.TAC.M1)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.007582796 0.007582796 0.007582796 0
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.001188575 0.001188575 0.001188575 0
## 600 ppm-0 ppm -0.001188038 -0.001188038 -0.001188038 0
## 600 ppm-1000 ppm -0.002376613 -0.002376613 -0.002376613 0- Las concentraciones de ácido áscórbico y los dos tipos de riego presenta diferencias significativas.
# Anova - TAC - Muestreo 1 - Comparación entre los tratamientos
A.TAC.M1.CT <- aov(TAC ~ Tratamiento, data=datos.TAC.M1)
summary(A.TAC.M1.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 0.0001838 6.126e-05 1.797e+31 <2e-16 ***
## Residuals 12 0.0000000 0.000e+00
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.TAC.M1.CT <- TukeyHSD(A.TAC.M1.CT)
Tukey_A.TAC.M1.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAC ~ Tratamiento, data = datos.TAC.M1)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 0.001188710 0.001188710 0.001188710 0
## 50 + AS 600ppm-50 - AS -0.001187903 -0.001187903 -0.001187903 0
## Control-50 - AS 0.007583065 0.007583065 0.007583065 0
## 50 + AS 600ppm-50 + AS 1000ppm -0.002376613 -0.002376613 -0.002376613 0
## Control-50 + AS 1000ppm 0.006394355 0.006394355 0.006394355 0
## Control-50 + AS 600ppm 0.008770968 0.008770968 0.008770968 0- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
Análisis estadístico - Muestreo 2
# Datos - Tasa absoluta de crecimiento - Muestreo 2
datos.TAC.M2 <- data.frame(Ind_M2$`Factor riego`, Ind_M2$`Factor ácido ascórbico`, Ind_M2$Tratamiento, Ind_M2$`Tasa absoluta de crecimiento (gr/dia)`)
colnames(datos.TAC.M2) <- c("F_riego", "F_AA", "Tratamiento", "TAC")
datos.TAC.M2datos.TAC.M2$F_riego <- as.factor(datos.TAC.M2$F_riego)
datos.TAC.M2$F_AA <- as.factor(datos.TAC.M2$F_AA)
str(datos.TAC.M2)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ TAC : num 0.082 0.082 0.082 0.082 0.0459 ...# Anova - Tasa absoluta de crecimiento - Muestreo 2
A.TAC.M2 <- aov(TAC ~ F_riego*F_AA, data = datos.TAC.M2)
summary(A.TAC.M2)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 0.003831 0.003831 1.095e+32 <2e-16 ***
## F_AA 2 0.000194 0.000097 2.779e+30 <2e-16 ***
## Residuals 12 0.000000 0.000000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Tanto las concentraciones de ácido ascórbico como los dos tipos de riego, como factores independientes, tuvieron diferencias entre si con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.TAC.M2$residuals)##
## Shapiro-Wilk normality test
##
## data: A.TAC.M2$residuals
## W = 0.57247, p-value = 9.003e-06- Los datos no tienen una distribución normal, por lo tanto, se debe aplicar una prueba que soporte la distribución de estos datos.
TukeyHSD(A.TAC.M2)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAC ~ F_riego * F_AA, data = datos.TAC.M2)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.0357369 0.0357369 0.0357369 0
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -0.004515476 -0.004515476 -0.004515476 0
## 600 ppm-0 ppm 0.005320238 0.005320238 0.005320238 0
## 600 ppm-1000 ppm 0.009835714 0.009835714 0.009835714 0- Las concentraciones de ácido áscórbico y los dos tipos de riego presenta diferencias significativas.
# Anova - TAC - Muestreo 2 - Comparación entre los tratamientos
A.TAC.M2.CT <- aov(TAC ~ Tratamiento, data=datos.TAC.M2)
summary(A.TAC.M2.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 0.004026 0.001342 1.776e+31 <2e-16 ***
## Residuals 12 0.000000 0.000000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.TAC.M2.CT <- TukeyHSD(A.TAC.M2.CT)
Tukey_A.TAC.M2.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAC ~ Tratamiento, data = datos.TAC.M2)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -0.004314286 -0.004314286 -0.004314286 0
## 50 + AS 600ppm-50 - AS 0.005521429 0.005521429 0.005521429 0
## Control-50 - AS 0.036139286 0.036139286 0.036139286 0
## 50 + AS 600ppm-50 + AS 1000ppm 0.009835714 0.009835714 0.009835714 0
## Control-50 + AS 1000ppm 0.040453571 0.040453571 0.040453571 0
## Control-50 + AS 600ppm 0.030617857 0.030617857 0.030617857 0- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
Análisis estadístico - Muestreo 3
# Datos - Tasa absoluta de crecimiento - Muestreo 3
datos.TAC.M3 <- data.frame(Ind_M3$`Factor riego`, Ind_M3$`Factor ácido ascórbico`, Ind_M3$Tratamiento, Ind_M3$`Tasa absoluta de crecimiento (gr/dia)`)
colnames(datos.TAC.M3) <- c("F_riego", "F_AA", "Tratamiento", "TAC")
datos.TAC.M3datos.TAC.M3$F_riego <- as.factor(datos.TAC.M3$F_riego)
datos.TAC.M3$F_AA <- as.factor(datos.TAC.M3$F_AA)
str(datos.TAC.M3)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ TAC : num 0.1182 0.1182 0.1182 0.1182 0.0263 ...# Anova - Tasa absoluta de crecimiento - Muestreo 3
A.TAC.M3 <- aov(TAC ~ F_riego*F_AA, data = datos.TAC.M3)
summary(A.TAC.M3)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 0.021738 0.021738 5.923e+31 <2e-16 ***
## F_AA 2 0.000935 0.000467 1.273e+30 <2e-16 ***
## Residuals 12 0.000000 0.000000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Tanto las concentraciones de ácido ascórbico como los dos tipos de riego, como factores independientes, tuvieron diferencias entre si con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.TAC.M3$residuals)##
## Shapiro-Wilk normality test
##
## data: A.TAC.M3$residuals
## W = 0.58229, p-value = 1.107e-05- Los datos no tienen una distribución normal, por lo tanto, se debe aplicar una prueba que soporte la distribución de estos datos.
TukeyHSD(A.TAC.M3)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAC ~ F_riego * F_AA, data = datos.TAC.M3)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.08512262 0.08512262 0.08512262 0
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.015847024 0.015847024 0.015847024 0
## 600 ppm-0 ppm -0.002320833 -0.002320833 -0.002320833 0
## 600 ppm-1000 ppm -0.018167857 -0.018167857 -0.018167857 0- Las concentraciones de ácido áscórbico y los dos tipos de riego presenta diferencias significativas.
# Anova - TAC - Muestreo 3 - Comparación entre los tratamientos
A.TAC.M3.CT <- aov(TAC ~ Tratamiento, data=datos.TAC.M3)
summary(A.TAC.M3.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 0.02267 0.007557 1.118e+31 <2e-16 ***
## Residuals 12 0.00000 0.000000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.TAC.M3.CT <- TukeyHSD(A.TAC.M3.CT)
Tukey_A.TAC.M3.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAC ~ Tratamiento, data = datos.TAC.M3)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 0.019228571 0.019228571 0.019228571 0
## 50 + AS 600ppm-50 - AS 0.001060714 0.001060714 0.001060714 0
## Control-50 - AS 0.091885714 0.091885714 0.091885714 0
## 50 + AS 600ppm-50 + AS 1000ppm -0.018167857 -0.018167857 -0.018167857 0
## Control-50 + AS 1000ppm 0.072657143 0.072657143 0.072657143 0
## Control-50 + AS 600ppm 0.090825000 0.090825000 0.090825000 0- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
Análisis estadístico - Muestreo 4
# Datos - Tasa absoluta de crecimiento - Muestreo 4
datos.TAC.M4 <- data.frame(Ind_M4$`Factor riego`, Ind_M4$`Factor ácido ascórbico`, Ind_M4$Tratamiento, Ind_M4$`Tasa absoluta de crecimiento (gr/dia)`)
colnames(datos.TAC.M4) <- c("F_riego", "F_AA", "Tratamiento", "TAC")
datos.TAC.M4datos.TAC.M4$F_riego <- as.factor(datos.TAC.M4$F_riego)
datos.TAC.M4$F_AA <- as.factor(datos.TAC.M4$F_AA)
str(datos.TAC.M4)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ TAC : num 0.1373 0.1373 0.1373 0.1373 0.0289 ...# Anova - Tasa absoluta de crecimiento - Muestreo 4
A.TAC.M4 <- aov(TAC ~ F_riego*F_AA, data = datos.TAC.M4)
summary(A.TAC.M4)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 0.030032 0.030032 6.105e+31 <2e-16 ***
## F_AA 2 0.000666 0.000333 6.774e+29 <2e-16 ***
## Residuals 12 0.000000 0.000000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Tanto las concentraciones de ácido ascórbico como los dos tipos de riego, como factores independientes, tuvieron diferencias entre si con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.TAC.M4$residuals)##
## Shapiro-Wilk normality test
##
## data: A.TAC.M4$residuals
## W = 0.7245, p-value = 0.0003184- Los datos no tienen una distribución normal, por lo tanto, se debe aplicar una prueba que soporte la distribución de estos datos.
