A.Gomez_pollos_pesos

R packages

if (!require(tidyverse)) install.packages("tidyverse")
if (!require(googlesheets4)) install.packages("googlesheets4")
if (!require(googledrive)) install.packages("googledrive")
if (!require(plotly)) install.packages("plotly")
if (!require(lme4)) install.packages("lme4")
if (!require(fitdistrplus)) install.packages("fitdistrplus")
if (!require(goft)) install.packages("goft")
if (!require(data.table)) install.packages("data.table")
if (!require(scales)) install.packages("scales")
if (!require(ggbreak)) install.packages("ggbreak")
if (!require(arm)) install.packages("arm")
if (!require(splines)) install.packages("splines")
if (!require(DT)) install.packages("DT")
if (!require(binom)) install.packages("binom")

##
library(DT)
library(binom)
library(splines)
library(arm)
library(ggbreak)
library(scales)
library(tidyverse)
library(googlesheets4)
library(googledrive)
library(plotly)
library(lme4)
library(fitdistrplus)
library(goft)
library(data.table)

Data

ss= "https://docs.google.com/spreadsheets/d/1fLePbF3flzafwVH_kOZh6o2_g5htf0nmrOB2TpSdZ_M/edit?usp=sharing"
hoja = "lico_1"
rango = "B2:I1188"

ss= "https://docs.google.com/spreadsheets/d/1fLePbF3flzafwVH_kOZh6o2_g5htf0nmrOB2TpSdZ_M/edit?usp=sharing"
hoja = 1
rango = "B2:I1188"
df <- read_sheet(ss,
                  sheet = hoja,
                  range = rango,
                  col_names = TRUE
                  )

✔ Reading from Copia de pesaje de pollos.
✔ Range ''lico_1'!B2:I1188'.

df <- data.frame(df)
df$g <- as.numeric(df$g)
df$grupo <- as.factor(df$grupo)
df$replica <- as.factor(df$replica)
df$pseudo.r.dia <- as.factor(df$pseudo.r.dia)
df$dia.grupo <- as.factor(df$dia.grupo)
df$dia.replica <- as.factor(df$dia.replica)
#########################
ss= "https://docs.google.com/spreadsheets/d/1fLePbF3flzafwVH_kOZh6o2_g5htf0nmrOB2TpSdZ_M/edit?usp=sharing"
hoja = 1
rango = "A1193:F1229"
df2 <- read_sheet(ss,
                  sheet = hoja,
                  range = rango,
                  col_names = TRUE
                  )

✔ Reading from Copia de pesaje de pollos.
✔ Range ''lico_1'!A1193:F1229'.

df2$grupo <- as.factor(df2$grupo)

Peso (g): Serie de tiempo

p <- ggplot(data= df,
            aes(x= dia, y= g)
            )
p <- p +
#  geom_boxplot( aes(col= pseudo.r), fill= NA,
 #              position = position_dodge(1),
  #             show.legend = FALSE) +
#   geom_point(aes(col= grupo), size = 1,
 #            position = position_jitterdodge(jitter.height = 0.01,
  #                                           jitter.width = 0.2,
   #                                          dodge.width = 1),
    #         show.legend = FALSE) +
  geom_smooth(aes(col=grupo),
              method = "glm", linewidth = 0.4,
              formula= y ~ bs(x, 3),
              method.args = list(family="gaussian"),
              show.legend = TRUE) +
  
  stat_summary(aes(fill= grupo), fun.data = 'mean_cl_normal' , 
               geom = , size= 0.2,
               col= "darkslategray", linewidth = 0.9,
               position = position_jitterdodge(jitter.height = 0,
                                               jitter.width = 0,
                                               dodge.width = 1),
               show.legend = FALSE) +
#  stat_summary(aes(shape= grupo), col= "darkslategray",
 #              fun = median, geom = "point", size = 3,
  #             position = position_jitterdodge(jitter.height = 0.2,
   #                                            jitter.width = 0.2,
    #                                           dodge.width = 1),
     #          show.legend = FALSE) +
   labs(y = "peso (g)",
       x= "dia",
       title = "Peso (g) de pollos .....") +
  scale_x_continuous(breaks= c(1,7,11, 14,21,28,35),
                     labels = levels(as.factor((df$dia))))
ggplotly(p)

