Analise de Palavras-Chave
Karina Marques
Junho de 2017
setwd(“C:/Users/Karina/Desktop/UFMG/Mestrado/2015 - 2016/Projetos/Projeto 2016/Instrumentos e Dados/Dados”)
Timeline <-read.csv("C:/Users/Karina/Desktop/UFMG/Mestrado/2015 - 2016/Projetos/Projeto 2016/Instrumentos e Dados/Dados/timeline2.csv", sep=",", dec=",",fill=TRUE, header=TRUE)
str(Timeline)## 'data.frame': 157 obs. of 14 variables:
## $ X : Factor w/ 157 levels "2004-01","2004-02",..: 1 2 3 4 5 6 7 8 9 10 ...
## $ Cartilha : int 38 46 58 47 38 44 34 49 42 38 ...
## $ DA : int 49 49 19 54 51 59 42 44 76 58 ...
## $ DAs : int 37 61 100 69 62 63 63 61 46 85 ...
## $ FE : int 31 43 75 79 89 90 55 51 69 64 ...
## $ PA : int 12 18 25 49 39 35 25 17 36 30 ...
## $ DE : int 12 12 25 20 8 16 8 7 20 10 ...
## $ TA : int 3 3 3 3 2 2 2 3 3 3 ...
## $ DA1 : int 10 16 26 18 16 16 16 16 12 22 ...
## $ Dislexia : int 35 26 33 40 82 72 57 49 65 61 ...
## $ Disgrafia : int 13 10 10 5 6 4 4 5 10 3 ...
## $ Discalculia: int 3 3 3 4 5 3 2 2 3 4 ...
## $ Tempo : Factor w/ 157 levels "2004-01-30","2004-03-01",..: 1 2 3 4 5 6 7 8 9 10 ...
## $ tempo : Factor w/ 157 levels "2004-01-29 22:00:00",..: 1 2 3 4 5 6 7 8 9 10 ...tail(Timeline)## X Cartilha DA DAs FE PA DE TA DA1 Dislexia Disgrafia Discalculia
## 152 2016-08 24 17 18 4 4 5 1 5 20 1 2
## 153 2016-09 25 19 22 8 6 5 1 6 31 2 2
## 154 2016-10 24 17 19 8 5 5 1 5 21 2 2
## 155 2016-11 23 22 23 8 6 7 1 6 20 2 3
## 156 2016-12 14 18 16 3 5 5 1 4 17 1 2
## 157 2017-01 15 9 11 3 3 2 0 3 14 1 1
## Tempo tempo
## 152 2016-08-30 2016-08-29 21:00:00
## 153 2016-09-30 2016-09-29 21:00:00
## 154 2016-10-30 2016-10-29 22:00:00
## 155 2016-11-30 2016-11-29 22:00:00
## 156 2016-12-30 2016-12-29 22:00:00
## 157 2017-01-30 2017-01-29 22:00:00Timeline_5anos <-read.csv("C:/Users/Karina/Desktop/UFMG/Mestrado/2015 - 2016/Projetos/Projeto 2016/Instrumentos e Dados/Dados/Timeline_5anos.csv", sep=",", dec=",",fill=TRUE, header=TRUE)summary(Timeline)## X Cartilha DA DAs
## 2004-01: 1 Min. : 14.00 Min. : 7.00 Min. : 9.00
## 2004-02: 1 1st Qu.: 24.00 1st Qu.:16.00 1st Qu.: 23.00
## 2004-03: 1 Median : 28.00 Median :19.00 Median : 29.00
## 2004-04: 1 Mean : 29.56 Mean :22.77 Mean : 32.03
## 2004-05: 1 3rd Qu.: 33.00 3rd Qu.:27.00 3rd Qu.: 38.00
## 2004-06: 1 Max. :100.00 Max. :76.00 Max. :100.00
## (Other):151
## FE PA DE TA
## Min. : 3.0 Min. : 2.00 Min. : 2.000 Min. :0.0000
## 1st Qu.: 7.0 1st Qu.: 5.00 1st Qu.: 5.000 1st Qu.:0.0000
## Median :12.0 Median : 7.00 Median : 6.000 Median :1.0000
## Mean :20.2 Mean :10.11 Mean : 6.739 Mean :0.9045
## 3rd Qu.:27.0 3rd Qu.:12.00 3rd Qu.: 8.000 3rd Qu.:1.0000
## Max. :90.0 Max. :49.00 Max. :25.000 Max. :4.0000
##
## DA1 Dislexia Disgrafia Discalculia
## Min. : 2.000 Min. : 13.00 Min. : 0.000 Min. :0.000
## 1st Qu.: 6.000 1st Qu.: 21.00 1st Qu.: 1.000 1st Qu.:1.000
## Median : 7.000 Median : 25.00 Median : 2.000 Median :2.000
## Mean : 8.344 Mean : 30.59 Mean : 2.363 Mean :2.089
## 3rd Qu.:10.000 3rd Qu.: 35.00 3rd Qu.: 3.000 3rd Qu.:2.000
## Max. :26.000 Max. :100.00 Max. :13.000 Max. :6.000
##
## Tempo tempo
## 2004-01-30: 1 2004-01-29 22:00:00: 1
## 2004-03-01: 1 2004-02-29 21:00:00: 1
## 2004-03-30: 1 2004-03-29 21:00:00: 1
## 2004-04-30: 1 2004-04-29 21:00:00: 1
## 2004-05-30: 1 2004-05-29 21:00:00: 1
## 2004-06-30: 1 2004-06-29 21:00:00: 1
## (Other) :151 (Other) :151Select Study period (months 97:156 - 5 year exactly)
Timeline_5anos <- Timeline[97:156, ]
head(Timeline_5anos)## X Cartilha DA DAs FE PA DE TA DA1 Dislexia Disgrafia Discalculia
## 97 2012-01 15 10 19 5 3 3 0 5 22 1 1
## 98 2012-02 20 15 19 6 7 4 0 5 20 1 1
## 99 2012-03 26 18 30 9 7 5 0 8 27 2 2
## 100 2012-04 25 25 29 11 8 5 1 8 27 2 1
## 101 2012-05 29 24 35 12 10 7 1 9 24 2 2
## 102 2012-06 25 18 31 12 7 6 1 8 23 2 2
## Tempo tempo
## 97 2012-01-30 2012-01-29 22:00:00
## 98 2012-03-01 2012-02-29 21:00:00
## 99 2012-03-30 2012-03-29 21:00:00
## 100 2012-04-30 2012-04-29 21:00:00
## 101 2012-05-30 2012-05-29 21:00:00
## 102 2012-06-30 2012-06-29 21:00:00tail(Timeline_5anos)## X Cartilha DA DAs FE PA DE TA DA1 Dislexia Disgrafia Discalculia
## 151 2016-07 21 14 15 4 3 4 1 4 22 1 1
## 152 2016-08 24 17 18 4 4 5 1 5 20 1 2
## 153 2016-09 25 19 22 8 6 5 1 6 31 2 2
## 154 2016-10 24 17 19 8 5 5 1 5 21 2 2
## 155 2016-11 23 22 23 8 6 7 1 6 20 2 3
## 156 2016-12 14 18 16 3 5 5 1 4 17 1 2
## Tempo tempo
## 151 2016-07-30 2016-07-29 21:00:00
## 152 2016-08-30 2016-08-29 21:00:00
## 153 2016-09-30 2016-09-29 21:00:00
## 154 2016-10-30 2016-10-29 22:00:00
## 155 2016-11-30 2016-11-29 22:00:00
## 156 2016-12-30 2016-12-29 22:00:00write.csv(Timeline_5anos,file="Timeline_5anos.csv")
summary(Timeline_5anos)## X Cartilha DA DAs
## 2012-01: 1 Min. :14.00 Min. : 7.00 Min. : 9.00
## 2012-02: 1 1st Qu.:21.75 1st Qu.:14.00 1st Qu.:19.00
## 2012-03: 1 Median :25.00 Median :17.00 Median :23.00
## 2012-04: 1 Mean :23.75 Mean :16.27 Mean :22.45
## 2012-05: 1 3rd Qu.:26.00 3rd Qu.:19.00 3rd Qu.:26.00
## 2012-06: 1 Max. :32.00 Max. :25.00 Max. :35.00
## (Other):54
## FE PA DE TA
## Min. : 3.000 Min. : 2.000 Min. :2.000 Min. :0.00
## 1st Qu.: 5.000 1st Qu.: 4.000 1st Qu.:4.000 1st Qu.:0.00
## Median : 7.000 Median : 5.000 Median :5.000 Median :1.00
## Mean : 6.833 Mean : 5.267 Mean :5.183 Mean :0.65
## 3rd Qu.: 8.000 3rd Qu.: 6.000 3rd Qu.:6.000 3rd Qu.:1.00
## Max. :12.000 Max. :10.000 Max. :8.000 Max. :1.00
##
## DA1 Dislexia Disgrafia Discalculia
## Min. :2.000 Min. :13.00 Min. :1.00 Min. :1.000
## 1st Qu.:5.000 1st Qu.:18.75 1st Qu.:1.00 1st Qu.:1.000
## Median :6.000 Median :21.00 Median :1.00 Median :2.000
## Mean :5.833 Mean :21.12 Mean :1.45 Mean :1.783
## 3rd Qu.:7.000 3rd Qu.:23.00 3rd Qu.:2.00 3rd Qu.:2.000
## Max. :9.000 Max. :35.00 Max. :2.00 Max. :3.000
##
## Tempo tempo
## 2012-01-30: 1 2012-01-29 22:00:00: 1
## 2012-03-01: 1 2012-02-29 21:00:00: 1
## 2012-03-30: 1 2012-03-29 21:00:00: 1
## 2012-04-30: 1 2012-04-29 21:00:00: 1
## 2012-05-30: 1 2012-05-29 21:00:00: 1
## 2012-06-30: 1 2012-06-29 21:00:00: 1
## (Other) :54 (Other) :54Organizando o banco por mes
#2004 a 2017
Timeline$date <- seq(as.Date("2004/01/30"), as.Date("2017/01/30"), by = "month")
Timeline$Date <- as.POSIXct(Timeline$date, "%d.%m.%y", tz="America/Sao_Paulo")
#2012 A 2016
Timeline_5anos$date <- seq(as.Date("2012/01/30"), as.Date("2017/01/01"), by = "month")
Timeline_5anos$Date <- as.POSIXct(Timeline_5anos$date, "%d.%m.%y", tz="America/Sao_Paulo")Retirando outliers # borrowed from aL3xa - # http://stackoverflow.com/questions/4787332/how-to-remove-outliers-from-a-dataset
remove_outliers <- function(x, na.rm = TRUE, ...) {
qnt <- quantile(x, probs=c(.25, .75), na.rm = na.rm, ...)