TukeyHSD(A.TAC.M4)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAC ~ F_riego * F_AA, data = datos.TAC.M4)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.1000536 0.1000536 0.1000536 0
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 0.002585714 0.002585714 0.002585714 0
## 600 ppm-0 ppm 0.013928571 0.013928571 0.013928571 0
## 600 ppm-1000 ppm 0.011342857 0.011342857 0.011342857 0- Las concentraciones de ácido áscórbico y los dos tipos de riego presenta diferencias significativas.
# Anova - TAC - Muestreo 4 - Comparación entre los tratamientos
A.TAC.M4.CT <- aov(TAC ~ Tratamiento, data=datos.TAC.M4)
summary(A.TAC.M4.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 0.0307 0.01023 1.158e+31 <2e-16 ***
## Residuals 12 0.0000 0.00000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.TAC.M4.CT <- TukeyHSD(A.TAC.M4.CT)
Tukey_A.TAC.M4.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = TAC ~ Tratamiento, data = datos.TAC.M4)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 0.006714286 0.006714286 0.006714286 0
## 50 + AS 600ppm-50 - AS 0.018057143 0.018057143 0.018057143 0
## Control-50 - AS 0.108310714 0.108310714 0.108310714 0
## 50 + AS 600ppm-50 + AS 1000ppm 0.011342857 0.011342857 0.011342857 0
## Control-50 + AS 1000ppm 0.101596429 0.101596429 0.101596429 0
## Control-50 + AS 600ppm 0.090253571 0.090253571 0.090253571 0- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
Análisis descriptivo - Tasa de asimilación neta
# Datos - Gráfico - Tasa de asimilación neta
Datos.G.TAC <- data.frame(factor(Tratamientos, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), Ind_M1$`Tasa absoluta de crecimiento (gr/dia)`, Ind_M2$`Tasa absoluta de crecimiento (gr/dia)`, Ind_M3$`Tasa absoluta de crecimiento (gr/dia)`, Ind_M4$`Tasa absoluta de crecimiento (gr/dia)`)
colnames(Datos.G.TAC) <- c("Tratamientos", "31", "38","45", "52")
Datos.G.TAC# Unificar la variable respuesta en una sola columna
df.long.TAC = gather(Datos.G.TAC, dds, TAC, 2:5)
df.long.TAC# Diferencias significativas con el control
# M1 M2 M3 M4
DS.TAC <- c("", "", "", "", #T1
"*", "*", "*", "*", #T2
"*", "*", "*", "*", #T3
"*", "*", "*", "*") #T4
# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.TAC = group_by(df.long.TAC, Tratamientos, dds, ) %>% summarise(mean = mean(TAC), sd = sd(TAC))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.TAC.ds <- data.frame(df.sumzd.TAC, DS.TAC)
df.sumzd.TAC.dsG.TAC <- ggplot(df.sumzd.TAC, aes(x = dds, y = mean, group = Tratamientos, colour = Tratamientos)) +
scale_colour_manual(values=c("red", "#EFA30C", "blue", "green"))+
geom_line() +
geom_point( size=2, shape = 21, fill="white") +
theme_test()+
ggtitle("Tasa absoluta de crecimiento \nen los diferentes muestreos")+ theme(plot.title = element_text(family = "Times New Roman",
size = rel(2),
vjust = 0.5,
hjust = 0.5))+
theme(axis.text.x = element_text(size=15),
axis.text.y = element_text(size=15))+
labs(x = "Días después de siembra", y = "gr / día")+
theme(axis.title.x = element_text(family = "Times New Roman", size = rel(1.5)), axis.title.y = element_text(family = "Times New Roman", size = rel(1.5)))+
theme(legend.title = element_text(family = "Times New Roman"))
G.TAC
Indice de área foliar
Análisis estadístico - Muestreo 1
# Datos - Indice de área foliar - Muestreo 1
datos.IAF.M1 <- data.frame(Ind_M1$`Factor riego`, Ind_M1$`Factor ácido ascórbico`, Ind_M1$Tratamiento, Ind_M1$`Indice de área foliar`)
colnames(datos.IAF.M1) <- c("F_riego", "F_AA", "Tratamiento", "IAF")
datos.IAF.M1datos.IAF.M1$F_riego <- as.factor(datos.IAF.M1$F_riego)
datos.IAF.M1$F_AA <- as.factor(datos.IAF.M1$F_AA)
str(datos.IAF.M1)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ IAF : num 0.472 0.472 0.472 0.472 0.35 ...# Anova - Indice de área foliar - Muestreo 1
A.IAF.M1 <- aov(IAF ~ F_riego*F_AA, data = datos.IAF.M1)
summary(A.IAF.M1)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 0.08831 0.08831 1.859e+31 <2e-16 ***
## F_AA 2 0.01726 0.00863 1.817e+30 <2e-16 ***
## Residuals 12 0.00000 0.00000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Tanto las concentraciones de ácido ascórbico como los dos tipos de riego, como factores independientes, tuvieron diferencias entre si con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.IAF.M1$residuals)##
## Shapiro-Wilk normality test
##
## data: A.IAF.M1$residuals
## W = 0.59302, p-value = 1.392e-05- Los datos no tienen una distribución normal, por lo tanto, se debe aplicar una prueba que soporte la distribución de estos datos.
TukeyHSD(A.IAF.M1)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = IAF ~ F_riego * F_AA, data = datos.IAF.M1)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.171575 0.171575 0.171575 0
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -0.0678375 -0.0678375 -0.0678375 0
## 600 ppm-0 ppm -0.0302625 -0.0302625 -0.0302625 0
## 600 ppm-1000 ppm 0.0375750 0.0375750 0.0375750 0- Las concentraciones de ácido áscórbico y los dos tipos de riego presenta diferencias significativas.
# Anova - IAF - Muestreo 1 - Comparación entre los tratamientos
A.IAF.M1.CT <- aov(IAF ~ Tratamiento, data=datos.IAF.M1)
summary(A.IAF.M1.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 0.1056 0.03519 5.281e+30 <2e-16 ***
## Residuals 12 0.0000 0.00000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.IAF.M1.CT <- TukeyHSD(A.IAF.M1.CT)
Tukey_A.IAF.M1.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = IAF ~ Tratamiento, data = datos.IAF.M1)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -0.0923625 -0.0923625 -0.0923625 0
## 50 + AS 600ppm-50 - AS -0.0547875 -0.0547875 -0.0547875 0
## Control-50 - AS 0.1225250 0.1225250 0.1225250 0
## 50 + AS 600ppm-50 + AS 1000ppm 0.0375750 0.0375750 0.0375750 0
## Control-50 + AS 1000ppm 0.2148875 0.2148875 0.2148875 0
## Control-50 + AS 600ppm 0.1773125 0.1773125 0.1773125 0- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
Análisis estadístico - Muestreo 2
# Datos - Indice de área foliar - Muestreo 2
datos.IAF.M2 <- data.frame(Ind_M2$`Factor riego`, Ind_M2$`Factor ácido ascórbico`, Ind_M2$Tratamiento, Ind_M2$`Indice de área foliar`)
colnames(datos.IAF.M2) <- c("F_riego", "F_AA", "Tratamiento", "IAF")
datos.IAF.M2datos.IAF.M2$F_riego <- as.factor(datos.IAF.M2$F_riego)
datos.IAF.M2$F_AA <- as.factor(datos.IAF.M2$F_AA)
str(datos.IAF.M2)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ IAF : num 1.317 1.317 1.317 1.317 0.789 ...# Anova - Indice de área foliar - Muestreo 2
A.IAF.M2 <- aov(IAF ~ F_riego*F_AA, data = datos.IAF.M2)
summary(A.IAF.M2)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 1.0991 1.0991 1.577e+31 <2e-16 ***
## F_AA 2 0.0385 0.0193 2.763e+29 <2e-16 ***
## Residuals 12 0.0000 0.0000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Tanto las concentraciones de ácido ascórbico como los dos tipos de riego, como factores independientes, tuvieron diferencias entre si con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.IAF.M2$residuals)##
## Shapiro-Wilk normality test
##
## data: A.IAF.M2$residuals
## W = 0.57127, p-value = 8.78e-06- Los datos no tienen una distribución normal, por lo tanto, se debe aplicar una prueba que soporte la distribución de estos datos.
TukeyHSD(A.IAF.M2)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = IAF ~ F_riego * F_AA, data = datos.IAF.M2)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.6052875 0.6052875 0.6052875 0
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -0.09549375 -0.09549375 -0.09549375 0
## 600 ppm-0 ppm -0.05918125 -0.05918125 -0.05918125 0
## 600 ppm-1000 ppm 0.03631250 0.03631250 0.03631250 0- Las concentraciones de ácido áscórbico y los dos tipos de riego presenta diferencias significativas.