Dias 21-35

df. <- df[df$dia>14,]
p <- ggplot(data= df.,
            aes(x= dia, y= g)
            )
p <- p +
  geom_boxplot( aes(col= pseudo.r.dia), fill= NA,
               position = position_dodge(6),
               show.legend = TRUE) +
   geom_point(aes(col= pseudo.r.dia), size = 1,
             position = position_jitterdodge(jitter.height = 0.01,
                                             jitter.width = 0.2,
                                             dodge.width = 6),
             show.legend = FALSE) +
#  stat_summary(aes(fill= pseudo.r.dia), fun.data = 'mean_cl_normal' , 
 #              geom = , size= 0.05, linewidth = 0.6,
  #             position = position_jitterdodge(jitter.height = 0,
   #                                            jitter.width = 0,
    #                                           dodge.width = 6),
     #          show.legend = FALSE) +
#  geom_smooth(aes(col=grupo),
 #             method = "glm", linewidth = 0.4,
  #            formula= y ~ bs(x, 3), se= FALSE,
   #           method.args = list(family="gaussian"),
    #          show.legend = FALSE) +
#  stat_summary(aes(shape= grupo), col= "darkslategray",
 #              fun = median, geom = "point", size = 3,
  #             position = position_jitterdodge(jitter.height = 0.2,
   #                                            jitter.width = 0.2,
    #                                           dodge.width = 1),
     #          show.legend = FALSE) +
   labs(y = "peso (g)",
       x= "dia",
       title = "Peso (g) de pollos .....") +
  scale_x_continuous(breaks= c(1,7,11, 14,21,28,35),
                     labels = levels(as.factor((df$dia))))
#ggplotly(p)
p

COVARIANZA : GRUPO X DIA

mod1 <-lm(formula = g ~ grupo * dia ,
           #family = gaussian,
           data = df)

summary(mod1)

Call:
lm(formula = g ~ grupo * dia, data = df)

Residuals:
    Min      1Q  Median      3Q     Max 
-420.75 -149.74  -31.98  150.47  750.49 

Coefficients:
             Estimate Std. Error t value Pr(>|t|)    
(Intercept) -268.2086    13.0008 -20.630   <2e-16 ***
grupoE       -17.6211    18.4379  -0.956   0.3394    
dia           65.5620     0.6739  97.292   <2e-16 ***
grupoE:dia     2.1607     0.9652   2.239   0.0254 *  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 178.6 on 1182 degrees of freedom
Multiple R-squared:  0.9416,    Adjusted R-squared:  0.9415 
F-statistic:  6357 on 3 and 1182 DF,  p-value: < 2.2e-16

• La pendiente de aumento de peso (positiva) es mayor para el grupo experimental que para el control. Esto se evidencia por la interacción positiva en el glm arriba. Incluso tomando en cuenta el efecto aleatorio de los tres subgrupos por grupo (modelo con efectos aleatorios abajo), el patrón se mantiene.

mod2 <-lmer(formula = g ~ grupo * dia + (1|dia/replica),
           #family = gaussian,
           data = df)

boundary (singular) fit: see help('isSingular')

summary(mod2)

Linear mixed model fit by REML ['lmerMod']
Formula: g ~ grupo * dia + (1 | dia/replica)
   Data: df

REML criterion at convergence: 14440.6

Scaled residuals: 
    Min      1Q  Median      3Q     Max 
-5.3950 -0.2567 -0.0221  0.2312  5.5860 

Random effects:
 Groups      Name        Variance Std.Dev.
 replica:dia (Intercept)     0      0.0   
 dia         (Intercept) 28086    167.6   
 Residual                11203    105.8   
Number of obs: 1186, groups:  replica:dia, 21; dia, 7

Fixed effects:
             Estimate Std. Error t value
(Intercept) -273.9727   114.8050  -2.386
grupoE       -20.8032    10.9271  -1.904
dia           66.1618     5.7240  11.559
grupoE:dia     2.5071     0.5721   4.382