H <- 1.5 * IQR(x, na.rm = na.rm)
y <- x
y[x < (qnt[1] - H)] <- NA
y[x > (qnt[2] + H)] <- NA
y
}
#2004 a 2017
Timeline$DA_clean <- remove_outliers(Timeline$DA)
Timeline$DAs_clean <- remove_outliers(Timeline$DAs)
Timeline$Dislexia_clean <- remove_outliers(Timeline$Dislexia)
Timeline$FE_clean <- remove_outliers(Timeline$FE)
Timeline$PA_clean <- remove_outliers(Timeline$PA)
Timeline$DE_clean <- remove_outliers(Timeline$DE)
Timeline$Discalculia_clean <- remove_outliers(Timeline$Discalculia)
Timeline$Disgrafia_clean <- remove_outliers(Timeline$Disgrafia)
Timeline$TA_clean <- remove_outliers(Timeline$TA)
#2012 a 2017
Timeline_5anos$DA_clean <- remove_outliers(Timeline_5anos$DA)
Timeline_5anos$DAs_clean <- remove_outliers(Timeline_5anos$DAs)
Timeline_5anos$Dislexia_clean <- remove_outliers(Timeline_5anos$Dislexia)
Timeline_5anos$FE_clean <- remove_outliers(Timeline_5anos$FE)
Timeline_5anos$PA_clean <- remove_outliers(Timeline_5anos$PA)
Timeline_5anos$DE_clean <- remove_outliers(Timeline_5anos$DE)
Timeline_5anos$Discalculia_clean <- remove_outliers(Timeline_5anos$Discalculia)
Timeline_5anos$Disgrafia_clean <- remove_outliers(Timeline_5anos$Disgrafia)
Timeline_5anos$TA_clean <- remove_outliers(Timeline_5anos$TA)Relacao das Palavras x Tempo
GRafico Comparativo de frequencia - 2004 a 2017
##Grafico
library(scales) ##instalar e citar o pacote scales
#grafico de frequencia
plot(Timeline$Dislexia~Timeline$Date, type = "l", lty=1, lwd=2, col="red", xaxt="n", axes=F, ylim=c(0,100), xlab="", ylab="Frequencia relativa", main = "Frequencia relativa pelo tempo, 2004 a 2017")
axis.POSIXct(1, at=seq(as.Date("2004/01/30"), as.Date("2017/01/30"), by = "4 month"), format="%Y %b", las=2, cex.axis=0.9)
lines(Timeline$DAs~Timeline$Date, type = "l", lty=1, lwd=2,col= "darkblue")
lines(Timeline$FE~Timeline$Date, type = "l", lty=1, lwd=2,col= "darkgray")
lines(Timeline$PA~Timeline$Date, type = "l", lty=1, lwd=2,col= "pink")
lines(Timeline$DE~Timeline$Date, type = "l", lty=1, lwd=2,col= "black")
lines(Timeline$DA~Timeline$Date, type = "l", lty=1, lwd=2,col= "darkred")
lines(Timeline$Discalculia~Timeline$Date, type = "l", lty=1, lwd=2,col= "green")
lines(Timeline$Disgrafia~Timeline$Date, type = "l", lty=1, lwd=2,col= "orange")
lines(Timeline$TA~Timeline$Date, type = "l", lty=1, lwd=2,col= "purple")
legend("topright", c("Dislexia", "Dif.s de Aprendizagem", "Dif. de Aprendizagem", "Fracasso Escolar", "Transt. da Aprendizagem", "Disgrafia", "Probl. de Aprendizagem", "Discalculia"), bty="n", lty=c(1),
col=c("red","darkblue", "darkred","darkgray", "purple", "orange", "pink", "green"), box.col="white",lwd=c(2,1.5), cex=0.9)
#removidos outliers
plot(Timeline$Dislexia_clean~Timeline$Date, type = "l", lty=1, lwd=2, col="red", xaxt="n", axes=F, ylim=c(0,100), xlab="", ylab="Frequencia relativa", main = "Frequencia relativa pelo tempo, 2004 a 2017, removidos outliers")
axis.POSIXct(1, at=seq(as.Date("2004/01/30"), as.Date("2017/01/30"), by = "4 month"), format="%Y %b", las=2, cex.axis=0.9)
lines(Timeline$DAs_clean~Timeline$Date, type = "h", lty=1, lwd=2,col= "darkblue")
lines(Timeline$FE_clean~Timeline$Date, type = "h", lty=1, lwd=2,col= "darkgray")
lines(Timeline$PA_clean~Timeline$Date, type = "h", lty=1, lwd=2,col= "pink")
lines(Timeline$DE_clean~Timeline$Date, type = "h", lty=1, lwd=2,col= "black")
lines(Timeline$DA_clean~Timeline$Date, type = "h", lty=1, lwd=2,col= "darkred")
lines(Timeline$Discalculia_clean~Timeline$Date, type = "h", lty=1, lwd=2,col= "green")
lines(Timeline$Disgrafia_clean~Timeline$Date, type = "h", lty=1, lwd=2,col= "orange")
lines(Timeline$TA_clean~Timeline$Date, type = "h", lty=1, lwd=2,col= "purple")
legend("topright", c("Dislexia", "Dif.s de Aprendizagem", "Dif. de Aprendizagem", "Fracasso Escolar", "Transt. da Aprendizagem", "Disgrafia", "Probl. de Aprendizagem", "Discalculia"), bty="n", lty=c(1),
col=c("red","darkblue", "darkred","darkgray", "purple", "orange", "pink", "green"), box.col="white",lwd=c(2,1.5), cex=0.9)
GRafico Comparativo de frequencia - ultimos 5 anos
plot(Timeline_5anos$Dislexia~Timeline_5anos$Date, type = "l", lty=1, lwd=2, col="red", xaxt="n", axes=F, ylim=c(0,100), xlab="", ylab="Frequencia relativa", main = "Frequencia relativa pelo tempo, 2012 a 2017")
axis.POSIXct(1, at=seq(as.Date("2012/01/30"), as.Date("2017/01/01"), by = "4 month"), format="%Y %b", las=2, cex.axis=0.9)
lines(Timeline_5anos$DAs~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "darkblue")
lines(Timeline_5anos$FE~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "darkgray")
lines(Timeline_5anos$PA~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "pink")
lines(Timeline_5anos$DE~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "black")
lines(Timeline_5anos$DA~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "darkred")
lines(Timeline_5anos$Discalculia~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "green")
lines(Timeline_5anos$Disgrafia~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "orange")
lines(Timeline_5anos$TA~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "purple")
legend("topright", c("Dislexia", "Dif.s de Aprendizagem", "Dif. de Aprendizagem", "Fracasso Escolar", "Transt. da Aprendizagem", "Disgrafia", "Probl. de Aprendizagem", "Discalculia"), bty="n", lty=c(1),
col=c("red","darkblue", "darkred","darkgray", "purple", "orange", "pink", "darkgreen"), box.col="white",lwd=c(2), cex=0.9)
#removidos outliers
plot(Timeline_5anos$Dislexia_clean~Timeline_5anos$Date, type = "l", lty=1, lwd=2, col="red", xaxt="n", axes=F, ylim=c(0,100), xlab="", ylab="Frequencia relativa", main = "Frequencia relativa pelo tempo, 2012 a 2017, removidos outliers")
axis.POSIXct(1, at=seq(as.Date("2012/01/30"), as.Date("2017/01/01"), by = "4 month"), format="%Y %b", las=2, cex.axis=0.9)
lines(Timeline_5anos$DAs_clean~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "darkblue")
lines(Timeline_5anos$FE~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "darkgray")
lines(Timeline_5anos$PA_clean~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "pink")
lines(Timeline_5anos$DE_clean~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "black")
lines(Timeline_5anos$DA_clean~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "darkred")
lines(Timeline_5anos$Discalculia_clean~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "green")
lines(Timeline_5anos$Disgrafia_clean~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "orange")
lines(Timeline_5anos$TA_clean~Timeline_5anos$Date, type = "h", lty=1, lwd=2,col= "purple")
legend("topright", c("Dislexia", "Dif.s de Aprendizagem", "Dif. de Aprendizagem", "Fracasso Escolar", "Transt. da Aprendizagem", "Disgrafia", "Probl. de Aprendizagem", "Discalculia"), bty="n", lty=c(1),
col=c("red","darkblue", "darkred","darkgray", "purple", "orange", "pink", "darkgreen"), box.col="white",lwd=c(2), cex=0.9)
Analise da densidade das palavras # borrowed from melina.leite- # http://ecologia.ib.usp.br/bie5782/doku.php?id=bie5782:03_apostila:05-exploratoria
#grafico de densidade
plot(density (Timeline$Dislexia, bw=1.5), col="red", xlab="", ylab="Densidade Probabilistica", main = "Densidade pela frequencia")
lines(density(Timeline$DAs, bw=1.5), col= "darkblue")
lines(density(Timeline$FE, bw=1.5),col= "darkgray")
lines(density(Timeline$PA, bw=1.5),col= "pink")
lines(density(Timeline$DE, bw=1.5),col= "black")
lines(density(Timeline$DA, bw=1.5),col= "darkred")
lines(density(Timeline$Discalculia, bw=1.5),col= "green")
lines(density(Timeline$Disgrafia, bw=1.5),col= "orange")
lines(density(Timeline$TA, bw=1.5),col= "purple")
legend("topright", c("Dislexia", "Dif.s de Aprendizagem", "Dif. de Aprendizagem", "Fracasso Escolar", "Transt. da Aprendizagem", "Disgrafia", "Probl. de Aprendizagem", "Discalculia"), bty="n", lty=c(1),
col=c("red","darkblue", "darkred","darkgray", "purple", "orange", "pink", "green"), box.col="white",lwd=c(2,2), cex=0.9)
Relacao entre as palavras (1) Dificiculdades de Aprendizagem x todas as palavras (2) Dislexia x todas as palavras
Estatistica Descritiva
#valores maximos e minimos
max = apply(Timeline[3:12], MARGIN=2, FUN= max)
max_5anos = apply(Timeline_5anos[3:12], MARGIN=2, FUN= max)
min = apply(Timeline[3:12], MARGIN=2, FUN= min)
min_5anos = apply(Timeline_5anos[3:12], MARGIN=2, FUN= min)
#mediana
mediana = apply(Timeline[3:12], MARGIN=2, FUN=median)
mediana_5anos = apply(Timeline_5anos[3:12], MARGIN=2, FUN=median)
#medias
medias = apply(Timeline[3:12], MARGIN=2, FUN=mean)
medias_5anos = apply(Timeline_5anos[3:12], MARGIN=2, FUN=mean)
#desvio padrao
dp = apply(Timeline[3:12], MARGIN=2, FUN=sd)
dp_5anos = apply(Timeline_5anos[3:12], MARGIN=2, FUN=sd)
#variance
var = apply(Timeline[3:12], 2, var)
var_5anos = apply(Timeline_5anos[3:12], 2, var)
#Tabela descritiva, 2004 a 2017
TabelaDes = data.frame( min, max, mediana, medias, dp, var)
TabelaDes## min max mediana medias dp var
## DA 7 76 19 22.7707006 11.1768262 124.9214437
## DAs 9 100 29 32.0254777 15.0298630 225.8967826
## FE 3 90 12 20.1974522 18.7211109 350.4799935
## PA 2 49 7 10.1146497 7.9775070 63.6406173
## DE 2 25 6 6.7388535 3.4271662 11.7454679
## TA 0 4 1 0.9044586 0.8147946 0.6638902
## DA1 2 26 7 8.3439490 3.9300182 15.4450433
## Dislexia 13 100 25 30.5859873 15.6745934 245.6928793
## Disgrafia 0 13 2 2.3630573 1.8402650 3.3865752
## Discalculia 0 6 2 2.0891720 1.1057878 1.2227666#Tabela descrita, 2012 a 2017
TabelaDes_5anos = data.frame(min_5anos, max_5anos, mediana_5anos, medias_5anos, dp_5anos, var_5anos)
TabelaDes_5anos## min_5anos max_5anos mediana_5anos medias_5anos dp_5anos