# Anova - IAF - Muestreo 2 - Comparación entre los tratamientos
A.IAF.M2.CT <- aov(IAF ~ Tratamiento, data=datos.IAF.M2)
summary(A.IAF.M2.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 1.138 0.3792 4.585e+30 <2e-16 ***
## Residuals 12 0.000 0.0000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.IAF.M2.CT <- TukeyHSD(A.IAF.M2.CT)
Tukey_A.IAF.M2.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = IAF ~ Tratamiento, data = datos.IAF.M2)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -0.1341625 -0.1341625 -0.1341625 0
## 50 + AS 600ppm-50 - AS -0.0978500 -0.0978500 -0.0978500 0
## Control-50 - AS 0.5279500 0.5279500 0.5279500 0
## 50 + AS 600ppm-50 + AS 1000ppm 0.0363125 0.0363125 0.0363125 0
## Control-50 + AS 1000ppm 0.6621125 0.6621125 0.6621125 0
## Control-50 + AS 600ppm 0.6258000 0.6258000 0.6258000 0- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
Análisis estadístico - Muestreo 3
# Datos - Indice de área foliar - Muestreo 3
datos.IAF.M3 <- data.frame(Ind_M3$`Factor riego`, Ind_M3$`Factor ácido ascórbico`, Ind_M3$Tratamiento, Ind_M3$`Indice de área foliar`)
colnames(datos.IAF.M3) <- c("F_riego", "F_AA", "Tratamiento", "IAF")
datos.IAF.M3datos.IAF.M3$F_riego <- as.factor(datos.IAF.M3$F_riego)
datos.IAF.M3$F_AA <- as.factor(datos.IAF.M3$F_AA)
str(datos.IAF.M3)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ IAF : num 2.027 2.027 2.027 2.027 0.984 ...# Anova - Indice de área foliar - Muestreo 3
A.IAF.M3 <- aov(IAF ~ F_riego*F_AA, data = datos.IAF.M3)
summary(A.IAF.M3)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 3.870 3.870 6.112e+30 <2e-16 ***
## F_AA 2 0.056 0.028 4.409e+28 <2e-16 ***
## Residuals 12 0.000 0.000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Tanto las concentraciones de ácido ascórbico como los dos tipos de riego, como factores independientes, tuvieron diferencias entre si con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.IAF.M3$residuals)##
## Shapiro-Wilk normality test
##
## data: A.IAF.M3$residuals
## W = 0.55562, p-value = 6.354e-06- Los datos no tienen una distribución normal, por lo tanto, se debe aplicar una prueba que soporte la distribución de estos datos.
TukeyHSD(A.IAF.M3)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = IAF ~ F_riego * F_AA, data = datos.IAF.M3)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 1.135852 1.135852 1.135853 0
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -0.11454688 -0.11454688 -0.11454687 0
## 600 ppm-0 ppm -0.07204063 -0.07204063 -0.07204062 0
## 600 ppm-1000 ppm 0.04250625 0.04250625 0.04250625 0- Las concentraciones de ácido áscórbico y los dos tipos de riego presenta diferencias significativas.
# Anova - IAF - Muestreo 3 - Comparación entre los tratamientos
A.IAF.M3.CT <- aov(IAF ~ Tratamiento, data=datos.IAF.M3)
summary(A.IAF.M3.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 3.926 1.309 6.044e+30 <2e-16 ***
## Residuals 12 0.000 0.000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.IAF.M3.CT <- TukeyHSD(A.IAF.M3.CT)
Tukey_A.IAF.M3.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = IAF ~ Tratamiento, data = datos.IAF.M3)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -0.16119375 -0.16119375 -0.16119375 0
## 50 + AS 600ppm-50 - AS -0.11868750 -0.11868750 -0.11868750 0
## Control-50 - AS 1.04255875 1.04255875 1.04255875 0
## 50 + AS 600ppm-50 + AS 1000ppm 0.04250625 0.04250625 0.04250625 0
## Control-50 + AS 1000ppm 1.20375250 1.20375250 1.20375250 0
## Control-50 + AS 600ppm 1.16124625 1.16124625 1.16124625 0- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
Análisis estadístico - Muestreo 4
# Datos - Indice de área foliar - Muestreo 4
datos.IAF.M4 <- data.frame(Ind_M4$`Factor riego`, Ind_M4$`Factor ácido ascórbico`, Ind_M4$Tratamiento, Ind_M4$`Indice de área foliar`)
colnames(datos.IAF.M4) <- c("F_riego", "F_AA", "Tratamiento", "IAF")
datos.IAF.M4datos.IAF.M4$F_riego <- as.factor(datos.IAF.M4$F_riego)
datos.IAF.M4$F_AA <- as.factor(datos.IAF.M4$F_AA)
str(datos.IAF.M4)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ IAF : num 2.319 2.319 2.319 2.319 0.965 ...# Anova - Indice de área foliar - Muestreo 4
A.IAF.M4 <- aov(IAF ~ F_riego*F_AA, data = datos.IAF.M4)
summary(A.IAF.M4)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 5.685 5.685 2.908e+32 <2e-16 ***
## F_AA 2 0.023 0.011 5.841e+29 <2e-16 ***
## Residuals 12 0.000 0.000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Tanto las concentraciones de ácido ascórbico como los dos tipos de riego, como factores independientes, tuvieron diferencias entre si con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.IAF.M4$residuals)##
## Shapiro-Wilk normality test
##
## data: A.IAF.M4$residuals
## W = 0.67844, p-value = 9.863e-05- Los datos no tienen una distribución normal, por lo tanto, se debe aplicar una prueba que soporte la distribución de estos datos.
TukeyHSD(A.IAF.M4)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = IAF ~ F_riego * F_AA, data = datos.IAF.M4)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 1.37659 1.37659 1.37659 0
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -0.07254833 -0.07254833 -0.07254833 0
## 600 ppm-0 ppm 0.02665167 0.02665167 0.02665167 0
## 600 ppm-1000 ppm 0.09920000 0.09920000 0.09920000 0- Las concentraciones de ácido áscórbico y los dos tipos de riego presenta diferencias significativas.
# Anova - IAF - Muestreo 4 - Comparación entre los tratamientos
A.IAF.M4.CT <- aov(IAF ~ Tratamiento, data=datos.IAF.M4)
summary(A.IAF.M4.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 5.708 1.903 8.284e+31 <2e-16 ***
## Residuals 12 0.000 0.000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.IAF.M4.CT <- TukeyHSD(A.IAF.M4.CT)
Tukey_A.IAF.M4.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = IAF ~ Tratamiento, data = datos.IAF.M4)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -0.0840225 -0.0840225 -0.0840225 0
## 50 + AS 600ppm-50 - AS 0.0151775 0.0151775 0.0151775 0
## Control-50 - AS 1.3536413 1.3536412 1.3536413 0
## 50 + AS 600ppm-50 + AS 1000ppm 0.0992000 0.0992000 0.0992000 0
## Control-50 + AS 1000ppm 1.4376638 1.4376637 1.4376638 0
## Control-50 + AS 600ppm 1.3384638 1.3384637 1.3384638 0- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
Análisis descriptivo - Tasa de asimilación neta
# Datos - Gráfico - Tasa de asimilación neta
Datos.G.IAF <- data.frame(factor(Tratamientos, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), Ind_M1$`Indice de área foliar`, Ind_M2$`Indice de área foliar`, Ind_M3$`Indice de área foliar`, Ind_M4$`Indice de área foliar`)
colnames(Datos.G.IAF) <- c("Tratamientos", "31", "38","45", "52")
Datos.G.IAF# Unificar la variable respuesta en una sola columna
df.long.IAF = gather(Datos.G.IAF, dds, IAF, 2:5)
df.long.IAF# Diferencias significativas con el control
# M1 M2 M3 M4
DS.IAF <- c("", "", "", "", #T1
"*", "*", "*", "*", #T2
"*", "*", "*", "*", #T3
"*", "*", "*", "*") #T4
# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.IAF = group_by(df.long.IAF, Tratamientos, dds, ) %>% summarise(mean = mean(IAF), sd = sd(IAF))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.IAF.ds <- data.frame(df.sumzd.IAF, DS.IAF)
df.sumzd.IAF.dsG.IAF <- ggplot(df.sumzd.IAF, aes(x = dds, y = mean, group = Tratamientos, colour = Tratamientos)) +
scale_colour_manual(values=c("red", "yellow", "blue", "green"))+
geom_line() +
geom_point( size=2, shape = 21, fill="white") +
theme_test()+
ggtitle("Indice de área foliar \nen los diferentes muestreos")+ theme(plot.title = element_text(family = "Times New Roman",
size = rel(1.5),
vjust = 0.5,
hjust = 0.5))+
labs(x = "Días después de siembra", y = "Indice")+
theme(axis.title.x = element_text(family = "Times New Roman"), axis.title.y = element_text(family = "Times New Roman"))+
theme(legend.title = element_text(family = "Times New Roman"))
G.IAF
Área foliar
Análisis estadístico - Muestreo 1
# Datos - Área foliar - Muestreo 1
datos.AF.M1 <- data.frame(Var_M1$`Factor riego`, Var_M1$`Factor ácido ascórbico`, Var_M1$Tratamiento, Var_M1$`Área Foliar (cm2)`)
colnames(datos.AF.M1) <- c("F_riego", "F_AA", "Tratamiento", "AF")
datos.AF.M1datos.AF.M1$F_riego <- as.factor(datos.AF.M1$F_riego)
datos.AF.M1$F_AA <- as.factor(datos.AF.M1$F_AA)
str(datos.AF.M1)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ AF : num 86.7 95.7 86.9 108.7 56.3 ...# Anova - Área foliar - Muestreo 1
A.AF.M1 <- aov(AF ~ F_riego*F_AA, data = datos.AF.M1)
summary(A.AF.M1)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 3533 3533 21.396 0.000585 ***
## F_AA 2 690 345 2.091 0.166342
## Residuals 12 1981 165
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factores independientes, tuvieron diferencias entre si con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.AF.M1$residuals)##
## Shapiro-Wilk normality test
##
## data: A.AF.M1$residuals
## W = 0.95392, p-value = 0.5542- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.AF.M1)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = AF ~ F_riego * F_AA, data = datos.AF.M1)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 34.315 18.15126 50.47874 0.0005846
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -13.5675 -34.55988 7.424882 0.2364517