Correlation of Fixed Effects:
           (Intr) grupoE dia   
grupoE     -0.047              
dia        -0.833  0.041       
grupoE:dia  0.039 -0.827 -0.049
optimizer (nloptwrap) convergence code: 0 (OK)
boundary (singular) fit: see help('isSingular')

##################################
p <- ggplot(data= df,
            aes(x= dia, y= g)
            )
p <- p +
  geom_boxplot( aes(col= dia.grupo), fill= NA,
               position = position_dodge(1),
               show.legend = TRUE) +
#  geom_point(aes(col= dia.grupo), size = 0.7,
 #            position = position_jitterdodge(jitter.height = 0.01,
  #                                           jitter.width = 0.2,
   #                                          dodge.width = 1),
    #         show.legend = FALSE) +
  geom_smooth(aes(group=grupo),
              method = "lm", linewidth= 0.2,
              formula= y ~ bs(x, 3), se= FALSE,
              method.args = list(family="poisson"),
              show.legend = FALSE) +
  
  stat_summary(aes(fill= dia.grupo), fun.data = 'mean_cl_normal' , 
               geom = , size= 0.2,
               col= "darkslategray", linewidth = 0.9,
               position = position_jitterdodge(jitter.height = 0,
                                               jitter.width = 0,
                                               dodge.width = 1),
               show.legend = FALSE) +
#  stat_summary(aes(shape= grupo), col= "darkslategray",
 #              fun = median, geom = "point", size = 3,
  #             position = position_jitterdodge(jitter.height = 0.2,
   #                                            jitter.width = 0.2,
    #                                           dodge.width = 1),
     #          show.legend = FALSE) +
   labs(y = "peso (g)",
       x= "dia",
       title = "Peso (g) de pollos .....") +
  scale_x_continuous(breaks= c(1,7,11, 14,21,28,35),
                     labels = levels(as.factor((df$dia))))
p

AOV (GLM) por DIA

DIA 1

dia = 1
df. <- df[df$dia==dia,]

mod1 <-lmer(formula = g ~ grupo + (1|replica),
           #family = gaussian,
           data = df.)

boundary (singular) fit: see help('isSingular')

summary(mod1)

Linear mixed model fit by REML ['lmerMod']
Formula: g ~ grupo + (1 | replica)
   Data: df.

REML criterion at convergence: 790.3

Scaled residuals: 
    Min      1Q  Median      3Q     Max 
-2.2708 -0.8206  0.2154  0.8419  1.8028 

Random effects:
 Groups   Name        Variance Std.Dev.
 replica  (Intercept) 0.000    0.000   
 Residual             4.717    2.172   
Number of obs: 180, groups:  replica, 3

Fixed effects:
            Estimate Std. Error t value
(Intercept) 35.13222    0.22895 153.451
grupoE      -0.04778    0.32378  -0.148

Correlation of Fixed Effects:
       (Intr)
grupoE -0.707
optimizer (nloptwrap) convergence code: 0 (OK)
boundary (singular) fit: see help('isSingular')

#############  AOV ######################

p <- ggplot(data= df.,
            aes(x= grupo, y= g)
            )
p <- p +
  geom_violin( aes(group=grupo), fill= NA,
               position = position_dodge(1), 
               show.legend = FALSE) +
  geom_point(aes(col= pseudo.r.dia), size = 1.5,
             position = position_jitterdodge(jitter.height = 0,
                                             jitter.width = 2,
                                             dodge.width =0.5),
             show.legend = TRUE) +
#  geom_smooth(aes(col=grupo),
 #             method = "lm",
  #            formula= y ~ bs(x, 3),
   #           method.args = list(family="poisson"),
    #          show.legend = TRUE) +
  