## DA 7 25 17 16.266667 3.8658289
## DAs 9 35 23 22.450000 5.6521332
## FE 3 12 7 6.833333 2.3733454
## PA 2 10 5 5.266667 1.6142025
## DE 2 8 5 5.183333 1.2952543
## TA 0 1 1 0.650000 0.4809947
## DA1 2 9 6 5.833333 1.4976441
## Dislexia 13 35 21 21.116667 4.2469996
## Disgrafia 1 2 1 1.450000 0.5016921
## Discalculia 1 3 2 1.783333 0.5551505
## var_5anos
## DA 14.9446328
## DAs 31.9466102
## FE 5.6327684
## PA 2.6056497
## DE 1.6776836
## TA 0.2313559
## DA1 2.2429379
## Dislexia 18.0370056
## Disgrafia 0.2516949
## Discalculia 0.3081921#correlacao entre as variaveis
explicativas = Timeline[,3:12]
cor(explicativas, method = "spearman")## DA DAs FE PA DE TA
## DA 1.0000000 0.8540771 0.8209625 0.7936141 0.6392457 0.5180165
## DAs 0.8540771 1.0000000 0.8524418 0.8269852 0.6485852 0.4435616
## FE 0.8209625 0.8524418 1.0000000 0.8653836 0.5799170 0.4378294
## PA 0.7936141 0.8269852 0.8653836 1.0000000 0.6016096 0.4329864
## DE 0.6392457 0.6485852 0.5799170 0.6016096 1.0000000 0.3825523
## TA 0.5180165 0.4435616 0.4378294 0.4329864 0.3825523 1.0000000
## DA1 0.8503754 0.9935315 0.8494201 0.8172536 0.6425757 0.4314246
## Dislexia 0.7587646 0.7326280 0.8566252 0.7994667 0.4988009 0.3492648
## Disgrafia 0.7968748 0.8085197 0.7878822 0.7858998 0.5889306 0.5114184
## Discalculia 0.5809042 0.5118086 0.4850222 0.5163804 0.5044344 0.5069295
## DA1 Dislexia Disgrafia Discalculia
## DA 0.8503754 0.7587646 0.7968748 0.5809042
## DAs 0.9935315 0.7326280 0.8085197 0.5118086
## FE 0.8494201 0.8566252 0.7878822 0.4850222
## PA 0.8172536 0.7994667 0.7858998 0.5163804
## DE 0.6425757 0.4988009 0.5889306 0.5044344
## TA 0.4314246 0.3492648 0.5114184 0.5069295
## DA1 1.0000000 0.7325950 0.8125736 0.5189100
## Dislexia 0.7325950 1.0000000 0.7344780 0.4313871
## Disgrafia 0.8125736 0.7344780 1.0000000 0.5902121
## Discalculia 0.5189100 0.4313871 0.5902121 1.0000000pairs(explicativas)
Testes normalidade das variaveis
library(nortest)
#Teste de Shapiro-Francis para normalidade - n< 50
sf.test = apply(Timeline[17:25], MARGIN=2, FUN=sf.test)
sf.test## $DA_clean
##
## Shapiro-Francia normality test
##
## data: newX[, i]
## W = 0.95516, p-value = 0.0002326
##
##
## $DAs_clean
##
## Shapiro-Francia normality test
##
## data: newX[, i]
## W = 0.96458, p-value = 0.001309
##
##
## $Dislexia_clean
##
## Shapiro-Francia normality test
##
## data: newX[, i]
## W = 0.92617, p-value = 3.676e-06
##
##
## $FE_clean
##
## Shapiro-Francia normality test
##
## data: newX[, i]
## W = 0.84329, p-value = 7.899e-10
##
##
## $PA_clean
##
## Shapiro-Francia normality test
##
## data: newX[, i]
## W = 0.87275, p-value = 1.165e-08
##
##
## $DE_clean
##
## Shapiro-Francia normality test
##
## data: newX[, i]
## W = 0.95808, p-value = 0.0003847
##
##
## $Discalculia_clean
##
## Shapiro-Francia normality test
##
## data: newX[, i]
## W = 0.82941, p-value = 3.222e-10
##
##
## $Disgrafia_clean
##
## Shapiro-Francia normality test
##
## data: newX[, i]
## W = 0.82525, p-value = 1.137e-10
##
##
## $TA_clean
##
## Shapiro-Francia normality test
##
## data: newX[, i]
## W = 0.72974, p-value = 4.65e-13#Testes de Shapiro-Wilk para normalidade
shapiro.test = apply(Timeline[17:25], MARGIN=2, FUN=shapiro.test)
shapiro.test## $DA_clean
##
## Shapiro-Wilk normality test
##
## data: newX[, i]
## W = 0.95302, p-value = 7.019e-05
##
##
## $DAs_clean
##
## Shapiro-Wilk normality test
##
## data: newX[, i]
## W = 0.96262, p-value = 0.0005275
##
##
## $Dislexia_clean
##
## Shapiro-Wilk normality test
##
## data: newX[, i]
## W = 0.92287, p-value = 4.782e-07
##
##
## $FE_clean
##
## Shapiro-Wilk normality test
##
## data: newX[, i]
## W = 0.8404, p-value = 2.398e-11
##
##
## $PA_clean
##
## Shapiro-Wilk normality test
##
## data: newX[, i]
## W = 0.87041, p-value = 6.218e-10
##
##
## $DE_clean
##
## Shapiro-Wilk normality test
##
## data: newX[, i]
## W = 0.95616, p-value = 0.0001306
##
##
## $Discalculia_clean
##
## Shapiro-Wilk normality test
##
## data: newX[, i]
## W = 0.82723, p-value = 8.761e-12
##
##
## $Disgrafia_clean
##
## Shapiro-Wilk normality test
##
## data: newX[, i]
## W = 0.82471, p-value = 2.961e-12
##
##
## $TA_clean
##
## Shapiro-Wilk normality test
##
## data: newX[, i]
## W = 0.72669, p-value = 3.211e-15#Teste de normalidade Anderson-Darling
ad.test = apply(Timeline[17:25], MARGIN=2, FUN=ad.test)
ad.test## $DA_clean
##
## Anderson-Darling normality test
##
## data: newX[, i]
## A = 2.2692, p-value = 8.88e-06
##
##
## $DAs_clean
##
## Anderson-Darling normality test
##
## data: newX[, i]
## A = 1.5723, p-value = 0.0004605
##
##
## $Dislexia_clean
##
## Anderson-Darling normality test
##
## data: newX[, i]
## A = 3.6684, p-value = 3.375e-09
##
##
## $FE_clean
##
## Anderson-Darling normality test
##
## data: newX[, i]
## A = 8.4665, p-value < 2.2e-16
##
##
## $PA_clean
##
## Anderson-Darling normality test
##
## data: newX[, i]
## A = 6.3436, p-value = 1.2e-15
##
##
## $DE_clean
##
## Anderson-Darling normality test
##
## data: newX[, i]
## A = 2.3517, p-value = 5.571e-06
##
##
## $Discalculia_clean
##
## Anderson-Darling normality test
##
## data: newX[, i]
## A = 11.851, p-value < 2.2e-16
##
##
## $Disgrafia_clean
##
## Anderson-Darling normality test
##
## data: newX[, i]
## A = 9.308, p-value < 2.2e-16
##
##
## $TA_clean
##
## Anderson-Darling normality test
##
## data: newX[, i]
## A = 20.155, p-value < 2.2e-16#Teste de Cramer-von Mises para normalidade
cvm.test = apply(Timeline[17:25], MARGIN=2, FUN=cvm.test)## Warning in FUN(newX[, i], ...): p-value is smaller than 7.37e-10, cannot be
## computed more accurately
## Warning in FUN(newX[, i], ...): p-value is smaller than 7.37e-10, cannot be
## computed more accurately
## Warning in FUN(newX[, i], ...): p-value is smaller than 7.37e-10, cannot be
## computed more accurately
## Warning in FUN(newX[, i], ...): p-value is smaller than 7.37e-10, cannot be
## computed more accurately
## Warning in FUN(newX[, i], ...): p-value is smaller than 7.37e-10, cannot be
## computed more accuratelycvm.test## $DA_clean
##
## Cramer-von Mises normality test
##
## data: newX[, i]
## W = 0.43175, p-value = 1.219e-05
##
##
## $DAs_clean
##
## Cramer-von Mises normality test
##
## data: newX[, i]
## W = 0.27493, p-value = 0.0006369
##
##
## $Dislexia_clean
##
## Cramer-von Mises normality test
##
## data: newX[, i]
## W = 0.64513, p-value = 1.552e-07
##
##
## $FE_clean
##
## Cramer-von Mises normality test
##
## data: newX[, i]
## W = 1.5153, p-value = 7.37e-10
##
##
## $PA_clean
##
## Cramer-von Mises normality test
##
## data: newX[, i]
## W = 1.1283, p-value = 7.37e-10
##
##
## $DE_clean
##
## Cramer-von Mises normality test
##
## data: newX[, i]
## W = 0.43382, p-value = 1.162e-05
##
##
## $Discalculia_clean
##
## Cramer-von Mises normality test
##
## data: newX[, i]
## W = 2.1864, p-value = 7.37e-10
##
##
## $Disgrafia_clean
##
## Cramer-von Mises normality test
##
## data: newX[, i]
## W = 1.6243, p-value = 7.37e-10
##
##
## $TA_clean
##
## Cramer-von Mises normality test
##
## data: newX[, i]
## W = 3.7942, p-value = 7.37e-10#Teste de Lilliefors (correção do KS) para normalidade do ajuste, n < 50
lillie.test = apply(Timeline[17:25], MARGIN=2, FUN=lillie.test)
lillie.test## $DA_clean
##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: newX[, i]
## D = 0.14452, p-value = 5.78e-08
##
##
## $DAs_clean
##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: newX[, i]
## D = 0.10969, p-value = 0.000191
##
##
## $Dislexia_clean
##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: newX[, i]
## D = 0.13847, p-value = 3.542e-07
##
##
## $FE_clean
##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: newX[, i]
## D = 0.2038, p-value < 2.2e-16
##
##
## $PA_clean
##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: newX[, i]
## D = 0.20385, p-value = 3.174e-16
##
##
## $DE_clean
##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: newX[, i]
## D = 0.14544, p-value = 4.518e-08
##
##
## $Discalculia_clean
##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: newX[, i]
## D = 0.27422, p-value < 2.2e-16
##
##
## $Disgrafia_clean
##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: newX[, i]
## D = 0.2652, p-value < 2.2e-16
##
##
## $TA_clean
##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: newX[, i]
## D = 0.3607, p-value < 2.2e-16#Teste Qui-quadrado de Pearson para normalidade considerando o número de classes
pearson.test = apply(Timeline[17:25], MARGIN=2, FUN=pearson.test)
pearson.test## $DA_clean
##
## Pearson chi-square normality test
##
## data: newX[, i]
## P = 26.367, p-value = 0.009519
##
##
## $DAs_clean
##
## Pearson chi-square normality test
##
## data: newX[, i]
## P = 29.685, p-value = 0.003114
##
##
## $Dislexia_clean
##
## Pearson chi-square normality test
##
## data: newX[, i]
## P = 46.483, p-value = 5.728e-06
##
##
## $FE_clean
##
## Pearson chi-square normality test
##
## data: newX[, i]
## P = 137.59, p-value < 2.2e-16
##
##
## $PA_clean
##
## Pearson chi-square normality test
##
## data: newX[, i]
## P = 76.483, p-value = 1.925e-11
##
##
## $DE_clean
##
## Pearson chi-square normality test
##
## data: newX[, i]
## P = 172.49, p-value < 2.2e-16
##
##
## $Discalculia_clean
##
## Pearson chi-square normality test
##
## data: newX[, i]
## P = 693.24, p-value < 2.2e-16
##
##
## $Disgrafia_clean
##
## Pearson chi-square normality test
##
## data: newX[, i]
## P = 504.94, p-value < 2.2e-16
##
##
## $TA_clean
##
## Pearson chi-square normality test
##
## data: newX[, i]
## P = 940.86, p-value < 2.2e-16CONCLUSAO: a distribuicao dos dados nao e normal e, portanto, deve se adotar metodos nao-parametricos.