## 600 ppm-0 ppm -6.0525 -27.04488 14.939882 0.7281723
## 600 ppm-1000 ppm 7.5150 -16.72492 31.754915 0.6940372- Los dos tipos de riego presenta diferencias significativas.
# Anova - AF - Muestreo 1 - Comparación entre los tratamientos
A.AF.M1.CT <- aov(AF ~ Tratamiento, data=datos.AF.M1)
summary(A.AF.M1.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 4223 1407.6 8.526 0.00265 **
## Residuals 12 1981 165.1
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.AF.M1.CT <- TukeyHSD(A.AF.M1.CT)
Tukey_A.AF.M1.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = AF ~ Tratamiento, data = datos.AF.M1)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -18.4725 -45.447608 8.502608 0.2296943
## 50 + AS 600ppm-50 - AS -10.9575 -37.932608 16.017608 0.6348450
## Control-50 - AS 24.5050 -2.470108 51.480108 0.0795494
## 50 + AS 600ppm-50 + AS 1000ppm 7.5150 -19.460108 34.490108 0.8406491
## Control-50 + AS 1000ppm 42.9775 16.002392 69.952608 0.0023759
## Control-50 + AS 600ppm 35.4625 8.487392 62.437608 0.0097795- Podemos observar que los tratamientos azul (50 + AS 600ppm) y verde (50 + AS 1000ppm) tienen diferencias signifciativas con respecto a su control. Adicionalmente, el tratamiento amarillo (50 - AS), no llegó al umbral del 5%, sin embargo, puede que estos resultados se puedan tener en cuenta de manera agronómica.
A.AF.M1.DT <- duncan.test(A.AF.M1.CT, 'Tratamiento', console = T)##
## Study: A.AF.M1.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for AF
##
## Mean Square Error: 165.1068
##
## Tratamiento, means
##
## AF std r se Min Max Q25 Q50
## 50 - AS 69.9900 13.001769 4 6.424694 56.30 87.36 63.4100 68.150
## 50 + AS 1000ppm 51.5175 17.459451 4 6.424694 26.78 66.50 46.0625 56.395
## 50 + AS 600ppm 59.0325 8.898945 4 6.424694 51.25 71.77 54.3925 56.555
## Control 94.4950 10.361346 4 6.424694 86.69 108.71 86.8475 91.290
## Q75
## 50 - AS 74.7300
## 50 + AS 1000ppm 61.8500
## 50 + AS 600ppm 61.1950
## Control 98.9375
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 19.79645 20.72120 21.28150
##
## Means with the same letter are not significantly different.
##
## AF groups
## Control 94.4950 a
## 50 - AS 69.9900 b
## 50 + AS 600ppm 59.0325 b
## 50 + AS 1000ppm 51.5175 bA.AF.M1.DT## $statistics
## MSerror Df Mean CV
## 165.1068 12 68.75875 18.68764
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 19.79645
## 3 3.225244 20.72120
## 4 3.312453 21.28150
##
## $means
## AF std r se Min Max Q25 Q50
## 50 - AS 69.9900 13.001769 4 6.424694 56.30 87.36 63.4100 68.150
## 50 + AS 1000ppm 51.5175 17.459451 4 6.424694 26.78 66.50 46.0625 56.395
## 50 + AS 600ppm 59.0325 8.898945 4 6.424694 51.25 71.77 54.3925 56.555
## Control 94.4950 10.361346 4 6.424694 86.69 108.71 86.8475 91.290
## Q75
## 50 - AS 74.7300
## 50 + AS 1000ppm 61.8500
## 50 + AS 600ppm 61.1950
## Control 98.9375
##
## $comparison
## NULL
##
## $groups
## AF groups
## Control 94.4950 a
## 50 - AS 69.9900 b
## 50 + AS 600ppm 59.0325 b
## 50 + AS 1000ppm 51.5175 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 2
# Datos - Área foliar - Muestreo 2
datos.AF.M2 <- data.frame(Var_M2$`Factor riego`, Var_M2$`Factor ácido ascórbico`, Var_M2$Tratamiento, Var_M2$`Área Foliar (cm2)`)
colnames(datos.AF.M2) <- c("F_riego", "F_AA", "Tratamiento", "AF")
datos.AF.M2datos.AF.M2$F_riego <- as.factor(datos.AF.M2$F_riego)
datos.AF.M2$F_AA <- as.factor(datos.AF.M2$F_AA)
str(datos.AF.M2)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ AF : num 108 180 175 213 121 ...# Anova - Área foliar - Muestreo 2
A.AF.M2 <- aov(AF ~ F_riego*F_AA, data = datos.AF.M2)
summary(A.AF.M2)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 22573 22573 23.46 0.000402 ***
## F_AA 2 192 96 0.10 0.905687
## Residuals 12 11544 962
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factores independientes, tuvieron diferencias entre si con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.AF.M2$residuals)##
## Shapiro-Wilk normality test
##
## data: A.AF.M2$residuals
## W = 0.96114, p-value = 0.6825- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.AF.M2)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = AF ~ F_riego * F_AA, data = datos.AF.M2)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 86.7425 47.72663 125.7584 0.0004023
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -5.53125 -56.20246 45.13996 0.9544999
## 600 ppm-0 ppm -5.78375 -56.45496 44.88746 0.9503765
## 600 ppm-1000 ppm -0.25250 -58.76258 58.25758 0.9999269- Los dos tipos de riego presenta diferencias significativas.
# Anova - AF - Muestreo 2 - Comparación entre los tratamientos
A.AF.M2.CT <- aov(AF ~ Tratamiento, data=datos.AF.M2)
summary(A.AF.M2.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 22765 7588 7.888 0.00359 **
## Residuals 12 11544 962
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.AF.M2.CT <- TukeyHSD(A.AF.M2.CT)
Tukey_A.AF.M2.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = AF ~ Tratamiento, data = datos.AF.M2)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -8.3600 -73.47226 56.75226 0.9802648
## 50 + AS 600ppm-50 - AS -8.6125 -73.72476 56.49976 0.9785096
## Control-50 - AS 81.0850 15.97274 146.19726 0.0140108
## 50 + AS 600ppm-50 + AS 1000ppm -0.2525 -65.36476 64.85976 0.9999994
## Control-50 + AS 1000ppm 89.4450 24.33274 154.55726 0.0072115
## Control-50 + AS 600ppm 89.6975 24.58524 154.80976 0.0070692- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.AF.M2.DT <- duncan.test(A.AF.M2.CT, 'Tratamiento', console = T)##