  stat_summary(aes(shape= grupo), fun.data = 'mean_cl_normal' , 
               geom = , size= 0.4,
               col= "darkslategray", linewidth = 1.4,
               position = position_jitterdodge(jitter.height = 0,
                                               jitter.width = 0,
                                               dodge.width = 0.8),
               show.legend = FALSE) +
#  stat_summary(aes(shape= grupo), col= "darkslategray",
 #              fun = median, geom = "point", size = 3,
  #             position = position_jitterdodge(jitter.height = 0.2,
   #                                            jitter.width = 0.2,
    #                                           dodge.width = 1),
     #          show.legend = FALSE) +
   labs(y = paste("Peso (g) de pollos al dia", dia),
        x = "Grupo",
       title = paste("Peso (g) de pollos al dia", dia) ) 
  #scale_x_continuous(breaks= c(1,7,11, 14,21,28,35),
   #                  labels = levels(as.factor((df$dia))))
p

DIA 7

dia = 7
df. <- df[df$dia==dia,]

mod1 <-lmer(formula = g ~ grupo + (1|replica),
           #family = gaussian,
           data = df.)

boundary (singular) fit: see help('isSingular')

summary(mod1)

Linear mixed model fit by REML ['lmerMod']
Formula: g ~ grupo + (1 | replica)
   Data: df.

REML criterion at convergence: 1555.3

Scaled residuals: 
     Min       1Q   Median       3Q      Max 
-2.49623 -0.66986  0.06682  0.68165  2.25275 

Random effects:
 Groups   Name        Variance Std.Dev.
 replica  (Intercept)   0       0.00   
 Residual             347      18.63   
Number of obs: 180, groups:  replica, 3

Fixed effects:
            Estimate Std. Error t value
(Intercept)  177.803      1.963  90.555
grupoE        -8.845      2.777  -3.185

Correlation of Fixed Effects:
       (Intr)
grupoE -0.707
optimizer (nloptwrap) convergence code: 0 (OK)
boundary (singular) fit: see help('isSingular')

#############  AOV ######################

p <- ggplot(data= df.,
            aes(x= grupo, y= g)
            )
p <- p +
  geom_violin( aes(group=grupo), fill= NA,
               position = position_dodge(1), 
               show.legend = FALSE) +
  geom_point(aes(col= pseudo.r.dia), size = 1.5,
             position = position_jitterdodge(jitter.height = 0,
                                             jitter.width = 2,
                                             dodge.width =0.5),
             show.legend = TRUE) +
#  geom_smooth(aes(col=grupo),
 #             method = "lm",
  #            formula= y ~ bs(x, 3),
   #           method.args = list(family="poisson"),
    #          show.legend = TRUE) +
  
  stat_summary(aes(shape= grupo), fun.data = 'mean_cl_normal' , 
               geom = , size= 0.4,
               col= "darkslategray", linewidth = 1.4,
               position = position_jitterdodge(jitter.height = 0,
                                               jitter.width = 0,
                                               dodge.width = 0.8),
               show.legend = FALSE) +
#  stat_summary(aes(shape= grupo), col= "darkslategray",
 #              fun = median, geom = "point", size = 3,
  #             position = position_jitterdodge(jitter.height = 0.2,
   #                                            jitter.width = 0.2,
    #                                           dodge.width = 1),
     #          show.legend = FALSE) +
   labs(y = paste("Peso (g) de pollos al dia", dia),
        x = "Grupo",
       title = paste("Peso (g) de pollos al dia", dia)) 
  #scale_x_continuous(breaks= c(1,7,11, 14,21,28,35),
   #                  labels = levels(as.factor((df$dia))))
p

DIA 11

dia = 11
df. <- df[df$dia==dia,]

mod1 <-lmer(formula = g ~ grupo + (1|replica),
           #family = gaussian,
           data = df.)

boundary (singular) fit: see help('isSingular')

summary(mod1)

Linear mixed model fit by REML ['lmerMod']
Formula: g ~ grupo + (1 | replica)
   Data: df.