Graficos de diagnostico da variavel - Dif.s de Aprendizagem
#Gráfico de diagnóstico e Histograma com ajuste não paramétrico de densidade probabilística mais curva normal, com e sem outliers
par(mfrow=c(2,2))
boxplot(Timeline$DAs, notch = T, main = "Dif.s de Aprendizagem", ylab = "valores observados")
boxplot(Timeline$DAs_clean, notch = T, main = "removidos outliers", ylab = "valores observados")
hist(Timeline$DAs, prob=T, main = "Densid. probabilística e curva normal", xlab = "valores observados", ylab = "Densidade probabilística")
lines( density(Timeline$DAs),col="darkblue", lwd=c(2))
curve(expr = dnorm(x,mean=mean(Timeline$DAs),sd=sd(Timeline$DAs)),add=T, col="darkred", lwd=c(2))
legend("topright", c("Dens. prob.", "curva normal"), bty="n", lty=c(1),
col=c("darkblue","darkred"), box.col="white",lwd=c(2), cex=0.9)
hist(Timeline$DAs_clean, prob=T, main = "removidos outliers", xlab = "valores observados", ylab = "Densidade probabilística")
lines( density(Timeline$DAs),col="darkblue", lwd=c(2) )
curve(expr = dnorm(x,mean=mean(Timeline$DAs),sd=sd(Timeline$DAs)),add=T, col="darkred", lwd=c(2))
legend("topright", c("Dens. prob.", "curva normal"), bty="n", lty=c(1),
col=c("darkblue","darkred"), box.col="white",lwd=c(2), cex=0.9)
par(mfrow=c(1,1))
#testes de normalidade
Teste_norm_DAs = c(sf.test$DAs_clean$method, ad.test$DAs_clean$method, cvm.test$DAs_clean$method, lillie.test$DAs_clean$method, pearson.test$DAs_clean$method, shapiro.test$DAs_clean$method)
pvalor_norm_DAs = c(sf.test$DAs_clean$p.value, ad.test$DAs_clean$p.value, cvm.test$DAs_clean$p.value, lillie.test$DAs_clean$p.value, pearson.test$DAs_clean$p.value, shapiro.test$DAs_clean$p.value)
Tab_norm_DAs = data.frame(Teste_norm_DAs, pvalor_norm_DAs)
Tab_norm_DAs## Teste_norm_DAs pvalor_norm_DAs
## 1 Shapiro-Francia normality test 0.0013094786
## 2 Anderson-Darling normality test 0.0004604626
## 3 Cramer-von Mises normality test 0.0006369276
## 4 Lilliefors (Kolmogorov-Smirnov) normality test 0.0001909866
## 5 Pearson chi-square normality test 0.0031139409
## 6 Shapiro-Wilk normality test 0.0005274689Correlacao entre a variavel dependente e as demais, metodo Spearman (não parametrico), retirados outliers
cor.das.da = cor.test(Timeline$DAs_clean, Timeline$DA_clean, method = "spearman") ## DificuldadeS de Aprendizagem## Warning in cor.test.default(Timeline$DAs_clean, Timeline$DA_clean, method =
## "spearman"): Cannot compute exact p-value with tiescor.das.fe = cor.test(Timeline$DAs_clean, Timeline$FE_clean, method = "spearman") ## Fracasso Escolar## Warning in cor.test.default(Timeline$DAs_clean, Timeline$FE_clean, method =
## "spearman"): Cannot compute exact p-value with tiescor.das.pa = cor.test(Timeline$DAs_clean, Timeline$PA_clean, method = "spearman") ## Problemas de Aprendizagem## Warning in cor.test.default(Timeline$DAs_clean, Timeline$PA_clean, method =
## "spearman"): Cannot compute exact p-value with tiescor.das.de = cor.test(Timeline$DAs_clean, Timeline$DE_clean, method = "spearman") ## Desempenho escolar## Warning in cor.test.default(Timeline$DAs_clean, Timeline$DE_clean, method =
## "spearman"): Cannot compute exact p-value with tiescor.das.dis1 = cor.test(Timeline$DAs_clean, Timeline$Dislexia_clean, method = "spearman") #Dislexia## Warning in cor.test.default(Timeline$DAs_clean, Timeline$Dislexia_clean, :
## Cannot compute exact p-value with tiescor.das.dis2 = cor.test(Timeline$DAs_clean, Timeline$Discalculia_clean, method = "spearman") #Discalculia## Warning in cor.test.default(Timeline$DAs_clean, Timeline
## $Discalculia_clean, : Cannot compute exact p-value with tiescor.das.dis3 = cor.test(Timeline$DAs_clean, Timeline$Disgrafia_clean, method = "spearman") #Digrafia## Warning in cor.test.default(Timeline$DAs_clean, Timeline$Disgrafia_clean, :
## Cannot compute exact p-value with tiescor.das.ta = cor.test(Timeline$DAs_clean, Timeline$TA_clean, method = "spearman") ## Transtorno de Aprendizagem## Warning in cor.test.default(Timeline$DAs_clean, Timeline$TA_clean, method =
## "spearman"): Cannot compute exact p-value with tiesNome = c("Dif.s de aprend. com Dif. de Aprend.", " com Fracasso Escolar", " com Problema de Aprend.", " com Desempenho Escolar", " com Dislexia", " com Discalculia", " com Disgrafia", " com Transt. de Aprend.")
pvalor_cor = c(cor.das.da$p.value, cor.das.fe$p.value, cor.das.pa$p.value, cor.das.de$p.value, cor.das.dis1$p.value, cor.das.dis2$p.value, cor.das.dis3$p.value, cor.das.ta$p.value)
valor_cor = c(cor.das.da$estimate, cor.das.fe$estimate, cor.das.pa$estimate, cor.das.de$estimate, cor.das.dis1$estimate, cor.das.dis2$estimate, cor.das.dis3$estimate, cor.das.ta$estimate)
tab_cor_DAs = data.frame(Nome, valor_cor, pvalor_cor)
tab_cor_DAs## Nome valor_cor pvalor_cor
## 1 Dif.s de aprend. com Dif. de Aprend. 0.8284254 4.629895e-37
## 2 com Fracasso Escolar 0.8092017 2.240052e-34
## 3 com Problema de Aprend. 0.7745652 3.061280e-29
## 4 com Desempenho Escolar 0.5891858 2.355150e-14
## 5 com Dislexia 0.7775874 3.442925e-29
## 6 com Discalculia 0.3835854 3.416791e-06
## 7 com Disgrafia 0.7638980 8.502890e-29
## 8 com Transt. de Aprend. 0.2659142 1.381377e-03Grafico de dispersão com linha de regressão, método loess (não parametrico) - Dif.s de Aprendizagem, retirados outliers
library("ggplot2")
library("devtools")
ggplot(Timeline, aes(y=DAs_clean, x=DA_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 15 rows containing non-finite values (stat_smooth).## Warning: Removed 15 rows containing missing values (geom_point).
qplot(DA_clean, DAs_clean, data = Timeline, geom = c("point", "smooth"), main = "Dif.s x Dif. de Aprendizagem")## `geom_smooth()` using method = 'loess'## Warning: Removed 15 rows containing non-finite values (stat_smooth).
## Warning: Removed 15 rows containing missing values (geom_point).
ggplot(Timeline, aes(y=DAs_clean, x=FE_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 14 rows containing non-finite values (stat_smooth).## Warning: Removed 14 rows containing missing values (geom_point).
qplot(FE_clean, DAs_clean, data = Timeline, geom = c("point", "smooth"), main = "Dif.s x Fracasso Escolar")## `geom_smooth()` using method = 'loess'## Warning: Removed 14 rows containing non-finite values (stat_smooth).
## Warning: Removed 14 rows containing missing values (geom_point).
ggplot(Timeline, aes(y=DAs_clean, x=PA_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 17 rows containing non-finite values (stat_smooth).## Warning: Removed 17 rows containing missing values (geom_point).
qplot(PA_clean, DAs_clean, data = Timeline, geom = c("point", "smooth"), main = "Dif.s x Problemas de Aprendizagem")## `geom_smooth()` using method = 'loess'## Warning: Removed 17 rows containing non-finite values (stat_smooth).
## Warning: Removed 17 rows containing missing values (geom_point).
ggplot(Timeline, aes(y=DAs_clean, x=DE_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 18 rows containing non-finite values (stat_smooth).## Warning: Removed 18 rows containing missing values (geom_point).
qplot(DE_clean, DAs_clean, data = Timeline, geom = c("point", "smooth"), main = "Dif.s x Desempenho Escolar")## `geom_smooth()` using method = 'loess'## Warning: Removed 18 rows containing non-finite values (stat_smooth).
## Warning: Removed 18 rows containing missing values (geom_point).
ggplot(Timeline, aes(y=DAs_clean, x=Dislexia_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 19 rows containing non-finite values (stat_smooth).## Warning: Removed 19 rows containing missing values (geom_point).
qplot(Dislexia_clean, DAs_clean, data = Timeline, geom = c("point", "smooth"), main = "Dif.s x Dislexia")## `geom_smooth()` using method = 'loess'## Warning: Removed 19 rows containing non-finite values (stat_smooth).
## Warning: Removed 19 rows containing missing values (geom_point).
ggplot(Timeline, aes(y=DAs_clean, x=Discalculia_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 19 rows containing non-finite values (stat_smooth).## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : pseudoinverse used at -0.015## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : neighborhood radius 2.015## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : reciprocal condition number 3.0341e-016## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : There are other near singularities as well. 4.0602## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : pseudoinverse used
## at -0.015## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : neighborhood radius
## 2.015## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : reciprocal
## condition number 3.0341e-016## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : There are other
## near singularities as well. 4.0602## Warning: Removed 19 rows containing missing values (geom_point).
qplot(Discalculia_clean, DAs_clean, data = Timeline, geom = c("point", "smooth"), main = "Dif.s x Discalculia")## `geom_smooth()` using method = 'loess'## Warning: Removed 19 rows containing non-finite values (stat_smooth).## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : pseudoinverse used at -0.015## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : neighborhood radius 2.015## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : reciprocal condition number 3.0341e-016## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : There are other near singularities as well. 4.0602## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : pseudoinverse used
## at -0.015## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : neighborhood radius
## 2.015## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : reciprocal
## condition number 3.0341e-016## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : There are other
## near singularities as well. 4.0602## Warning: Removed 19 rows containing missing values (geom_point).
ggplot(Timeline, aes(y=DAs_clean, x=Disgrafia_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 13 rows containing non-finite values (stat_smooth).## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : pseudoinverse used at -0.03## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : neighborhood radius 2.03## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : reciprocal condition number 1.2507e-016## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : There are other near singularities as well. 1## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : pseudoinverse used
## at -0.03## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : neighborhood radius
## 2.03## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : reciprocal
## condition number 1.2507e-016## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : There are other
## near singularities as well. 1## Warning: Removed 13 rows containing missing values (geom_point).
qplot(Disgrafia_clean, DAs_clean, data = Timeline, geom = c("point", "smooth"), main = "Dif.s x Disgrafia")## `geom_smooth()` using method = 'loess'## Warning: Removed 13 rows containing non-finite values (stat_smooth).## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : pseudoinverse used at -0.03## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : neighborhood radius 2.03## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : reciprocal condition number 1.2507e-016## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : There are other near singularities as well. 1## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : pseudoinverse used
## at -0.03## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : neighborhood radius
## 2.03## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : reciprocal
## condition number 1.2507e-016## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : There are other
## near singularities as well. 1## Warning: Removed 13 rows containing missing values (geom_point).
ggplot(Timeline, aes(y=DAs_clean, x=TA_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 15 rows containing non-finite values (stat_smooth).## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : pseudoinverse used at -0.01## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : neighborhood radius 1.01## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : reciprocal condition number 0## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : There are other near singularities as well. 4.0401## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : pseudoinverse used
## at -0.01## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : neighborhood radius
## 1.01## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : reciprocal
## condition number 0## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : There are other
## near singularities as well. 4.0401## Warning: Removed 15 rows containing missing values (geom_point).
qplot(TA_clean, DAs_clean, data = Timeline, geom = c("point", "smooth"), main = "Dif.s x Transtorno de Aprendizagem")## `geom_smooth()` using method = 'loess'## Warning: Removed 15 rows containing non-finite values (stat_smooth).## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : pseudoinverse used at -0.01## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : neighborhood radius 1.01## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : reciprocal condition number 0## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : There are other near singularities as well. 4.0401## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : pseudoinverse used
