Statistica
Correlacion entre variables
round(cor(flomel2.5[,-1],use="complete.obs"),2)
## Peso Ala.sin.aplanar rectriz.externa largo.cola.
## Peso 1.00 0.21 0.09 0.23
## Ala.sin.aplanar 0.21 1.00 0.66 0.69
## rectriz.externa 0.09 0.66 1.00 0.77
## largo.cola. 0.23 0.69 0.77 1.00
## Ancho.nar.f 0.15 0.20 -0.04 0.21
## Alto.nar.f 0.20 0.37 0.20 0.28
## Largo.f -0.14 -0.27 -0.30 -0.12
## Curvatura -0.12 -0.19 -0.45 -0.36
## tarso 0.41 0.40 0.12 0.40
## Ancho.nar.f Alto.nar.f Largo.f Curvatura tarso
## Peso 0.15 0.20 -0.14 -0.12 0.41
## Ala.sin.aplanar 0.20 0.37 -0.27 -0.19 0.40
## rectriz.externa -0.04 0.20 -0.30 -0.45 0.12
## largo.cola. 0.21 0.28 -0.12 -0.36 0.40
## Ancho.nar.f 1.00 0.17 0.31 0.13 0.42
## Alto.nar.f 0.17 1.00 0.06 0.06 0.18
## Largo.f 0.31 0.06 1.00 0.05 -0.17
## Curvatura 0.13 0.06 0.05 1.00 0.11
## tarso 0.42 0.18 -0.17 0.11 1.00
Analisis de cluster K-means (calcula cual es el # de grupos mas probable)
## *** : The Hubert index is a graphical method of determining the number of clusters.
## In the plot of Hubert index, we seek a significant knee that corresponds to a
## significant increase of the value of the measure i.e the significant peak in Hubert
## index second differences plot.
##
## *** : The D index is a graphical method of determining the number of clusters.
## In the plot of D index, we seek a significant knee (the significant peak in Dindex
## second differences plot) that corresponds to a significant increase of the value of
## the measure.
##
## All 39 observations were used.
##
## *******************************************************************
## * Among all indices:
## * 8 proposed 2 as the best number of clusters
## * 3 proposed 3 as the best number of clusters
## * 1 proposed 4 as the best number of clusters
## * 6 proposed 5 as the best number of clusters
## * 2 proposed 8 as the best number of clusters
## * 3 proposed 10 as the best number of clusters
##
## ***** Conclusion *****
##
## * According to the majority rule, the best number of clusters is 2
##
##
## *******************************************************************
##
## pred Macho pred Hembra
## hembra-hembra 0 12
## hembra-macho 0 8
## macho-macho 17 2
MEDIDAS QUE VARIAN ENTRE SEXOS/PLUMAJES:
MANOVA con todas las variables
#MANOVA##########
# str(flomel2.5)
fit <- manova(cbind(Peso,Ala.sin.aplanar,rectriz.externa,largo.cola.,Ancho.nar.f,
Alto.nar.f,Largo.f,Curvatura,tarso) ~ sexo.plum,flomel2.5)
summary(fit)
## Df Pillai approx F num Df den Df Pr(>F)
## sexo.plum 2 0.918 2.73 18 58 0.002 **
## Residuals 36
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Analisis univariados (no hay grafico o posthoc para las q no varian significativamente)
Ala sin aplanar
summary(anv<-aov(flomel2.5$Ala.sin.aplanar~flomel2.5$sexo.plum))
## Df Sum Sq Mean Sq F value Pr(>F)
## flomel2.5$sexo.plum 2 70.3 35.1 15.7 1.2e-05 ***
## Residuals 36 80.4 2.2
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
TukeyHSD(anv, "flomel2.5$sexo.plum", ordered = TRUE)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
## factor levels have been ordered
##
## Fit: aov(formula = flomel2.5$Ala.sin.aplanar ~ flomel2.5$sexo.plum)
##
## $`flomel2.5$sexo.plum`
## diff lwr upr p adj
## hembra-hembra-hembra-macho 0.04167 -1.625 1.709 0.9979
## macho-macho-hembra-macho 2.71053 1.171 4.250 0.0004
## macho-macho-hembra-hembra 2.66886 1.322 4.016 0.0001
dev.off()
## null device
## 1
boxplot(flomel2.5$Ala.sin.aplanar~flomel2.5$sexo.plum,main="Largo de ala",xlab="Sexo/plumaje",ylab="Largo de ala (mm)",col=sample(cols,1))
Peso
summary(anv<-aov(flomel2.5$Peso~flomel2.5$sexo.plum))