## Study: A.AF.M2.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for AF
##
## Mean Square Error: 961.9756
##
## Tratamiento, means
##
## AF std r se Min Max Q25 Q50
## 50 - AS 87.7875 31.41195 4 15.50787 45.69 120.72 76.5450 92.370
## 50 + AS 1000ppm 79.4275 25.78645 4 15.50787 56.81 111.22 59.3300 74.840
## 50 + AS 600ppm 79.1750 15.63701 4 15.50787 60.54 97.07 70.4925 79.545
## Control 168.8725 44.17845 4 15.50787 107.52 212.80 158.4225 177.585
## Q75
## 50 - AS 103.6125
## 50 + AS 1000ppm 94.9375
## 50 + AS 600ppm 88.2275
## Control 188.0350
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 47.78449 50.01665 51.36908
##
## Means with the same letter are not significantly different.
##
## AF groups
## Control 168.8725 a
## 50 - AS 87.7875 b
## 50 + AS 1000ppm 79.4275 b
## 50 + AS 600ppm 79.1750 bA.AF.M2.DT## $statistics
## MSerror Df Mean CV
## 961.9756 12 103.8156 29.87578
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 47.78449
## 3 3.225244 50.01665
## 4 3.312453 51.36908
##
## $means
## AF std r se Min Max Q25 Q50
## 50 - AS 87.7875 31.41195 4 15.50787 45.69 120.72 76.5450 92.370
## 50 + AS 1000ppm 79.4275 25.78645 4 15.50787 56.81 111.22 59.3300 74.840
## 50 + AS 600ppm 79.1750 15.63701 4 15.50787 60.54 97.07 70.4925 79.545
## Control 168.8725 44.17845 4 15.50787 107.52 212.80 158.4225 177.585
## Q75
## 50 - AS 103.6125
## 50 + AS 1000ppm 94.9375
## 50 + AS 600ppm 88.2275
## Control 188.0350
##
## $comparison
## NULL
##
## $groups
## AF groups
## Control 168.8725 a
## 50 - AS 87.7875 b
## 50 + AS 1000ppm 79.4275 b
## 50 + AS 600ppm 79.1750 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 3
# Datos - Área foliar - Muestreo 3
datos.AF.M3 <- data.frame(Var_M3$`Factor riego`, Var_M3$`Factor ácido ascórbico`, Var_M3$Tratamiento, Var_M3$`Área Foliar (cm2)`)
colnames(datos.AF.M3) <- c("F_riego", "F_AA", "Tratamiento", "AF")
datos.AF.M3datos.AF.M3$F_riego <- as.factor(datos.AF.M3$F_riego)
datos.AF.M3$F_AA <- as.factor(datos.AF.M3$F_AA)
str(datos.AF.M3)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ AF : num 225 226 295 199 118 ...# Anova - Área foliar - Muestreo 3
A.AF.M3 <- aov(AF ~ F_riego*F_AA, data = datos.AF.M3)
summary(A.AF.M3)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 59160 59160 83.881 9.18e-07 ***
## F_AA 2 1167 584 0.828 0.461
## Residuals 12 8463 705
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factores independientes, tuvieron diferencias entre si con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.AF.M3$residuals)##
## Shapiro-Wilk normality test
##
## data: A.AF.M3$residuals
## W = 0.94459, p-value = 0.409- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.AF.M3)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = AF ~ F_riego * F_AA, data = datos.AF.M3)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 140.428 107.0208 173.8352 9e-07
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -17.378125 -60.76522 26.00897 0.5503353
## 600 ppm-0 ppm -8.624375 -52.01147 34.76272 0.8581478
## 600 ppm-1000 ppm 8.753750 -41.34535 58.85285 0.8881978- Los dos tipos de riego presenta diferencias significativas.
# Anova - AF - Muestreo 3 - Comparación entre los tratamientos
A.AF.M3.CT <- aov(AF ~ Tratamiento, data=datos.AF.M3)
summary(A.AF.M3.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 60328 20109 28.51 9.62e-06 ***
## Residuals 12 8463 705
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.AF.M3.CT <- TukeyHSD(A.AF.M3.CT)
Tukey_A.AF.M3.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = AF ~ Tratamiento, data = datos.AF.M3)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -23.87875 -79.63096 31.87346 0.5965738
## 50 + AS 600ppm-50 - AS -15.12500 -70.87721 40.62721 0.8506567
## Control-50 - AS 127.42675 71.67454 183.17896 0.0000993
## 50 + AS 600ppm-50 + AS 1000ppm 8.75375 -46.99846 64.50596 0.9650692
## Control-50 + AS 1000ppm 151.30550 95.55329 207.05771 0.0000181
## Control-50 + AS 600ppm 142.55175 86.79954 198.30396 0.0000331- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.AF.M3.DT <- duncan.test(A.AF.M3.CT, 'Tratamiento', console = T)##
## Study: A.AF.M3.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for AF
##
## Mean Square Error: 705.2817
##
## Tratamiento, means
##
## AF std r se Min Max Q25
## 50 - AS 109.03425 22.838105 4 13.27857 75.192 125.306 106.8682
## 50 + AS 1000ppm 85.15550 9.616382 4 13.27857 76.353 95.340 77.2605
## 50 + AS 600ppm 93.90925 23.015248 4 13.27857 70.058 123.768 80.1155
## Control 236.46100 40.955725 4 13.27857 199.471 295.030 218.6778
## Q50 Q75
## 50 - AS 117.8195 119.9855
## 50 + AS 1000ppm 84.4645 92.3595
## 50 + AS 600ppm 90.9055 104.6993
## Control 225.6715 243.4547
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 40.91535 42.82662 43.98464
##
## Means with the same letter are not significantly different.
##
## AF groups
## Control 236.46100 a
## 50 - AS 109.03425 b
## 50 + AS 600ppm 93.90925 b
## 50 + AS 1000ppm 85.15550 bA.AF.M3.DT## $statistics
## MSerror Df Mean CV
## 705.2817 12 131.14 20.25098
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 40.91535
## 3 3.225244 42.82662
## 4 3.312453 43.98464
##
## $means
## AF std r se Min Max Q25
## 50 - AS 109.03425 22.838105 4 13.27857 75.192 125.306 106.8682
## 50 + AS 1000ppm 85.15550 9.616382 4 13.27857 76.353 95.340 77.2605
## 50 + AS 600ppm 93.90925 23.015248 4 13.27857 70.058 123.768 80.1155
## Control 236.46100 40.955725 4 13.27857 199.471 295.030 218.6778
## Q50 Q75
## 50 - AS 117.8195 119.9855
## 50 + AS 1000ppm 84.4645 92.3595
## 50 + AS 600ppm 90.9055 104.6993
## Control 225.6715 243.4547
##
## $comparison
## NULL
##
## $groups
## AF groups
## Control 236.46100 a
## 50 - AS 109.03425 b
## 50 + AS 600ppm 93.90925 b
## 50 + AS 1000ppm 85.15550 b
##
## attr(,"class")
## [1] "group"Análisis estadístico - Muestreo 4
# Datos - Área foliar - Muestreo 4
datos.AF.M4 <- data.frame(Var_M4$`Factor riego`, Var_M4$`Factor ácido ascórbico`, Var_M4$Tratamiento, Var_M4$`Área Foliar (cm2)`)
colnames(datos.AF.M4) <- c("F_riego", "F_AA", "Tratamiento", "AF")
datos.AF.M4datos.AF.M4$F_riego <- as.factor(datos.AF.M4$F_riego)
datos.AF.M4$F_AA <- as.factor(datos.AF.M4$F_AA)
str(datos.AF.M4)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ AF : num 274.1 214.2 249.2 171.5 81.8 ...# Anova - Área foliar - Muestreo 4
A.AF.M4 <- aov(AF ~ F_riego*F_AA, data = datos.AF.M4)
summary(A.AF.M4)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 54586 54586 84.321 8.93e-07 ***
## F_AA 2 670 335 0.518 0.609
## Residuals 12 7768 647
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Tanto las concentraciones de ácido ascórbico como los dos tipos de riego, como factores independientes, tuvieron diferencias entre si con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.AF.M4$residuals)##
## Shapiro-Wilk normality test
##
## data: A.AF.M4$residuals
## W = 0.94759, p-value = 0.4525- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.AF.M4)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = AF ~ F_riego * F_AA, data = datos.AF.M4)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 134.8899 102.8839 166.896 9e-07
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm 2.868458 -38.69887 44.43578 0.9815169
## 600 ppm-0 ppm 13.954708 -27.61262 55.52203 0.6531372
## 600 ppm-1000 ppm 11.086250 -36.91156 59.08406 0.8141630- Los dos tipos de riego presenta diferencias significativas.
# Anova - AF - Muestreo 4 - Comparación entre los tratamientos
A.AF.M4.CT <- aov(AF ~ Tratamiento, data=datos.AF.M4)
summary(A.AF.M4.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 55256 18419 28.45 9.73e-06 ***
## Residuals 12 7768 647
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.AF.M4.CT <- TukeyHSD(A.AF.M4.CT)
Tukey_A.AF.M4.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = AF ~ Tratamiento, data = datos.AF.M4)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS 7.07425 -46.33956 60.48806 0.9784302
## 50 + AS 600ppm-50 - AS 18.16050 -35.25331 71.57431 0.7471537
## Control-50 - AS 143.30150 89.88769 196.71531 0.0000204
## 50 + AS 600ppm-50 + AS 1000ppm 11.08625 -42.32756 64.50006 0.9249392
## Control-50 + AS 1000ppm 136.22725 82.81344 189.64106 0.0000339
## Control-50 + AS 600ppm 125.14100 71.72719 178.55481 0.0000782- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.AF.M4.DT <- duncan.test(A.AF.M4.CT, 'Tratamiento', console = T)##