## at -0.01## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : neighborhood radius
## 1.01## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : reciprocal
## condition number 0## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : There are other
## near singularities as well. 4.0401## Warning: Removed 15 rows containing missing values (geom_point).
Modelo de Regressao linear multipla (prescinde de testes de normalidade, homogeneidade e dependência dos resíduos) - Dificuldades de Aprendizagem
#Regressão Linear Múltipla
## observar o R2 - coeficiente de determinacao, quanto o modelo consegue explicar os valores observados valores entre 0 e 1, convertido em porcentagem. E a distribuição F - relação entre a variável dependente e independente, p < 0,05 ao menos uma variável explicativa contribui significativamente para o modelo
RL_DAs1 = lm( DAs ~ DA + FE + PA + DE + Dislexia + Discalculia + Disgrafia + TA, data=Timeline)
summary( RL_DAs1 )##
## Call:
## lm(formula = DAs ~ DA + FE + PA + DE + Dislexia + Discalculia +
## Disgrafia + TA, data = Timeline)
##
## Residuals:
## Min 1Q Median 3Q Max
## -27.7796 -4.4051 -0.2952 3.2602 23.5186
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 15.73371 2.16432 7.270 1.93e-11 ***
## DA 0.12492 0.11400 1.096 0.2750
## FE 0.60800 0.08085 7.520 4.89e-12 ***
## PA -0.14275 0.16756 -0.852 0.3956
## DE 0.48093 0.25784 1.865 0.0641 .
## Dislexia -0.11573 0.05563 -2.080 0.0392 *
## Discalculia 0.30267 0.69808 0.434 0.6652
## Disgrafia 0.54605 0.55880 0.977 0.3301
## TA 1.09128 1.11393 0.980 0.3288
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 7.531 on 148 degrees of freedom
## Multiple R-squared: 0.7618, Adjusted R-squared: 0.7489
## F-statistic: 59.16 on 8 and 148 DF, p-value: < 2.2e-16#Testes de normalidade de ajuste do residuo
lillie.test(RL_DAs1$residuals)##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: RL_DAs1$residuals
## D = 0.091273, p-value = 0.002761#Regressão Linear, removidos os outliers
RL_DAs = lm( DAs_clean ~ DA_clean + FE_clean + PA_clean + DE_clean + Dislexia_clean + Discalculia_clean + Disgrafia_clean + TA_clean, data=Timeline)
summary( RL_DAs )##
## Call:
## lm(formula = DAs_clean ~ DA_clean + FE_clean + PA_clean + DE_clean +
## Dislexia_clean + Discalculia_clean + Disgrafia_clean + TA_clean,
## data = Timeline)
##
## Residuals:
## Min 1Q Median 3Q Max
## -15.5469 -3.0775 0.0795 2.7979 15.5779
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.93634 2.01764 2.447 0.01594 *
## DA_clean 0.58738 0.12460 4.714 6.87e-06 ***
## FE_clean 0.07087 0.08772 0.808 0.42083
## PA_clean 0.36834 0.19833 1.857 0.06585 .
## DE_clean 0.27743 0.27266 1.017 0.31105
## Dislexia_clean 0.06703 0.10516 0.637 0.52512
## Discalculia_clean 0.06120 0.74686 0.082 0.93484
## Disgrafia_clean 2.70418 0.81104 3.334 0.00115 **
## TA_clean -1.10400 0.95187 -1.160 0.24852
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.707 on 115 degrees of freedom
## (33 observations deleted due to missingness)
## Multiple R-squared: 0.7557, Adjusted R-squared: 0.7387
## F-statistic: 44.46 on 8 and 115 DF, p-value: < 2.2e-16#Testes de normalidade de ajuste do residuo
lillie.test(RL_DAs$residuals)##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: RL_DAs$residuals
## D = 0.062922, p-value = 0.2666par(mfrow=c(2,2))
plot(RL_DAs)
library(car)
par(mfrow=c(1,2))
residualPlot(RL_DAs1, main = "Gráfico de Residuos")
residualPlot(RL_DAs, main = "Gráfico de Residuos, sem outliers")
#Ajuste de modelo pelo método Stepwise Backward
#retirando dislcaculia, p = 0,93484
RL_DAs_aj1 = lm( DAs_clean ~ DA_clean + FE_clean + PA_clean + DE_clean + Dislexia_clean + Disgrafia_clean + TA_clean, data=Timeline)
summary(RL_DAs_aj1)##
## Call:
## lm(formula = DAs_clean ~ DA_clean + FE_clean + PA_clean + DE_clean +
## Dislexia_clean + Disgrafia_clean + TA_clean, data = Timeline)
##
## Residuals:
## Min 1Q Median 3Q Max
## -16.6171 -3.0958 -0.1424 2.9445 15.1224
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.33080 1.92615 2.768 0.00655 **
## DA_clean 0.57281 0.12157 4.712 6.72e-06 ***
## FE_clean 0.13517 0.07651 1.767 0.07984 .
## PA_clean 0.28327 0.19356 1.463 0.14597
## DE_clean 0.32855 0.26187 1.255 0.21208
## Dislexia_clean 0.04887 0.10183 0.480 0.63220
## Disgrafia_clean 2.52339 0.77280 3.265 0.00143 **
## TA_clean -0.63608 0.91365 -0.696 0.48767
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.708 on 119 degrees of freedom
## (30 observations deleted due to missingness)
## Multiple R-squared: 0.7592, Adjusted R-squared: 0.745
## F-statistic: 53.59 on 7 and 119 DF, p-value: < 2.2e-16#retirando Dislexia, p = 0,63220
RL_DAs_aj2 = lm( DAs_clean ~ DA_clean + FE_clean + PA_clean + DE_clean + Disgrafia_clean + TA_clean, data=Timeline)
summary(RL_DAs_aj2)##
## Call:
## lm(formula = DAs_clean ~ DA_clean + FE_clean + PA_clean + DE_clean +
## Disgrafia_clean + TA_clean, data = Timeline)
##
## Residuals:
## Min 1Q Median 3Q Max
## -15.3148 -3.2140 -0.2363 3.0104 16.4652
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.18463 1.58960 3.262 0.001425 **
## DA_clean 0.66417 0.11270 5.893 3.25e-08 ***
## FE_clean 0.11066 0.06831 1.620 0.107726
## PA_clean 0.20945 0.18241 1.148 0.253045
## DE_clean 0.33812 0.25668 1.317 0.190124
## Disgrafia_clean 2.66941 0.76217 3.502 0.000639 ***
## TA_clean -0.65964 0.89343 -0.738 0.461688
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.774 on 126 degrees of freedom
## (24 observations deleted due to missingness)
## Multiple R-squared: 0.7476, Adjusted R-squared: 0.7356
## F-statistic: 62.21 on 6 and 126 DF, p-value: < 2.2e-16#retirando Transtorno de Aprendizagem, p = 0,461688
RL_DAs_aj3 = lm(DAs_clean ~ DA_clean + FE_clean + PA_clean + DE_clean + Disgrafia_clean, data=Timeline)
summary(RL_DAs_aj3)##
## Call:
## lm(formula = DAs_clean ~ DA_clean + FE_clean + PA_clean + DE_clean +
## Disgrafia_clean, data = Timeline)
##
## Residuals:
## Min 1Q Median 3Q Max
## -15.5361 -3.1722 -0.0087 2.9604 16.3183
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.04710 1.57582 3.203 0.001720 **
## DA_clean 0.65120 0.11113 5.860 3.75e-08 ***
## FE_clean 0.11182 0.06817 1.640 0.103400
## PA_clean 0.21728 0.18178 1.195 0.234193
## DE_clean 0.33401 0.25615 1.304 0.194616
## Disgrafia_clean 2.61139 0.75675 3.451 0.000759 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.766 on 127 degrees of freedom
## (24 observations deleted due to missingness)
## Multiple R-squared: 0.7465, Adjusted R-squared: 0.7366
## F-statistic: 74.81 on 5 and 127 DF, p-value: < 2.2e-16#Retirando Problema de Aprendizagem, p = 0,234193
RL_DAs_aj4 = lm( DAs_clean ~ DA_clean + FE_clean + DE_clean + Disgrafia_clean, data=Timeline)
summary(RL_DAs_aj4)##
## Call:
## lm(formula = DAs_clean ~ DA_clean + FE_clean + DE_clean + Disgrafia_clean,
## data = Timeline)
##
## Residuals:
## Min 1Q Median 3Q Max
## -16.8319 -2.8564 -0.3094 3.0077 17.4098
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.74149 1.58509 4.253 3.96e-05 ***
## DA_clean 0.65757 0.11435 5.750 5.89e-08 ***
## FE_clean 0.16225 0.06418 2.528 0.01265 *
## DE_clean 0.37681 0.25370 1.485 0.13986
## Disgrafia_clean 1.89478 0.70271 2.696 0.00792 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.996 on 132 degrees of freedom
## (20 observations deleted due to missingness)
## Multiple R-squared: 0.7318, Adjusted R-squared: 0.7237
## F-statistic: 90.04 on 4 and 132 DF, p-value: < 2.2e-16#Retirando Desempenho Escolar, p = 0,13986
RL_DAs_aj5 = lm( DAs_clean ~ DA_clean + FE_clean + Disgrafia_clean, data=Timeline)
summary(RL_DAs_aj5)##
## Call:
## lm(formula = DAs_clean ~ DA_clean + FE_clean + Disgrafia_clean,
## data = Timeline)
##
## Residuals:
## Min 1Q Median 3Q Max
## -16.5990 -3.1871 -0.3789 3.1076 16.8447
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 7.20862 1.42196 5.069 1.27e-06 ***
## DA_clean 0.76572 0.10467 7.316 1.95e-11 ***
## FE_clean 0.15344 0.06429 2.387 0.0184 *
## Disgrafia_clean 1.75992 0.70563 2.494 0.0138 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.08 on 137 degrees of freedom
## (16 observations deleted due to missingness)
## Multiple R-squared: 0.7448, Adjusted R-squared: 0.7392
## F-statistic: 133.3 on 3 and 137 DF, p-value: < 2.2e-16par(mfrow=c(2,2))
plot(RL_DAs_aj5)
#Teste de normalidade de ajuste do residuo
shapiro.test(RL_DAs_aj5$residuals)##
## Shapiro-Wilk normality test
##
## data: RL_DAs_aj5$residuals
## W = 0.97898, p-value = 0.02861sf.test(RL_DAs_aj5$residuals)##
## Shapiro-Francia normality test
##
## data: RL_DAs_aj5$residuals
## W = 0.97438, p-value = 0.01106ad.test(RL_DAs_aj5$residuals)##
## Anderson-Darling normality test
##
## data: RL_DAs_aj5$residuals
## A = 0.57491, p-value = 0.1331cvm.test(RL_DAs_aj5$residuals)##
## Cramer-von Mises normality test
##
## data: RL_DAs_aj5$residuals
## W = 0.070668, p-value = 0.2721lillie.test(RL_DAs_aj5$residuals)##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: RL_DAs_aj5$residuals
## D = 0.04998, p-value = 0.5265pearson.test(RL_DAs_aj5$residuals)##
## Pearson chi-square normality test
##
## data: RL_DAs_aj5$residuals
## P = 12.723, p-value = 0.3895#Teste de homoscedasticidade dos resíduos
residualPlot(RL_DAs_aj5)
library(lmtest)## Warning: package 'lmtest' was built under R version 3.3.3## Loading required package: zoo##
## Attaching package: 'zoo'## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numeric#Teste Goldfeld-Quandt test against heteroskedasticity (homoscedasticidade dos resíduos) n > 30
gqtest(RL_DAs_aj5)##