REML criterion at convergence: 1672.2

Scaled residuals: 
     Min       1Q   Median       3Q      Max 
-2.18134 -0.71277  0.01036  0.62419  2.90324 

Random effects:
 Groups   Name        Variance Std.Dev.
 replica  (Intercept)   0.0     0.00   
 Residual             669.3    25.87   
Number of obs: 180, groups:  replica, 3

Fixed effects:
            Estimate Std. Error t value
(Intercept)  292.292      2.727 107.185
grupoE        -6.837      3.857  -1.773

Correlation of Fixed Effects:
       (Intr)
grupoE -0.707
optimizer (nloptwrap) convergence code: 0 (OK)
boundary (singular) fit: see help('isSingular')

#############  AOV ######################

p <- ggplot(data= df.,
            aes(x= grupo, y= g)
            )
p <- p +
  geom_violin( aes(group=grupo), fill= NA,
               position = position_dodge(1), 
               show.legend = FALSE) +
  geom_point(aes(col= pseudo.r.dia), size = 1.5,
             position = position_jitterdodge(jitter.height = 0,
                                             jitter.width = 2,
                                             dodge.width =0.5),
             show.legend = TRUE) +
#  geom_smooth(aes(col=grupo),
 #             method = "lm",
  #            formula= y ~ bs(x, 3),
   #           method.args = list(family="poisson"),
    #          show.legend = TRUE) +
  
  stat_summary(aes(shape= grupo), fun.data = 'mean_cl_normal' , 
               geom = , size= 0.4,
               col= "darkslategray", linewidth = 1.4,
               position = position_jitterdodge(jitter.height = 0,
                                               jitter.width = 0,
                                               dodge.width = 0.8),
               show.legend = FALSE) +
#  stat_summary(aes(shape= grupo), col= "darkslategray",
 #              fun = median, geom = "point", size = 3,
  #             position = position_jitterdodge(jitter.height = 0.2,
   #                                            jitter.width = 0.2,
    #                                           dodge.width = 1),
     #          show.legend = FALSE) +
   labs(y = paste("Peso (g) de pollos al dia",dia),
        x = "Grupo",
       title = paste("Peso (g) de pollos al dia",dia) ) 
  #scale_x_continuous(breaks= c(1,7,11, 14,21,28,35),
   #                  labels = levels(as.factor((df$dia))))
p

DIA 14

dia = 14
df. <- df[df$dia==dia,]

mod1 <-lmer(formula = g ~ grupo + (1|replica),
           #family = gaussian,
           data = df.)

boundary (singular) fit: see help('isSingular')

summary(mod1)

Linear mixed model fit by REML ['lmerMod']
Formula: g ~ grupo + (1 | replica)
   Data: df.

REML criterion at convergence: 1836.3

Scaled residuals: 
    Min      1Q  Median      3Q     Max 
-2.0417 -0.6040 -0.0234  0.6610  3.1837 

Random effects:
 Groups   Name        Variance Std.Dev.
 replica  (Intercept)    0      0.00   
 Residual             1682     41.02   
Number of obs: 180, groups:  replica, 3

Fixed effects:
            Estimate Std. Error t value
(Intercept)  496.502      4.323 114.840
grupoE         7.938      6.114   1.298

Correlation of Fixed Effects:
       (Intr)
grupoE -0.707
optimizer (nloptwrap) convergence code: 0 (OK)
boundary (singular) fit: see help('isSingular')

#############  AOV ######################

p <- ggplot(data= df.,
            aes(x= grupo, y= g)
            )
p <- p +
  geom_violin( aes(group=grupo), fill= NA,
               position = position_dodge(1), 
               show.legend = FALSE) +
  geom_point(aes(col= pseudo.r.dia), size = 1.5,
             position = position_jitterdodge(jitter.height = 0,
                                             jitter.width = 2,
                                             dodge.width =0.5),
             show.legend = TRUE) +
#  geom_smooth(aes(col=grupo),
 #             method = "lm",
  #            formula= y ~ bs(x, 3),
   #           method.args = list(family="poisson"),
    #          show.legend = TRUE) +
  