## Study: A.AF.M4.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for AF
##
## Mean Square Error: 647.3597
##
## Tratamiento, means
##
## AF std r se Min Max Q25
## 50 - AS 83.98825 12.25592 4 12.72163 67.704 95.400 78.25275
## 50 + AS 1000ppm 91.06250 16.18752 4 12.72163 75.722 113.441 81.92525
## 50 + AS 600ppm 102.14875 13.86812 4 12.72163 89.279 118.999 91.64150
## Control 227.28975 44.55189 4 12.72163 171.548 274.145 203.56100
## Q50 Q75
## 50 - AS 86.4245 92.16000
## 50 + AS 1000ppm 87.5435 96.68075
## 50 + AS 600ppm 100.1585 110.66575
## Control 231.7330 255.46175
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 39.19925 41.03036 42.13981
##
## Means with the same letter are not significantly different.
##
## AF groups
## Control 227.28975 a
## 50 + AS 600ppm 102.14875 b
## 50 + AS 1000ppm 91.06250 b
## 50 - AS 83.98825 bA.AF.M4.DT## $statistics
## MSerror Df Mean CV
## 647.3597 12 126.1223 20.17348
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 39.19925
## 3 3.225244 41.03036
## 4 3.312453 42.13981
##
## $means
## AF std r se Min Max Q25
## 50 - AS 83.98825 12.25592 4 12.72163 67.704 95.400 78.25275
## 50 + AS 1000ppm 91.06250 16.18752 4 12.72163 75.722 113.441 81.92525
## 50 + AS 600ppm 102.14875 13.86812 4 12.72163 89.279 118.999 91.64150
## Control 227.28975 44.55189 4 12.72163 171.548 274.145 203.56100
## Q50 Q75
## 50 - AS 86.4245 92.16000
## 50 + AS 1000ppm 87.5435 96.68075
## 50 + AS 600ppm 100.1585 110.66575
## Control 231.7330 255.46175
##
## $comparison
## NULL
##
## $groups
## AF groups
## Control 227.28975 a
## 50 + AS 600ppm 102.14875 b
## 50 + AS 1000ppm 91.06250 b
## 50 - AS 83.98825 b
##
## attr(,"class")
## [1] "group"Análisis descriptivo - Área foliar
# Datos - Gráfico - Área foliar
Datos.G.AF <- data.frame(factor(Tratamientos, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), Var_M1$`Área Foliar (cm2)`, Var_M2$`Área Foliar (cm2)`,Var_M3$`Área Foliar (cm2)`, Var_M4$`Área Foliar (cm2)`)
colnames(Datos.G.AF) <- c("Tratamientos", "31", "38","45", "52")
Datos.G.AF# Unificar la variable respuesta en una sola columna
df.long.AF = gather(Datos.G.AF, dds, AF, 2:5)
df.long.AF# Diferencias significativas con el control
# M1 M2 M3 M4
DS.AF <- c("a", "a", "a", "a", #T1
"b*", "b*", "b*", "b*", #T2
"b*", "b*", "b*", "b*", #T3
"b*", "b*", "b*", "b*") #T4
# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.AF = group_by(df.long.AF, Tratamientos, dds, ) %>% summarise(mean = mean(AF), sd = sd(AF))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.AF.ds <- data.frame(df.sumzd.AF, DS.AF)
df.sumzd.AF.dsG.AF = ggplot(df.sumzd.AF, aes(x=dds, y=mean, fill=Tratamientos))+
geom_bar(stat = "identity", position = "dodge", color = "black", alpha = 0.5)+
geom_errorbar(aes(ymin=mean-sd, ymax=mean+sd), width = 0.25, size=0.5, position = position_dodge(0.9), alpha = 0.5)+
scale_fill_manual(values=c("red", "yellow", "blue", "green")) +
theme_minimal()+
ggtitle("Área foliar promedio \nen los diferentes muestreos") + theme(plot.title = element_text(family = "Times New Roman",
size = rel(1.5),
vjust = 0.5,
hjust = 0.5))+
labs(x = "Días después de siembra", y = "cm2")+
theme(axis.title.x = element_text(family = "Times New Roman"), axis.title.y = element_text(family = "Times New Roman"))+
theme(legend.title = element_text(family = "Times New Roman"))+
geom_text(aes(x=dds, y= mean+55, label=DS.AF),
position = position_dodge(width = 1), size = 5)
G.AF
Pérdida de electrolitos
Análisis estadístico - Muestreo 3
# Datos - Tasa absoluta de crecimiento - Muestreo 1
datos.PE.2.M3 <- data.frame(PE_M3$`Factor riego`, PE_M3$`Factor ácido ascórbico`, PE_M3$Tratamiento, PE_M3$`Pérdida de electrolitos`)
colnames(datos.PE.2.M3) <- c("F_riego", "F_AA", "Tratamiento", "PE2")
datos.PE.2.M3datos.PE.2.M3$F_riego <- as.factor(datos.PE.2.M3$F_riego)
datos.PE.2.M3$F_AA <- as.factor(datos.PE.2.M3$F_AA)
str(datos.PE.2.M3)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ PE2 : num 13.6 13.6 13.6 13.6 25.6 ...# Anova - PE - Muestreo 3
A.PE.2.M3 <- aov(PE2 ~ F_riego*F_AA, data = datos.PE.2.M3)
summary(A.PE.2.M3)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 214.7 214.71 1.582e+31 <2e-16 ***
## F_AA 2 116.6 58.28 4.293e+30 <2e-16 ***
## Residuals 12 0.0 0.00
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Tanto las concentraciones de ácido ascórbico como los dos tipos de riego, como factores independientes, tuvieron diferencias entre si con una significancia del 99.9%.
# Prueba de normalidad de residuos
shapiro.test(A.PE.2.M3$residuals)##
## Shapiro-Wilk normality test
##
## data: A.PE.2.M3$residuals
## W = 0.58726, p-value = 1.23e-05- Los datos tienen una distribución normal.
TukeyHSD(A.PE.2.M3)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PE2 ~ F_riego * F_AA, data = datos.PE.2.M3)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 -8.459899 -8.459899 -8.459899 0
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -5.816071 -5.816071 -5.816071 0
## 600 ppm-0 ppm -1.239844 -1.239844 -1.239844 0
## 600 ppm-1000 ppm 4.576226 4.576226 4.576226 0- Las concentraciones de ácido áscórbico y los dos tipos de riego presenta diferencias significativas.
# Anova - PE2 - Muestreo 3 - Comparación entre los tratamientos
A.PE.2.M3.CT <- aov(PE2 ~ Tratamiento, data=datos.PE.2.M3)
summary(A.PE.2.M3.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 331.3 110.4 4.561e+30 <2e-16 ***
## Residuals 12 0.0 0.0
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.PE.2.M3.CT <- TukeyHSD(A.PE.2.M3.CT)
Tukey_A.PE.2.M3.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = PE2 ~ Tratamiento, data = datos.PE.2.M3)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -7.580049 -7.580049 -7.580049 0
## 50 + AS 600ppm-50 - AS -3.003823 -3.003823 -3.003823 0
## Control-50 - AS -11.987857 -11.987857 -11.987857 0
## 50 + AS 600ppm-50 + AS 1000ppm 4.576226 4.576226 4.576226 0
## Control-50 + AS 1000ppm -4.407807 -4.407807 -4.407807 0
## Control-50 + AS 600ppm -8.984033 -8.984033 -8.984033 0- Podemos observar que todos los tratamientos tienen diferencias signifciativas con respecto a su control.