## Goldfeld-Quandt test
##
## data: RL_DAs_aj5
## GQ = 0.19071, df1 = 67, df2 = 66, p-value = 1
## alternative hypothesis: variance increases from segment 1 to 2#Teste de Breusch-Pagan: hipótese nula de que as variâncias dos erros são iguais (homoscedasticidade)
bptest(RL_DAs_aj5)##
## studentized Breusch-Pagan test
##
## data: RL_DAs_aj5
## BP = 23.965, df = 3, p-value = 2.54e-05#Teste de independencia - Teste de Durbin Watson
plot(RL_DAs_aj5$residuals)
dwtest(RL_DAs_aj5)##
## Durbin-Watson test
##
## data: RL_DAs_aj5
## DW = 1.5931, p-value = 0.005695
## alternative hypothesis: true autocorrelation is greater than 0#Retirando Disgrafia, p = 0.0138
RL_DAs_aj6 = lm( DAs_clean ~ DA_clean + FE_clean, data=Timeline)
summary(RL_DAs_aj6)##
## Call:
## lm(formula = DAs_clean ~ DA_clean + FE_clean, data = Timeline)
##
## Residuals:
## Min 1Q Median 3Q Max
## -13.8988 -3.5731 -0.4507 2.9881 18.6894
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 7.56890 1.44111 5.252 5.56e-07 ***
## DA_clean 0.86677 0.09832 8.816 4.58e-15 ***
## FE_clean 0.21938 0.05970 3.675 0.00034 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.175 on 138 degrees of freedom
## (16 observations deleted due to missingness)
## Multiple R-squared: 0.7332, Adjusted R-squared: 0.7293
## F-statistic: 189.6 on 2 and 138 DF, p-value: < 2.2e-16par(mfrow=c(2,2))
plot(RL_DAs_aj6)
#Teste de normalidade de ajuste do residuo
shapiro.test(RL_DAs_aj6$residuals)##
## Shapiro-Wilk normality test
##
## data: RL_DAs_aj6$residuals
## W = 0.98169, p-value = 0.05599sf.test(RL_DAs_aj6$residuals)##
## Shapiro-Francia normality test
##
## data: RL_DAs_aj6$residuals
## W = 0.97817, p-value = 0.0247ad.test(RL_DAs_aj6$residuals)##
## Anderson-Darling normality test
##
## data: RL_DAs_aj6$residuals
## A = 0.42498, p-value = 0.3127cvm.test(RL_DAs_aj6$residuals)##
## Cramer-von Mises normality test
##
## data: RL_DAs_aj6$residuals
## W = 0.051552, p-value = 0.4864lillie.test(RL_DAs_aj6$residuals)##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: RL_DAs_aj6$residuals
## D = 0.050816, p-value = 0.4994pearson.test(RL_DAs_aj6$residuals)##
## Pearson chi-square normality test
##
## data: RL_DAs_aj6$residuals
## P = 12.298, p-value = 0.4221#Teste de homoscedasticidade dos resíduos
residualPlot(RL_DAs_aj6)
library(lmtest)
#Teste Goldfeld-Quandt - test against heteroskedasticity (homoscedasticidade, ou variância constante dos erros, dos resíduos) n > 30
## p > 0,05 não rejeitamos a hipótese de homoscedasticidade dos erros
gqtest(RL_DAs_aj6)##
## Goldfeld-Quandt test
##
## data: RL_DAs_aj6
## GQ = 0.17646, df1 = 68, df2 = 67, p-value = 1
## alternative hypothesis: variance increases from segment 1 to 2#Teste de Breusch-Pagan: hipótese nula de que as variâncias dos erros são iguais (homoscedasticidade)
##p > 0,05 não rejeitamos a hipótese de homoscedasticidade dos erros
bptest(RL_DAs_aj6)##
## studentized Breusch-Pagan test
##
## data: RL_DAs_aj6
## BP = 15.706, df = 2, p-value = 0.0003885#Teste de independencia - Teste de Durbin Watson
## p > 0,05, não rejeita Hipotese nula, os resíduos são independentes.
plot(RL_DAs_aj5$residuals)
dwtest(RL_DAs_aj6)##
## Durbin-Watson test
##
## data: RL_DAs_aj6
## DW = 1.5928, p-value = 0.005752
## alternative hypothesis: true autocorrelation is greater than 0
Graficos de diagnostico da variavel - Dislexia
#Gráfico de diagnóstico e Histograma com ajuste não paramétrico de densidade probabilística mais curva normal, com e sem outliers
par(mfrow=c(2,2))
boxplot(Timeline$Dislexia, notch = T, main = "Dislexia", ylab = "valores observados")
boxplot(Timeline$Dislexia_clean, notch = T, main = "removidos outliers", ylab = "valores observados")
hist(Timeline$Dislexia, prob=T, main = "Densid. probabilística e curva normal", xlab = "valores observados", ylab = "Densidade probabilística")
lines( density(Timeline$Dislexia),col="darkblue", lwd=c(2))
curve(expr = dnorm(x,mean=mean(Timeline$Dislexia),sd=sd(Timeline$Dislexia)),add=T, col="darkred", lwd=c(2))
legend("topright", c("Dens. prob.", "curva normal"), bty="n", lty=c(1),
col=c("darkblue","darkred"), box.col="white",lwd=c(2), cex=0.9)
hist(Timeline$Dislexia_clean, prob=T, main = "removidos outliers", xlab = "valores observados", ylab = "Densidade probabilística")
lines( density(Timeline$Dislexia),col="darkblue", lwd=c(2) )
curve(expr = dnorm(x,mean=mean(Timeline$Dislexia),sd=sd(Timeline$Dislexia)),add=T, col="darkred", lwd=c(2))
legend("topright", c("Dens. prob.", "curva normal"), bty="n", lty=c(1),
col=c("darkblue","darkred"), box.col="white",lwd=c(2), cex=0.9)
par(mfrow=c(1,1))
#testes de normalidade
Teste_norm_Dis = c(sf.test$Dislexia_clean$method, ad.test$Dislexia_clean$method, cvm.test$Dislexia_clean$method, lillie.test$Dislexia_clean$method, pearson.test$Dislexia_clean$method)
pvalor_norm_Dis = c(sf.test$Dislexia_clean$p.value, ad.test$Dislexia_clean$p.value, cvm.test$Dislexia_clean$p.value, lillie.test$Dislexia_clean$p.value, pearson.test$Dislexia_clean$p.value)
Tab_norm_Dis = data.frame(Teste_norm_Dis, pvalor_norm_Dis)
Tab_norm_Dis## Teste_norm_Dis pvalor_norm_Dis
## 1 Shapiro-Francia normality test 3.675661e-06
## 2 Anderson-Darling normality test 3.374538e-09
## 3 Cramer-von Mises normality test 1.552354e-07
## 4 Lilliefors (Kolmogorov-Smirnov) normality test 3.542177e-07
## 5 Pearson chi-square normality test 5.727799e-06Correlacao entre a variavel dependente e as demais, metodo Spearman (não parametrico), retirados outliers
cor.dis.das = cor.test(Timeline$Dislexia_clean, Timeline$DAs_clean, method = "spearman") ## DificuldadeS de Aprendizagem## Warning in cor.test.default(Timeline$Dislexia_clean, Timeline$DAs_clean, :
## Cannot compute exact p-value with tiescor.dis.fe = cor.test(Timeline$Dislexia_clean, Timeline$FE_clean, method = "spearman") ## Fracasso Escolar## Warning in cor.test.default(Timeline$Dislexia_clean, Timeline$FE_clean, :
## Cannot compute exact p-value with tiescor.dis.pa = cor.test(Timeline$Dislexia_clean, Timeline$PA_clean, method = "spearman") ## Problemas de Aprendizagem## Warning in cor.test.default(Timeline$Dislexia_clean, Timeline$PA_clean, :
## Cannot compute exact p-value with tiescor.dis.de = cor.test(Timeline$Dislexia_clean, Timeline$DE_clean, method = "spearman") ## Desempenho escolar## Warning in cor.test.default(Timeline$Dislexia_clean, Timeline$DE_clean, :
## Cannot compute exact p-value with tiescor.dis.da = cor.test(Timeline$Dislexia_clean, Timeline$DA_clean, method = "spearman") ## Dificuldade de Aprendizagem## Warning in cor.test.default(Timeline$Dislexia_clean, Timeline$DA_clean, :
## Cannot compute exact p-value with tiescor.dis.dis2 = cor.test(Timeline$Dislexia_clean, Timeline$Discalculia_clean, method = "spearman") #Discalculia## Warning in cor.test.default(Timeline$Dislexia_clean, Timeline
## $Discalculia_clean, : Cannot compute exact p-value with tiescor.dis.dis3 = cor.test(Timeline$Dislexia_clean, Timeline$Disgrafia_clean, method = "spearman") #Digrafia## Warning in cor.test.default(Timeline$Dislexia_clean, Timeline
## $Disgrafia_clean, : Cannot compute exact p-value with tiescor.dis.ta = cor.test(Timeline$Dislexia_clean, Timeline$TA_clean, method = "spearman") ## Transtorno de Aprendizagem## Warning in cor.test.default(Timeline$Dislexia_clean, Timeline$TA_clean, :
## Cannot compute exact p-value with tiesNome_dis = c("Dislexia com Dif.s de Aprend.", " com Fracasso Escolar", " com Problema de Aprend.", " com Desempenho Escolar", " com Dif. de Aprend.", " com Discalculia", " com Disgrafia", " com Transt. de Aprend.")
pvalor_cor_dis = c(cor.dis.das$p.value, cor.dis.fe$p.value, cor.dis.pa$p.value, cor.dis.de$p.value, cor.dis.da$p.value, cor.dis.dis2$p.value, cor.dis.dis3$p.value, cor.dis.ta$p.value)
valor_cor_dis = c(cor.dis.das$estimate, cor.dis.fe$estimate, cor.dis.pa$estimate, cor.dis.de$estimate, cor.dis.da$estimate, cor.dis.dis2$estimate, cor.dis.dis3$estimate, cor.dis.ta$estimate)
tab_cor_dis = data.frame(Nome_dis, valor_cor_dis, pvalor_cor_dis)
tab_cor_dis## Nome_dis valor_cor_dis pvalor_cor_dis
## 1 Dislexia com Dif.s de Aprend. 0.7775874 3.442925e-29
## 2 com Fracasso Escolar 0.8303787 1.304910e-36
## 3 com Problema de Aprend. 0.7696528 1.769850e-28
## 4 com Desempenho Escolar 0.4603439 1.515041e-08
## 5 com Dif. de Aprend. 0.7935912 2.345935e-31
## 6 com Discalculia 0.3494107 3.270965e-05
## 7 com Disgrafia 0.7623783 3.038961e-28
## 8 com Transt. de Aprend. 0.2593444 2.211352e-03Grafico de dispersão com linha de regressão, método loess (não parametrico) - Dif.s de Aprendizagem, retirados outliers
library("ggplot2")
library("devtools")
ggplot(Timeline, aes(y=Dislexia_clean, x=DAs_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 19 rows containing non-finite values (stat_smooth).## Warning: Removed 19 rows containing missing values (geom_point).
qplot(DAs_clean, Dislexia_clean, data = Timeline, geom = c("point", "smooth"))## `geom_smooth()` using method = 'loess'## Warning: Removed 19 rows containing non-finite values (stat_smooth).
## Warning: Removed 19 rows containing missing values (geom_point).
ggplot(Timeline, aes(y=Dislexia_clean, x=FE_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 18 rows containing non-finite values (stat_smooth).## Warning: Removed 18 rows containing missing values (geom_point).
qplot(FE_clean, Dislexia_clean, data = Timeline, geom = c("point", "smooth"))## `geom_smooth()` using method = 'loess'## Warning: Removed 18 rows containing non-finite values (stat_smooth).
## Warning: Removed 18 rows containing missing values (geom_point).
ggplot(Timeline, aes(y=Dislexia_clean, x=PA_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 18 rows containing non-finite values (stat_smooth).