  stat_summary(aes(shape= grupo), fun.data = 'mean_cl_normal' , 
               geom = , size= 0.4,
               col= "darkslategray", linewidth = 1.4,
               position = position_jitterdodge(jitter.height = 0,
                                               jitter.width = 0,
                                               dodge.width = 0.8),
               show.legend = FALSE) +
#  stat_summary(aes(shape= grupo), col= "darkslategray",
 #              fun = median, geom = "point", size = 3,
  #             position = position_jitterdodge(jitter.height = 0.2,
   #                                            jitter.width = 0.2,
    #                                           dodge.width = 1),
     #          show.legend = FALSE) +
   labs(y = paste("Peso (g) de pollos al dia",dia),
        x = "Grupo",
       title = paste("Peso (g) de pollos al dia",dia) ) 
  #scale_x_continuous(breaks= c(1,7,11, 14,21,28,35),
   #                  labels = levels(as.factor((df$dia))))
p

DIA 21

dia = 21
df. <- df[df$dia==dia,]

mod1 <-lmer(formula = g ~ grupo + (1|replica),
           #family = gaussian,
           data = df.)

summary(mod1)

Linear mixed model fit by REML ['lmerMod']
Formula: g ~ grupo + (1 | replica)
   Data: df.

REML criterion at convergence: 2123.1

Scaled residuals: 
    Min      1Q  Median      3Q     Max 
-2.9771 -0.8093  0.0054  0.7608  2.4606 

Random effects:
 Groups   Name        Variance Std.Dev.
 replica  (Intercept)   99.72   9.986  
 Residual             8961.76  94.667  
Number of obs: 179, groups:  replica, 3

Fixed effects:
            Estimate Std. Error t value
(Intercept)  1050.61      11.57  90.779
grupoE         26.11      14.15   1.845

Correlation of Fixed Effects:
       (Intr)
grupoE -0.615

#############  AOV ######################

p <- ggplot(data= df.,
            aes(x= grupo, y= g)
            )
p <- p +
  geom_violin( aes(group=grupo), fill= NA,
               position = position_dodge(1), 
               show.legend = FALSE) +
  geom_point(aes(col= pseudo.r.dia), size = 1.5,
             position = position_jitterdodge(jitter.height = 0,
                                             jitter.width = 2,
                                             dodge.width =0.5),
             show.legend = TRUE) +
#  geom_smooth(aes(col=grupo),
 #             method = "lm",
  #            formula= y ~ bs(x, 3),
   #           method.args = list(family="poisson"),
    #          show.legend = TRUE) +
  
  stat_summary(aes(shape= grupo), fun.data = 'mean_cl_normal' , 
               geom = , size= 0.4,
               col= "darkslategray", linewidth = 1.4,
               position = position_jitterdodge(jitter.height = 0,
                                               jitter.width = 0,
                                               dodge.width = 0.8),
               show.legend = FALSE) +
#  stat_summary(aes(shape= grupo), col= "darkslategray",
 #              fun = median, geom = "point", size = 3,
  #             position = position_jitterdodge(jitter.height = 0.2,
   #                                            jitter.width = 0.2,
    #                                           dodge.width = 1),
     #          show.legend = FALSE) +
  labs(y = paste("Peso (g) de pollos al dia",dia),
        x = "Grupo",
       title = paste("Peso (g) de pollos al dia",dia) ) 
  #scale_x_continuous(breaks= c(1,7,11, 14,21,28,35),
   #                  labels = levels(as.factor((df$dia))))
p

DIA 28

dia = 28
df. <- df[df$dia==dia,]

mod1 <-lmer(formula = g ~ grupo + (1|replica),
           #family = gaussian,
           data = df.)

summary(mod1)

Linear mixed model fit by REML ['lmerMod']
Formula: g ~ grupo + (1 | replica)
   Data: df.