A.PE.2.M3.DT <- duncan.test(A.PE.2.M3.CT, 'Tratamiento', console = T)##
## Study: A.PE.2.M3.CT ~ "Tratamiento"
##
## Duncan's new multiple range test
## for PE2
##
## Mean Square Error: 2.421029e-29
##
## Tratamiento, means
##
## PE2 std r se Min Max Q25 Q50
## 50 - AS 25.61576 0 4 2.460198e-15 25.61576 25.61576 25.61576 25.61576
## 50 + AS 1000ppm 18.03571 0 4 2.460198e-15 18.03571 18.03571 18.03571 18.03571
## 50 + AS 600ppm 22.61194 0 4 2.460198e-15 22.61194 22.61194 22.61194 22.61194
## Control 13.62791 0 4 2.460198e-15 13.62791 13.62791 13.62791 13.62791
## Q75
## 50 - AS 25.61576
## 50 + AS 1000ppm 18.03571
## 50 + AS 600ppm 22.61194
## Control 13.62791
##
## Alpha: 0.05 ; DF Error: 12
##
## Critical Range
## 2 3 4
## 7.580623e-15 7.934736e-15 8.149289e-15
##
## Means with the same letter are not significantly different.
##
## PE2 groups
## 50 - AS 25.61576 a
## 50 + AS 600ppm 22.61194 b
## 50 + AS 1000ppm 18.03571 c
## Control 13.62791 dA.PE.2.M3.DT## $statistics
## MSerror Df Mean CV
## 2.421029e-29 12 19.97283 2.463544e-14
##
## $parameters
## test name.t ntr alpha
## Duncan Tratamiento 4 0.05
##
## $duncan
## Table CriticalRange
## 2 3.081307 7.580623e-15
## 3 3.225244 7.934736e-15
## 4 3.312453 8.149289e-15
##
## $means
## PE2 std r se Min Max Q25 Q50
## 50 - AS 25.61576 0 4 2.460198e-15 25.61576 25.61576 25.61576 25.61576
## 50 + AS 1000ppm 18.03571 0 4 2.460198e-15 18.03571 18.03571 18.03571 18.03571
## 50 + AS 600ppm 22.61194 0 4 2.460198e-15 22.61194 22.61194 22.61194 22.61194
## Control 13.62791 0 4 2.460198e-15 13.62791 13.62791 13.62791 13.62791
## Q75
## 50 - AS 25.61576
## 50 + AS 1000ppm 18.03571
## 50 + AS 600ppm 22.61194
## Control 13.62791
##
## $comparison
## NULL
##
## $groups
## PE2 groups
## 50 - AS 25.61576 a
## 50 + AS 600ppm 22.61194 b
## 50 + AS 1000ppm 18.03571 c
## Control 13.62791 d
##
## attr(,"class")
## [1] "group"Análisis descriptivo de Perdida de electrolitos
# Datos - Gráfico - PE2
Tratamientos_3 <- c(
rep("Control", 4),
rep("50 - AS", 4),
rep("50 + AS 600ppm", 4),
rep("50 + AS 1000ppm", 4)
)
Datos.G.PE.2.M3 <- data.frame(factor(Tratamientos_3, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), PE_M3$`Pérdida de electrolitos`)
colnames(Datos.G.PE.2.M3) <- c("Tratamientos", "45")
Datos.G.PE.2.M3# Unificar la variable respuesta en una sola columna
df.long.PE.2 = gather(Datos.G.PE.2.M3, dds, PE2, 2:2)
df.long.PE.2# Diferencias significativas con el control
# M1 M2 M3 M4
DS.PE.2 <- c("d", "a*", "b*", "c*") #T4
# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.PE.2 = group_by(df.long.PE.2, Tratamientos, dds, ) %>% summarise(mean = mean(PE2), sd = sd(PE2))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.PE2.ds <- data.frame(df.sumzd.PE.2, DS.PE.2)
df.sumzd.PE2.dsG.PE2 = ggplot(df.sumzd.PE.2, aes(x=dds, y=mean, fill=Tratamientos, xmax = 4.7))+
geom_bar(stat = "identity", position = "dodge", color = "black", alpha = 0.5)+
geom_errorbar(aes(ymin=mean-sd, ymax=mean+sd), width = 0.25, size=0.5, position = position_dodge(0.9), alpha = 0.5)+
scale_fill_manual(values=c("red", "yellow", "blue", "green")) +
theme_minimal()+
theme(axis.text.x = element_text(size = 15),
axis.text.y = element_text(size = 15))+
ggtitle("Pérdida de electrolitos \nen los diferentes muestreos") + theme(plot.title = element_text(family = "Times New Roman",
size = rel(2),
vjust = 0.5,
hjust = 0.5))+
labs(x = "Días después de siembra", y = "% Pérdida de electrolitos")+
theme(axis.title.x = element_text(family = "Times New Roman", size = rel(1.5)), axis.title.y = element_text(family = "Times New Roman", size = rel(1.5)))+
theme(legend.title = element_text(family = "Times New Roman", size = rel(1.5)))+
geom_text(aes(x=dds, y= mean+2, label=DS.PE.2),
position = position_dodge(width = 1), size = 8)
G.PE2
Relación área foliar y área bajo el dosel
Índice de área foliar - Muestreo 1
# Datos - Índice de área foliar - Valor instantáneo - Muestreo 1
datos.RAA.M1 <- data.frame(Var_M1$`Factor riego`, Var_M1$`Factor ácido ascórbico`, Var_M1$Tratamiento, Var_M1$`Indice de área foliar (Relación)`)
colnames(datos.RAA.M1) <- c("F_riego", "F_AA", "Tratamiento", "RAA")
datos.RAA.M1datos.RAA.M1$F_riego <- as.factor(datos.RAA.M1$F_riego)
datos.RAA.M1$F_AA <- as.factor(datos.RAA.M1$F_AA)
str(datos.RAA.M1)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ RAA : num 0.968 0.652 0.839 1.146 1.61 ...# Anova - Índice de área foliar - Valor instantáneo - Muestreo 1
A.RAA.M1 <- aov(RAA ~ F_riego*F_AA, data = datos.RAA.M1)
summary(A.RAA.M1)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 0.2906 0.29063 3.362 0.0916 .
## F_AA 2 0.1493 0.07465 0.863 0.4463
## Residuals 12 1.0374 0.08645
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego como factor independiente, tiene diferencias entre si, con una significancia del 90%.
# Prueba de normalidad de residuos
shapiro.test(A.RAA.M1$residuals)##
## Shapiro-Wilk normality test
##
## data: A.RAA.M1$residuals
## W = 0.97518, p-value = 0.9139- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una distribución normal.
TukeyHSD(A.RAA.M1)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = RAA ~ F_riego * F_AA, data = datos.RAA.M1)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 -0.3112498 -0.6811069 0.05860733 0.0916275
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -0.20461826 -0.6849640 0.2757275 0.5111439
## 600 ppm-0 ppm -0.06973048 -0.5500762 0.4106152 0.9211799
## 600 ppm-1000 ppm 0.13488778 -0.4197677 0.6895432 0.7965370- Ninguno de los factores tienen diferencias significativas.
# Anova - Índice de área foliar - Valor instantáneo - Muestreo 1 - Comparación entre los tratamientos
A.RAA.M1.CT <- aov(RAA ~ Tratamiento, data=datos.RAA.M1)
summary(A.RAA.M1.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 0.4399 0.14664 1.696 0.221
## Residuals 12 1.0374 0.08645Tukey_A.RAA.M1.CT <- TukeyHSD(A.RAA.M1.CT)
Tukey_A.RAA.M1.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = RAA ~ Tratamiento, data = datos.RAA.M1)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -0.2732054 -0.8904474 0.3440365 0.5718689
## 50 + AS 600ppm-50 - AS -0.1383177 -0.7555596 0.4789242 0.9080910
## Control-50 - AS -0.4484241 -1.0656660 0.1688178 0.1906430
## 50 + AS 600ppm-50 + AS 1000ppm 0.1348878 -0.4823541 0.7521297 0.9139524
## Control-50 + AS 1000ppm -0.1752187 -0.7924606 0.4420232 0.8332379
## Control-50 + AS 600ppm -0.3101065 -0.9273484 0.3071354 0.4717307- Podemos observar que ninguno de los tratamientos tiene diferencias significativas con el control para el muestreo 1.
Índice de área foliar - Muestreo 2
# Datos - Índice de área foliar - Valor instantáneo - Muestreo 2
datos.RAA.M2 <- data.frame(Var_M2$`Factor riego`, Var_M2$`Factor ácido ascórbico`, Var_M2$Tratamiento, Var_M2$`Indice de área foliar (Relación)`)
colnames(datos.RAA.M2) <- c("F_riego", "F_AA", "Tratamiento", "RAA")
datos.RAA.M2datos.RAA.M2$F_riego <- as.factor(datos.RAA.M2$F_riego)
datos.RAA.M2$F_AA <- as.factor(datos.RAA.M2$F_AA)
str(datos.RAA.M2)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ RAA : num 0.904 1.014 0.814 1.094 0.94 ...# Anova - Índice de área foliar - Valor instantáneo - Muestreo 2
A.RAA.M2 <- aov(RAA ~ F_riego*F_AA, data = datos.RAA.M2)
summary(A.RAA.M2)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 0.07602 0.07602 4.945 0.0461 *
## F_AA 2 0.07500 0.03750 2.439 0.1292
## Residuals 12 0.18449 0.01537
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factores independientes, tienen diferencias significativas.
# Prueba de normalidad de residuos
shapiro.test(A.RAA.M2$residuals)##
## Shapiro-Wilk normality test
##
## data: A.RAA.M2$residuals
## W = 0.96058, p-value = 0.6723- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una organización normal.
TukeyHSD(A.RAA.M2)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = RAA ~ F_riego * F_AA, data = datos.RAA.M2)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 0.1591884 0.003213213 0.3151636 0.0461307
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -0.13061786 -0.33318804 0.07195233 0.2378586
## 600 ppm-0 ppm 0.05000436 -0.15256583 0.25257454 0.7911749
## 600 ppm-1000 ppm 0.18062222 -0.05328569 0.41453012 0.1403437- Solo los riegos presentan diferencias significativas entre si.
# Anova - Índice de área foliar - Valor instantáneo - Muestreo 1 - Comparación entre los tratamientos
A.RAA.M2.CT <- aov(RAA ~ Tratamiento, data=datos.RAA.M2)
summary(A.RAA.M2.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 0.1510 0.05034 3.274 0.0588 .
## Residuals 12 0.1845 0.01537
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.RAA.M2.CT <- TukeyHSD(A.RAA.M2.CT)
Tukey_A.RAA.M2.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = RAA ~ Tratamiento, data = datos.RAA.M2)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -0.15077123 -0.411072926 0.1095305 0.3560266
## 50 + AS 600ppm-50 - AS 0.02985098 -0.230450711 0.2901527 0.9857471
## Control-50 - AS 0.11888166 -0.141420030 0.3791834 0.5477590
## 50 + AS 600ppm-50 + AS 1000ppm 0.18062222 -0.079679479 0.4409239 0.2206680
## Control-50 + AS 1000ppm 0.26965290 0.009351201 0.5299546 0.0415677
## Control-50 + AS 600ppm 0.08903068 -0.171271014 0.3493324 0.7438303- Podemos observar que ninguno de los tratamientos tiene diferencias significativas con el control para el muestreo 2.