## Warning: Removed 18 rows containing missing values (geom_point).
qplot(PA_clean, Dislexia_clean, data = Timeline, geom = c("point", "smooth"))## `geom_smooth()` using method = 'loess'## Warning: Removed 18 rows containing non-finite values (stat_smooth).
## Warning: Removed 18 rows containing missing values (geom_point).
ggplot(Timeline, aes(y=Dislexia_clean, x=DE_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 20 rows containing non-finite values (stat_smooth).## Warning: Removed 20 rows containing missing values (geom_point).
qplot(DE_clean, Dislexia_clean, data = Timeline, geom = c("point", "smooth"))## `geom_smooth()` using method = 'loess'## Warning: Removed 20 rows containing non-finite values (stat_smooth).
## Warning: Removed 20 rows containing missing values (geom_point).
ggplot(Timeline, aes(y=Dislexia_clean, x=DA_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 18 rows containing non-finite values (stat_smooth).## Warning: Removed 18 rows containing missing values (geom_point).
qplot(DA_clean, Dislexia_clean, data = Timeline, geom = c("point", "smooth"))## `geom_smooth()` using method = 'loess'## Warning: Removed 18 rows containing non-finite values (stat_smooth).
## Warning: Removed 18 rows containing missing values (geom_point).
ggplot(Timeline, aes(y=Dislexia_clean, x=Discalculia_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 22 rows containing non-finite values (stat_smooth).## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : pseudoinverse used at -0.015## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : neighborhood radius 2.015## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : reciprocal condition number 1.03e-016## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : There are other near singularities as well. 4.0602## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : pseudoinverse used
## at -0.015## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : neighborhood radius
## 2.015## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : reciprocal
## condition number 1.03e-016## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : There are other
## near singularities as well. 4.0602## Warning: Removed 22 rows containing missing values (geom_point).
qplot(Discalculia_clean, Dislexia_clean, data = Timeline, geom = c("point", "smooth"))## `geom_smooth()` using method = 'loess'## Warning: Removed 22 rows containing non-finite values (stat_smooth).## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : pseudoinverse used at -0.015## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : neighborhood radius 2.015## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : reciprocal condition number 1.03e-016## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : There are other near singularities as well. 4.0602## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : pseudoinverse used
## at -0.015## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : neighborhood radius
## 2.015## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : reciprocal
## condition number 1.03e-016## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : There are other
## near singularities as well. 4.0602## Warning: Removed 22 rows containing missing values (geom_point).
ggplot(Timeline, aes(y=Dislexia_clean, x=Disgrafia_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 15 rows containing non-finite values (stat_smooth).## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : pseudoinverse used at 2## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : neighborhood radius 1## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : reciprocal condition number 0## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : pseudoinverse used
## at 2## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : neighborhood radius
## 1## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : reciprocal
## condition number 0## Warning: Removed 15 rows containing missing values (geom_point).
qplot(Disgrafia_clean, Dislexia_clean, data = Timeline, geom = c("point", "smooth"))## `geom_smooth()` using method = 'loess'## Warning: Removed 15 rows containing non-finite values (stat_smooth).## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : pseudoinverse used at 2## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : neighborhood radius 1## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : reciprocal condition number 0## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : pseudoinverse used
## at 2## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : neighborhood radius
## 1## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : reciprocal
## condition number 0## Warning: Removed 15 rows containing missing values (geom_point).
ggplot(Timeline, aes(y=Dislexia_clean, x=TA_clean)) + geom_point(shape=1) + geom_smooth(method=loess)## Warning: Removed 20 rows containing non-finite values (stat_smooth).## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : pseudoinverse used at -0.01## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : neighborhood radius 1.01## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : reciprocal condition number 0## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : There are other near singularities as well. 4.0401## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : pseudoinverse used
## at -0.01## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : neighborhood radius
## 1.01## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : reciprocal
## condition number 0## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : There are other
## near singularities as well. 4.0401## Warning: Removed 20 rows containing missing values (geom_point).
qplot(TA_clean, Dislexia_clean, data = Timeline, geom = c("point", "smooth"))## `geom_smooth()` using method = 'loess'## Warning: Removed 20 rows containing non-finite values (stat_smooth).## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : pseudoinverse used at -0.01## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : neighborhood radius 1.01## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : reciprocal condition number 0## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : There are other near singularities as well. 4.0401## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : pseudoinverse used
## at -0.01## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : neighborhood radius
## 1.01## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : reciprocal
## condition number 0## Warning in predLoess(object$y, object$x, newx = if
## (is.null(newdata)) object$x else if (is.data.frame(newdata))
## as.matrix(model.frame(delete.response(terms(object)), : There are other
## near singularities as well. 4.0401## Warning: Removed 20 rows containing missing values (geom_point).
Modelo de Regressao linear multipla (com testes de normalidade, homogeneidade e dependencia dos residuos) - Dislexia
#Regressão Linear
RL_Dis1 = lm(Dislexia ~ DAs + FE + PA + DE + Discalculia + Disgrafia + TA, data=Timeline)
summary( RL_Dis1 )##
## Call:
## lm(formula = Dislexia ~ DAs + FE + PA + DE + Discalculia + Disgrafia +
## TA, data = Timeline)
##
## Residuals:
## Min 1Q Median 3Q Max
## -27.441 -5.510 -0.710 2.743 59.849
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.35399 2.82475 8.976 1.13e-15 ***
## DAs -0.23414 0.11800 -1.984 0.0491 *
## FE 0.71912 0.12302 5.846 3.07e-08 ***
## PA 0.41386 0.23537 1.758 0.0807 .
## DE -0.44224 0.37831 -1.169 0.2443
## Discalculia -0.01647 1.01752 -0.016 0.9871
## Disgrafia 0.03539 0.79757 0.044 0.9647
## TA -3.37112 1.59567 -2.113 0.0363 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 11 on 149 degrees of freedom
## Multiple R-squared: 0.5298, Adjusted R-squared: 0.5077
## F-statistic: 23.99 on 7 and 149 DF, p-value: < 2.2e-16#Testes de normalidade de ajuste do residuo
lillie.test(RL_Dis1$residuals)##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: RL_Dis1$residuals
## D = 0.17193, p-value = 2.682e-12residualPlot(RL_Dis1)
#Regressão Linear, revidos outliers
RL_Dis = lm(Dislexia_clean ~ DAs_clean + FE_clean + PA_clean + DE_clean + Discalculia_clean + Disgrafia_clean + TA_clean, data=Timeline)
summary( RL_Dis )##
## Call:
## lm(formula = Dislexia_clean ~ DAs_clean + FE_clean + PA_clean +
## DE_clean + Discalculia_clean + Disgrafia_clean + TA_clean,
## data = Timeline)
##
## Residuals:
## Min 1Q Median 3Q Max
## -12.587 -2.578 -0.374 2.544 15.027
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 10.52645 1.60088 6.575 1.45e-09 ***
## DAs_clean 0.14754 0.07633 1.933 0.0557 .
## FE_clean 0.39563 0.06842 5.782 6.34e-08 ***
## PA_clean 0.33230 0.18022 1.844 0.0678 .
## DE_clean 0.39148 0.24356 1.607 0.1107
## Discalculia_clean -0.02204 0.66670 -0.033 0.9737
## Disgrafia_clean 0.47063 0.76995 0.611 0.5422
## TA_clean -0.31212 0.85288 -0.366 0.7151
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.276 on 116 degrees of freedom
## (33 observations deleted due to missingness)
## Multiple R-squared: 0.7328, Adjusted R-squared: 0.7167
## F-statistic: 45.45 on 7 and 116 DF, p-value: < 2.2e-16#Testes de normalidade de ajuste do residuo
lillie.test(RL_Dis$residuals)##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: RL_Dis$residuals
## D = 0.067493, p-value = 0.18residualPlot(RL_Dis)
#Ajuste de modelo pelo método Backward
#retirando dislcaculia, p = 0,9737
RL_Dis_aj1 = lm( Dislexia_clean ~ DAs_clean + FE_clean + PA_clean + DE_clean + Disgrafia_clean + TA_clean, data=Timeline)
summary(RL_Dis_aj1)##
## Call:
## lm(formula = Dislexia_clean ~ DAs_clean + FE_clean + PA_clean +
## DE_clean + Disgrafia_clean + TA_clean, data = Timeline)
##
## Residuals:
## Min 1Q Median 3Q Max
## -11.9415 -2.5155 -0.5345 2.5451 15.5681
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 10.75901 1.55378 6.924 2.32e-10 ***
## DAs_clean 0.14681 0.07713 1.903 0.0594 .
## FE_clean 0.33840 0.06408 5.281 5.82e-07 ***
## PA_clean 0.44907 0.17694 2.538 0.0124 *
## DE_clean 0.30316 0.24020 1.262 0.2093
## Disgrafia_clean 0.67710 0.74728 0.906 0.3667
## TA_clean -0.76740 0.83231 -0.922 0.3584
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.38 on 120 degrees of freedom
## (30 observations deleted due to missingness)
## Multiple R-squared: 0.7183, Adjusted R-squared: 0.7043
## F-statistic: 51.01 on 6 and 120 DF, p-value: < 2.2e-16#retirando disgrafia, p = 0,3667
RL_Dis_aj2 = lm( Dislexia_clean ~ DAs_clean + FE_clean + PA_clean + DE_clean + TA_clean, data=Timeline)
summary(RL_Dis_aj2)##
## Call:
## lm(formula = Dislexia_clean ~ DAs_clean + FE_clean + PA_clean +
## DE_clean + TA_clean, data = Timeline)
##
## Residuals:
## Min 1Q Median 3Q Max
## -11.833 -2.752 -0.384 2.503 15.128
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 10.80233 1.55190 6.961 1.88e-10 ***
## DAs_clean 0.17132 0.07218 2.373 0.01920 *
## FE_clean 0.34973 0.06280 5.569 1.58e-07 ***
## PA_clean 0.48416 0.17252 2.806 0.00584 **
## DE_clean 0.30986 0.23991 1.292 0.19896
## TA_clean -0.69829 0.82819 -0.843 0.40081
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.376 on 121 degrees of freedom
## (30 observations deleted due to missingness)
## Multiple R-squared: 0.7164, Adjusted R-squared: 0.7047
## F-statistic: 61.13 on 5 and 121 DF, p-value: < 2.2e-16#retirando Transtorno de Aprendizagem, p = 0,40081
RL_Dis_aj3 = lm( Dislexia_clean ~ DAs_clean + FE_clean + PA_clean + DE_clean, data=Timeline)
summary(RL_Dis_aj3)##
## Call:
## lm(formula = Dislexia_clean ~ DAs_clean + FE_clean + PA_clean +
## DE_clean, data = Timeline)
##
## Residuals:
## Min 1Q Median 3Q Max
## -12.0563 -2.6252 -0.3376 2.4554 15.8577
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 10.61148 1.50119 7.069 1.03e-10 ***
## DAs_clean 0.17203 0.07146 2.407 0.01755 *
## FE_clean 0.34293 0.06188 5.542 1.73e-07 ***
## PA_clean 0.48198 0.17134 2.813 0.00572 **
## DE_clean 0.27460 0.22984 1.195 0.23450
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.356 on 123 degrees of freedom
## (29 observations deleted due to missingness)
## Multiple R-squared: 0.7176, Adjusted R-squared: 0.7084
## F-statistic: 78.15 on 4 and 123 DF, p-value: < 2.2e-16#Retirando Desempenho Escolar, p = 0,23450
RL_Dis_aj4 = lm( Dislexia_clean ~ DAs_clean + FE_clean + PA_clean, data=Timeline)
summary(RL_Dis_aj4)##
## Call:
## lm(formula = Dislexia_clean ~ DAs_clean + FE_clean + PA_clean,
## data = Timeline)
##
## Residuals:
## Min 1Q Median 3Q Max
## -12.485 -2.823 -0.171 2.425 17.152
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 12.51395 1.39305 8.983 2.89e-15 ***
## DAs_clean 0.15267 0.06987 2.185 0.0307 *
## FE_clean 0.33976 0.06520 5.211 7.31e-07 ***
## PA_clean 0.51939 0.16696 3.111 0.0023 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.666 on 128 degrees of freedom
## (25 observations deleted due to missingness)
## Multiple R-squared: 0.6858, Adjusted R-squared: 0.6785
## F-statistic: 93.15 on 3 and 128 DF, p-value: < 2.2e-16par(mfrow=c(2,2))
plot(RL_Dis_aj4)
#Teste de normalidade de ajuste do residuo
shapiro.test(RL_Dis_aj4$residuals)##
## Shapiro-Wilk normality test
##
## data: RL_Dis_aj4$residuals
## W = 0.95136, p-value = 0.0001274sf.test(RL_Dis_aj4$residuals)##
## Shapiro-Francia normality test
##
## data: RL_Dis_aj4$residuals
## W = 0.94688, p-value = 0.0001391ad.test(RL_Dis_aj4$residuals)##
## Anderson-Darling normality test
##
## data: RL_Dis_aj4$residuals
## A = 1.3736, p-value = 0.001418cvm.test(RL_Dis_aj4$residuals)##
## Cramer-von Mises normality test
##
## data: RL_Dis_aj4$residuals
## W = 0.18547, p-value = 0.008073lillie.test(RL_Dis_aj4$residuals)##
## Lilliefors (Kolmogorov-Smirnov) normality test
##
## data: RL_Dis_aj4$residuals
## D = 0.076317, p-value = 0.05712pearson.test(RL_Dis_aj4$residuals)##
## Pearson chi-square normality test
##
## data: RL_Dis_aj4$residuals
## P = 16.409, p-value = 0.1732#Teste Goldfeld-Quandt - test against heteroskedasticity (homoscedasticidade, ou variância constante dos erros, dos resíduos) n > 30
## p > 0,05 não rejeitamos a hipótese de homoscedasticidade dos erros
gqtest(RL_Dis_aj4)##
## Goldfeld-Quandt test
##
## data: RL_Dis_aj4
## GQ = 0.33945, df1 = 62, df2 = 62, p-value = 1
## alternative hypothesis: variance increases from segment 1 to 2#Teste de Breusch-Pagan: hipótese nula de que as variâncias dos erros são iguais (homoscedasticidade)