REML criterion at convergence: 1838.2

Scaled residuals: 
    Min      1Q  Median      3Q     Max 
-2.3974 -0.7758 -0.1284  0.8951  1.9984 

Random effects:
 Groups   Name        Variance Std.Dev.
 replica  (Intercept)   184.2   13.57  
 Residual             22979.1  151.59  
Number of obs: 144, groups:  replica, 3

Fixed effects:
            Estimate Std. Error t value
(Intercept)  1594.82      19.21  83.006
grupoE         60.04      25.52   2.352

Correlation of Fixed Effects:
       (Intr)
grupoE -0.632

#############  AOV ######################

p <- ggplot(data= df.,
            aes(x= grupo, y= g)
            )
p <- p +
  geom_violin( aes(group=grupo), fill= NA,
               position = position_dodge(1), 
               show.legend = FALSE) +
  geom_point(aes(col= pseudo.r.dia), size = 1.5,
             position = position_jitterdodge(jitter.height = 0,
                                             jitter.width = 2,
                                             dodge.width =0.5),
             show.legend = TRUE) +
#  geom_smooth(aes(col=grupo),
 #             method = "lm",
  #            formula= y ~ bs(x, 3),
   #           method.args = list(family="poisson"),
    #          show.legend = TRUE) +
  
  stat_summary(aes(shape= grupo), fun.data = 'mean_cl_normal' , 
               geom = , size= 0.4,
               col= "darkslategray", linewidth = 1.4,
               position = position_jitterdodge(jitter.height = 0,
                                               jitter.width = 0,
                                               dodge.width = 0.8),
               show.legend = FALSE) +
#  stat_summary(aes(shape= grupo), col= "darkslategray",
 #              fun = median, geom = "point", size = 3,
  #             position = position_jitterdodge(jitter.height = 0.2,
   #                                            jitter.width = 0.2,
    #                                           dodge.width = 1),
     #          show.legend = FALSE) +
    labs(y = paste("Peso (g) de pollos al dia",dia),
        x = "Grupo",
       title = paste("Peso (g) de pollos al dia",dia) ) 
  #scale_x_continuous(breaks= c(1,7,11, 14,21,28,35),
   #                  labels = levels(as.factor((df$dia))))
p

DIA 35

dia = 35
df. <- df[df$dia==dia,]

mod1 <-lmer(formula = g ~ grupo + (1|replica),
           #family = gaussian,
           data = df.)

boundary (singular) fit: see help('isSingular')

summary(mod1)

Linear mixed model fit by REML ['lmerMod']
Formula: g ~ grupo + (1 | replica)
   Data: df.

REML criterion at convergence: 1948.4

Scaled residuals: 
    Min      1Q  Median      3Q     Max 
-2.4249 -0.7270 -0.1096  0.8189  2.5071 

Random effects:
 Groups   Name        Variance Std.Dev.
 replica  (Intercept)     0      0.0   
 Residual             55268    235.1   
Number of obs: 143, groups:  replica, 3

Fixed effects:
            Estimate Std. Error t value
(Intercept)  2175.79      27.15  80.152
grupoE         69.76      39.37   1.772

Correlation of Fixed Effects:
       (Intr)
grupoE -0.690
optimizer (nloptwrap) convergence code: 0 (OK)
boundary (singular) fit: see help('isSingular')

#############  AOV ######################

p <- ggplot(data= df.,
            aes(x= grupo, y= g)
            )
p <- p +
  geom_violin( aes(group=grupo), fill= NA,
               position = position_dodge(1), 
               show.legend = FALSE) +
  geom_point(aes(col= pseudo.r.dia), size = 1.5,
             position = position_jitterdodge(jitter.height = 0,
                                             jitter.width = 2,
                                             dodge.width =0.5),
             show.legend = TRUE) +
#  geom_smooth(aes(col=grupo),
 #             method = "lm",
  #            formula= y ~ bs(x, 3),
   #           method.args = list(family="poisson"),
    #          show.legend = TRUE) +
  
  stat_summary(aes(shape= grupo), fun.data = 'mean_cl_normal' , 
               geom = , size= 0.4,
               col= "darkslategray", linewidth = 1.4,
               position = position_jitterdodge(jitter.height = 0,
                                               jitter.width = 0,
                                               dodge.width = 0.8),
               show.legend = FALSE) +
#  stat_summary(aes(shape= grupo), col= "darkslategray",
 #              fun = median, geom = "point", size = 3,
  #             position = position_jitterdodge(jitter.height = 0.2,
   #                                            jitter.width = 0.2,
    #                                           dodge.width = 1),
     #          show.legend = FALSE) +
  labs(y = paste("Peso (g) de pollos al dia",dia),
        x = "Grupo",
       title = paste("Peso (g) de pollos al dia",dia) ) 
  #scale_x_continuous(breaks= c(1,7,11, 14,21,28,35),
   #                  labels = levels(as.factor((df$dia))))
p