## Df Sum Sq Mean Sq F value Pr(>F)
## flomel2.5$sexo.plum 2 0.55 0.277 1.71 0.2
## Residuals 36 5.82 0.162
# boxplot(flomel2.5$Peso~flomel2.5$sexo.plum,main="Peso",xlab="Sexo/plumaje",ylab="Peso (g)")
Rectriz externa
summary(anv<-aov(flomel2.5$rectriz.externa~flomel2.5$sexo.plum))
## Df Sum Sq Mean Sq F value Pr(>F)
## flomel2.5$sexo.plum 2 194 97.1 28.2 4.2e-08 ***
## Residuals 36 124 3.4
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
TukeyHSD(anv, "flomel2.5$sexo.plum", ordered = TRUE)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
## factor levels have been ordered
##
## Fit: aov(formula = flomel2.5$rectriz.externa ~ flomel2.5$sexo.plum)
##
## $`flomel2.5$sexo.plum`
## diff lwr upr p adj
## hembra-hembra-hembra-macho 1.354 -0.7149 3.423 0.2588
## macho-macho-hembra-macho 5.174 3.2638 7.085 0.0000
## macho-macho-hembra-hembra 3.820 2.1487 5.492 0.0000
boxplot(flomel2.5$rectriz.externa~flomel2.5$sexo.plum,main="Rectriz externa",xlab="Sexo/plumaje",ylab="Rectriz externa (mm)",col=sample(cols,1))
Largo cola
summary(anv<-aov(flomel2.5$largo.cola.~flomel2.5$sexo.plum))
## Df Sum Sq Mean Sq F value Pr(>F)
## flomel2.5$sexo.plum 2 140 70.2 18 3.8e-06 ***
## Residuals 36 140 3.9
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
TukeyHSD(anv, "flomel2.5$sexo.plum", ordered = TRUE)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
## factor levels have been ordered
##
## Fit: aov(formula = flomel2.5$largo.cola. ~ flomel2.5$sexo.plum)
##
## $`flomel2.5$sexo.plum`
## diff lwr upr p adj
## hembra-hembra-hembra-macho 1.792 -0.4105 3.994 0.1296
## macho-macho-hembra-macho 4.658 2.6245 6.691 0.0000
## macho-macho-hembra-hembra 2.866 1.0872 4.645 0.0010
boxplot(flomel2.5$largo.cola.~flomel2.5$sexo.plum,main="Largo de cola",xlab="Sexo/plumaje",ylab="Largo de cola (mm)",col=sample(cols,1))
Alto narinas
summary(anv<-aov(flomel2.5$Alto.nar.f~flomel2.5$sexo.plum))
## Df Sum Sq Mean Sq F value Pr(>F)
## flomel2.5$sexo.plum 2 0.069 0.0346 1.01 0.38
## Residuals 36 1.240 0.0344
# boxplot(flomel2.5$Alto.nar.f~flomel2.5$sexo.plum,main="Alto narinas",xlab="Sexo/plumaje",ylab="Alto de las narinas (mm)")
Ancho narinas
summary(anv<-aov(flomel2.5$Ancho.nar.f~flomel2.5$sexo.plum))
## Df Sum Sq Mean Sq F value Pr(>F)
## flomel2.5$sexo.plum 2 0.055 0.0276 0.56 0.58
## Residuals 36 1.772 0.0492
# TukeyHSD(anv, "flomel2.5$sexo.plum", ordered = TRUE)
# boxplot(flomel2.5$Ancho.nar.f~flomel2.5$sexo.plum,main="Ancho narinas",xlab="Sexo/plumaje",ylab="Ancho de las narinas (mm)")
Culmen expuesto
summary(anv<-aov(flomel2.5$Largo.f~flomel2.5$sexo.plum))
## Df Sum Sq Mean Sq F value Pr(>F)
## flomel2.5$sexo.plum 2 7.44 3.72 4.5 0.018 *
## Residuals 36 29.73 0.83
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
TukeyHSD(anv, "flomel2.5$sexo.plum", ordered = TRUE)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
## factor levels have been ordered
##
## Fit: aov(formula = flomel2.5$Largo.f ~ flomel2.5$sexo.plum)
##
## $`flomel2.5$sexo.plum`
## diff lwr upr p adj
## hembra-macho-macho-macho 0.85699 -0.07921 1.793 0.0784
## hembra-hembra-macho-macho 0.88424 0.06517 1.703 0.0320
## hembra-hembra-hembra-macho 0.02725 -0.98663 1.041 0.9976
boxplot(flomel2.5$Largo.f~flomel2.5$sexo.plum,main="Culmen expuesto",xlab="Sexo/plumaje",ylab="Culmen expuesto (mm)",col=sample(cols,1))
Tarso
summary(anv<-aov(flomel2.5$tarso~flomel2.5$sexo.plum))
## Df Sum Sq Mean Sq F value Pr(>F)
## flomel2.5$sexo.plum 2 0.074 0.0370 0.63 0.54
## Residuals 36 2.102 0.0584
# boxplot(flomel2.5$tarso~flomel2.5$sexo.plum,main="Tarso",xlab="Sexo/plumaje",ylab="Tarso (mm)")