Índice de área foliar - Muestreo 3
# Datos - Índice de área foliar - Valor instantáneo - Muestreo 3
datos.RAA.M3 <- data.frame(Var_M3$`Factor riego`, Var_M3$`Factor ácido ascórbico`, Var_M3$Tratamiento,Var_M3$`Indice de área foliar (Relación)`)
colnames(datos.RAA.M3) <- c("F_riego", "F_AA", "Tratamiento", "RAA")
datos.RAA.M3datos.RAA.M3$F_riego <- as.factor(datos.RAA.M3$F_riego)
datos.RAA.M3$F_AA <- as.factor(datos.RAA.M3$F_AA)
str(datos.RAA.M3)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ RAA : num 0.926 1.454 0.965 0.947 2.565 ...# Anova - Índice de área foliar - Valor instantáneo - Muestreo 3
A.RAA.M3 <- aov(RAA ~ F_riego*F_AA, data = datos.RAA.M3)
summary(A.RAA.M3)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 9.017 9.017 16.754 0.00149 **
## F_AA 2 0.717 0.358 0.666 0.53188
## Residuals 12 6.459 0.538
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factores independientes, tienen diferencias con una significancia del 95%.
# Prueba de normalidad de residuos
shapiro.test(A.RAA.M3$residuals)##
## Shapiro-Wilk normality test
##
## data: A.RAA.M3$residuals
## W = 0.91672, p-value = 0.1493- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una organización normal.
TukeyHSD(A.RAA.M3)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = RAA ~ F_riego * F_AA, data = datos.RAA.M3)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 -1.733695 -2.656553 -0.8108368 0.0014905
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -0.3205264 -1.5190728 0.878020 0.7603629
## 600 ppm-0 ppm 0.2768678 -0.9216785 1.475414 0.8141218
## 600 ppm-1000 ppm 0.5973942 -0.7865679 1.981356 0.5025482- Solo los riegos tienen diferencias significativas.
# Anova - Temperatura de la hoja - Muestreo 3 - Comparación entre los tratamientos
A.RAA.M3.CT <- aov(RAA ~ Tratamiento, data=datos.RAA.M3)
summary(A.RAA.M3.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 9.734 3.245 6.028 0.00957 **
## Residuals 12 6.459 0.538
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.RAA.M3.CT <- TukeyHSD(A.RAA.M3.CT)
Tukey_A.RAA.M3.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = RAA ~ Tratamiento, data = datos.RAA.M3)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -0.3314410 -1.871567 1.2086852 0.9173678
## 50 + AS 600ppm-50 - AS 0.2659532 -1.274173 1.8060794 0.9544731
## Control-50 - AS -1.7555241 -3.295650 -0.2153978 0.0242533
## 50 + AS 600ppm-50 + AS 1000ppm 0.5973942 -0.942732 2.1375205 0.6664979
## Control-50 + AS 1000ppm -1.4240830 -2.964209 0.1160432 0.0733284
## Control-50 + AS 600ppm -2.0214773 -3.561603 -0.4813510 0.0098863- Podemos observar solo los tratamientos amarillo y azul, tienen diferencias significativas con respecto al control.
Índice de área foliar - Muestreo 4
# Datos - Índice de área foliar - Valor instantáneo - Muestreo 4
datos.RAA.M4 <- data.frame(Var_M4$`Factor riego`, Var_M4$`Factor ácido ascórbico`, Var_M4$Tratamiento, Var_M4$`Indice de área foliar (Relación)`)
colnames(datos.RAA.M4) <- c("F_riego", "F_AA", "Tratamiento", "RAA")
datos.RAA.M4datos.RAA.M4$F_riego <- as.factor(datos.RAA.M4$F_riego)
datos.RAA.M4$F_AA <- as.factor(datos.RAA.M4$F_AA)
str(datos.RAA.M4)## 'data.frame': 16 obs. of 4 variables:
## $ F_riego : Factor w/ 2 levels "0.5","1": 2 2 2 2 1 1 1 1 1 1 ...
## $ F_AA : Factor w/ 3 levels "0 ppm","1000 ppm",..: 1 1 1 1 1 1 1 1 3 3 ...
## $ Tratamiento: chr "Control" "Control" "Control" "Control" ...
## $ RAA : num 1.518 0.749 0.739 0.633 1.304 ...# Anova - Índice de área foliar - Valor instantáneo - Muestreo 4
A.RAA.M4 <- aov(RAA ~ F_riego*F_AA, data = datos.RAA.M4)
summary(A.RAA.M4)## Df Sum Sq Mean Sq F value Pr(>F)
## F_riego 1 1.661 1.661 6.891 0.0222 *
## F_AA 2 1.588 0.794 3.294 0.0724 .
## Residuals 12 2.892 0.241
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- Los dos tipos de riego, como factores independientes, tiene una diferencia significativa del 95%. Los tres tipos de concetraciones de ácido ascórbico, como factor independiente, presenta diferencias significativas del 90%.
# Prueba de normalidad de residuos
shapiro.test(A.RAA.M4$residuals)##
## Shapiro-Wilk normality test
##
## data: A.RAA.M4$residuals
## W = 0.95502, p-value = 0.5732- El hecho de que no sea significativo (p-value > 0.05), representa que los datos tienen una organización normal.
TukeyHSD(A.RAA.M4)## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = RAA ~ F_riego * F_AA, data = datos.RAA.M4)
##
## $F_riego
## diff lwr upr p adj
## 1-0.5 -0.7440306 -1.361597 -0.1264646 0.0221799
##
## $F_AA
## diff lwr upr p adj
## 1000 ppm-0 ppm -0.5835355 -1.3855890 0.2185180 0.1697387
## 600 ppm-0 ppm -0.4299584 -1.2320119 0.3720951 0.3572084
## 600 ppm-1000 ppm 0.1535772 -0.7725544 1.0797088 0.8986181- Solo los riegos presentan diferencias significativas.
# Anova - Temperatura de la hoja - Muestreo 4 - Comparación entre los tratamientos
A.RAA.M4.CT <- aov(RAA ~ Tratamiento, data=datos.RAA.M4)
summary(A.RAA.M4.CT)## Df Sum Sq Mean Sq F value Pr(>F)
## Tratamiento 3 3.249 1.083 4.493 0.0247 *
## Residuals 12 2.892 0.241
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Tukey_A.RAA.M4.CT <- TukeyHSD(A.RAA.M4.CT)
Tukey_A.RAA.M4.CT## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = RAA ~ Tratamiento, data = datos.RAA.M4)
##
## $Tratamiento
## diff lwr upr p adj
## 50 + AS 1000ppm-50 - AS -0.8369090 -1.8675438 0.1937258 0.1277319
## 50 + AS 600ppm-50 - AS -0.6833318 -1.7139667 0.3473030 0.2524823
## Control-50 - AS -1.2507775 -2.2814124 -0.2201427 0.0165230
## 50 + AS 600ppm-50 + AS 1000ppm 0.1535772 -0.8770576 1.1842120 0.9698481
## Control-50 + AS 1000ppm -0.4138685 -1.4445034 0.6167663 0.6428806
## Control-50 + AS 600ppm -0.5674457 -1.5980805 0.4631891 0.3970450- Podemos observar que solo el tratamiento amarillo tuvo diferencias significativas con respecto al control.
Análisis descriptivo - Índice de área foliar - Valor instantáneo
# Datos - Gráfico - Índice de área foliar - Valor instantáneo
Datos.G.RAA <- data.frame(factor(Tratamientos, ordered = TRUE, levels = c("Control", "50 - AS", "50 + AS 600ppm", "50 + AS 1000ppm")), Var_M1$`Indice de área foliar (Relación)`, Var_M2$`Indice de área foliar (Relación)`, Var_M3$`Indice de área foliar (Relación)`, Var_M4$`Indice de área foliar (Relación)`)
colnames(Datos.G.RAA) <- c("Tratamientos", "31", "38","45", "52")
Datos.G.RAA# Unificar la variable respuesta en una sola columna
df.long.RAA = gather(Datos.G.RAA, dds, RAA, 2:5)
df.long.RAA# Media y desviación estandar de la media de los tratamientos en los diferentes muestreos.
df.sumzd.RAA = group_by(df.long.RAA, Tratamientos, dds, ) %>% summarise(mean = mean(RAA), sd = sd(RAA))## `summarise()` has grouped output by 'Tratamientos'. You can override using the
## `.groups` argument.# Con diferencias significativas de los tratamientos con su control
df.sumzd.RAA.ds <- data.frame(df.sumzd.RAA)
df.sumzd.RAA.dsG.RAA <- ggplot(df.sumzd.RAA, aes(x = dds, y = mean, group = Tratamientos, colour = Tratamientos)) +
scale_colour_manual(values=c("red", "#EFA30C", "blue", "green"))+
geom_line() +
geom_point( size=2, shape = 21, fill="white") +
theme_test()+
ggtitle("Indice de área foliar \nen los diferentes muestreos")+ theme(plot.title = element_text(family = "Times New Roman",
size = rel(2),
vjust = 0.5,
hjust = 0.5))+
theme(axis.text.x = element_text(size=15),
axis.text.y = element_text(size=15))+
labs(x = "Días después de siembra", y = "Indice")+
theme(axis.title.x = element_text(family = "Times New Roman", size = rel(1.5)), axis.title.y = element_text(family = "Times New Roman", size=rel(1.5)))+
theme(legend.title = element_text(family = "Times New Roman"))
G.RAA
# Para colocar una imágen: ""Estomas abiertos y cerrados
- Realizado en Excel.
Escala de BBCH
Anexos
Ecuaciones de variables indirectas
Afiche