Mortalidad

#dia = 35
#df. <- df[df$dia==dia,]
### Pooled tables

## pooled by DIA
by_grupo <- df2 %>% group_by(grupo)
mort_by_grupo<- as.data.frame(by_grupo %>% summarise( muertes = sum(muertes), n = sum(n)))
#
by_dia <- df2 %>% group_by(d)
mort_by_dia<- as.data.frame(by_dia %>% summarise( muertes = sum(muertes), n = sum(n)))
#
by_grupo.dia <- df2 %>% group_by(d, grupo)
mort_by_grupo.dia <- as.data.frame(by_grupo.dia %>% summarise(muertes = sum(muertes), n= sum(n)))

`summarise()` has grouped output by 'd'. You can override using the `.groups`
argument.

mort_by_grupo.dia$Pr.mortal <- mort_by_grupo.dia$muertes/ 
  (mort_by_grupo.dia$muertes + mort_by_grupo.dia$n)

CIs <- binom.confint(x= mort_by_grupo.dia$muertes, n= mort_by_grupo.dia$n + mort_by_grupo.dia$muertes, methods="wilson")
mort_by_grupo.dia$lower <- CIs[,5]
mort_by_grupo.dia$upper<- CIs[,6]
datatable(mort_by_grupo.dia)

size= 3
width = 1

p <- ggplot(aes(x=d , y=Pr.mortal, 
                col=grupo, ymin = lower, ymax = upper), 
            data= mort_by_grupo.dia)
p <- p +
  geom_hline(yintercept = mort_by_grupo.dia$lower[9],
               linetype= "dashed", col="dark gray") +
  geom_point(position= position_dodge(width=width), 
               size=size) +
  geom_linerange(position = position_dodge(width=width)) +
  scale_x_continuous(n.breaks= 6,
                     breaks= c(1,7,11,21,28,35),
                     labels= levels(as.factor(mort_by_grupo.dia$d)),
                     name= "Día") +
  scale_y_continuous(n.breaks = 5, 
                     breaks=c(0, 0.25, 0.50,0.75, 1), limits = c(0,1),
                     labels=c("0%","25%", "50%", "75%", "100%"),
                     name= "Mortalidad (%)") +
  labs(title= "Mortalidad (%) de pollos (IC.95% de Wilson)") +
  ylim(0,0.5) +
  scale_colour_grey(start=0, end=0.6) + 
  theme_bw(base_size = 14)

Scale for y is already present.
Adding another scale for y, which will replace the existing scale.

p

### ######
Y <- cbind(mort_by_grupo.dia$muertes, mort_by_grupo.dia$n)
mod1 <-glm(formula = Y ~ d * grupo,
           family = binomial,
           data = mort_by_grupo.dia)

summary(mod1)

Call:
glm(formula = Y ~ d * grupo, family = binomial, data = mort_by_grupo.dia)

Coefficients:
             Estimate Std. Error z value Pr(>|z|)    
(Intercept) -5.515906   0.816304  -6.757 1.41e-11 ***
d            0.086616   0.028287   3.062   0.0022 ** 
grupoe       0.209388   1.082033   0.194   0.8466    
d:grupoe     0.007278   0.037363   0.195   0.8456    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 133.483  on 11  degrees of freedom
Residual deviance:  99.518  on  8  degrees of freedom
AIC: 118.48

Number of Fisher Scoring iterations: 6

• la mortalidad ocurrió mayormente entre los dias 21 y 28, y aunque numéricamente, fue mayor para el grupo experimental, esa diferencia no es “significativa”. no se logra encontrar diferencias entre esas dos proporciones, se ve que algo ocurrió para el grupo control también, no solo para el experimental. *