##p > 0,05 não rejeitamos a hipótese de homoscedasticidade dos erros
bptest(RL_Dis_aj4)##
## studentized Breusch-Pagan test
##
## data: RL_Dis_aj4
## BP = 15.84, df = 3, p-value = 0.001223#Teste de independencia - Teste de Durbin Watson
## p > 0,05, não rejeita Hipotese nula, os resíduos são independentes.
plot(RL_Dis_aj4$residuals)
dwtest(RL_Dis_aj4)##
## Durbin-Watson test
##
## data: RL_Dis_aj4
## DW = 1.5059, p-value = 0.001571
## alternative hypothesis: true autocorrelation is greater than 0
Grafico de tendencia utilizando Regressao linear # borrowed from amunatagui- # http://amunategui.github.io/google-trends-walkthrough/
library(ggplot2)
ggplot(Timeline,aes(x = Date)) +
stat_smooth(aes(y = DA_clean, group=1, colour="Dif. de Aprendizagem"), method=lm, formula = y ~ poly(x,1), level=0.95) +
stat_smooth(aes(y = DAs_clean, group=1, colour= "Dif.s de Aprendizagem"), method=lm, formula = y ~ poly(x,2), level=0.95) +
stat_smooth(aes(y = Dislexia_clean, group=1, colour="Dislexia"), method=lm, formula = y ~ poly(x,3), level=0.95) +
stat_smooth(aes(y = FE_clean, group=1, colour="Fracasso Escolar"), method=lm, formula = y ~ poly(x,4), level=0.95) +
stat_smooth(aes(y = PA_clean, group=1, colour= "Problemas de Aprendizagem"), method=lm, formula = y ~ poly(x,5), level=0.95) +
stat_smooth(aes(y = DE_clean, group=1, colour= "Desempenho Escolar"), method=lm, formula = y ~ poly(x,6), level=0.95) +
stat_smooth(aes(y = Discalculia_clean, group=1, colour="Discalculia"), method=lm, formula = y ~ poly(x,7), level=0.95) +
stat_smooth(aes(y = Disgrafia_clean, group=1, colour="Disgrafia"), method=lm, formula = y ~ poly(x,8), level=0.95) +
stat_smooth(aes(y = TA_clean, group=1, colour="Transtorno da Aprendizagem"), method=lm, formula = y ~ poly(x,9), level=0.95) +
geom_point (aes(y = DA_clean, colour = "Dif. de Aprendizagem"), size=1) +
geom_point (aes(y = DAs_clean, colour = "Dif.s de Aprendizagem"), size=1) +
geom_point (aes(y = Dislexia_clean, colour ="Dislexia"), size=1) +
geom_point (aes(y = FE_clean, colour ="Fracasso Escolar"), size=1) +
geom_point (aes(y = PA_clean, colour ="Problemas de Aprendizagem"), size=1) +
geom_point (aes(y = DE_clean, colour ="Desempenho Escolar"), size=1) +
geom_point (aes(y = Discalculia_clean, colour ="Discalculia"), size=1) +
geom_point (aes(y = Disgrafia_clean, colour ="Disgrafia"), size=1) +
geom_point (aes(y = TA_clean, colour = "Transtorno da Aprendizagem"), size=1) +
scale_colour_manual("Termos de Pesquisa", breaks = c("Dif. de Aprendizagem", "Dif.s de Aprendizagem", "Dislexia", "Fracasso Escolar", "Problemas de Aprendizagem", "Desempenho Escolar", "Discalculia", "Disgrafia", "Transtorno da Aprendizagem"), values = c("blue","green", "red", "darkgreen", "darkred", "darkblue", "purple", "orange", "pink")) +
theme_bw() +
xlab("Periodo") +
ylab("frequencia relativa") +
ggtitle("Regressao linear das Palavras-chave")## Warning: Removed 10 rows containing non-finite values (stat_smooth).## Warning: Removed 11 rows containing non-finite values (stat_smooth).## Warning: Removed 12 rows containing non-finite values (stat_smooth).## Warning: Removed 10 rows containing non-finite values (stat_smooth).## Warning: Removed 12 rows containing non-finite values (stat_smooth).## Warning: Removed 10 rows containing non-finite values (stat_smooth).## Warning: Removed 12 rows containing non-finite values (stat_smooth).## Warning: Removed 4 rows containing non-finite values (stat_smooth).## Warning: Removed 10 rows containing non-finite values (stat_smooth).## Warning: Removed 10 rows containing missing values (geom_point).## Warning: Removed 11 rows containing missing values (geom_point).## Warning: Removed 12 rows containing missing values (geom_point).## Warning: Removed 10 rows containing missing values (geom_point).## Warning: Removed 12 rows containing missing values (geom_point).## Warning: Removed 10 rows containing missing values (geom_point).## Warning: Removed 12 rows containing missing values (geom_point).## Warning: Removed 4 rows containing missing values (geom_point).## Warning: Removed 10 rows containing missing values (geom_point).
Grafico de tendencia utilizando Regressao Linear com a retirada de outliers - ultimos 5 anos # borrowed from aL3xa - # http://stackoverflow.com/questions/4787332/how-to-remove-outliers-from-a-dataset
library(ggplot2)
ggplot(Timeline_5anos,aes(x=Date)) +
stat_smooth(aes(y = DA_clean, group=1, colour="Dif. de Aprendizagem"), method=lm, formula = y ~ poly(x,1), level=0.95) +
stat_smooth(aes(y = DAs_clean, group=1, colour= "Dif.s de Aprendizagem"), method=lm, formula = y ~ poly(x,2), level=0.95) +
stat_smooth(aes(y = Dislexia_clean, group=1, colour="Dislexia"), method=lm, formula = y ~ poly(x,3), level=0.95) +
stat_smooth(aes(y = FE_clean, group=1, colour="Fracasso Escolar"), method=lm, formula = y ~ poly(x,4), level=0.95) +
stat_smooth(aes(y = PA_clean, group=1, colour= "Problemas de Aprendizagem"), method=lm, formula = y ~ poly(x,5), level=0.95) +
stat_smooth(aes(y = DE_clean, group=1, colour= "Desempenho Escolar"), method=lm, formula = y ~ poly(x,6), level=0.95) +
stat_smooth(aes(y = Discalculia_clean, group=1, colour="Discalculia"), method=lm, formula = y ~ poly(x,7), level=0.95) +
stat_smooth(aes(y = Disgrafia_clean, group=1, colour="Disgrafia"), method=lm, formula = y ~ poly(x,8), level=0.95) +
stat_smooth(aes(y = TA_clean, group=1, colour="Transtorno da Aprendizagem"), method=lm, formula = y ~ poly(x,9), level=0.95) +
geom_point (aes(y = DA_clean, colour = "Dif. de Aprendizagem"), size=1) +
geom_point (aes(y = DAs_clean, colour = "Dif.s de Aprendizagem"), size=1) +
geom_point (aes(y = Dislexia_clean, colour ="Dislexia"), size=1) +
geom_point (aes(y = FE_clean, colour ="Fracasso Escolar"), size=1) +
geom_point (aes(y = PA_clean, colour ="Problemas de Aprendizagem"), size=1) +
geom_point (aes(y = DE_clean, colour ="Desempenho Escolar"), size=1) +
geom_point (aes(y = Discalculia_clean, colour ="Discalculia"), size=1) +
geom_point (aes(y = Disgrafia_clean, colour ="Disgrafia"), size=1) +
geom_point (aes(y = TA_clean, colour = "Transtorno da Aprendizagem"), size=1) +
scale_colour_manual("Termos de Pesquisa", breaks = c("Dif. de Aprendizagem", "Dif.s de Aprendizagem", "Dislexia", "Fracasso Escolar", "Problemas de Aprendizagem", "Desempenho Escolar", "Discalculia", "Disgrafia", "Transtorno da Aprendizagem"), values = c("blue","green", "red", "darkgreen", "darkred", "darkblue", "purple", "orange", "pink")) +
theme_bw() +
xlab("Periodo") +
ylab("frequencia relativa") +
ggtitle("Regressao linear das Palavras-chave, sem outliers, últimos 5 anos")## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 1 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 1 rows containing missing values (geom_point).
Series temporais: avaliacao de sazonalidade e tendencia “Série Temporal é um conjunto de observações sobre uma variável, ordenado no tempo”, em períodos regulares. Intui identificar padrões não aleatórios de uma variável de interesse permitindo previsões e orientando a tomada de decisões. # borrowed from marcos-lopes #https://gist.github.com/marco-lopes/8550667
# Ativando o módulo de séries temporais
library(tseries)## Warning: package 'tseries' was built under R version 3.3.3library(forecast)## Warning: package 'forecast' was built under R version 3.3.3# Gráfico da serie - apresente o comportamento da média dos dados referente as váriaveis
ts.plot(Timeline$DAs, main = "Dif.s de Aprendizagem")
ts.plot(Timeline$DA, main = "Dif. de Aprendizagem" )
ts.plot(Timeline$Dislexia, main = "Dislexia")
ts.plot(Timeline$FE, main = "Fracasso Escolar")
ts.plot(Timeline$DE, main = "Desempenho Escolar")
ts.plot(Timeline$PA, main = "Problemas de Aprendizagem")
ts.plot(Timeline$Discalculia, main = "Discalculia")
ts.plot(Timeline$Disgrafia, main = "Disgrafia")
ts.plot(Timeline$TA, main = "Transtorno de Aprendizagem")
#TESTE DE COX-STUART: distribuicao binomial para avaliar tendencia, ou evolução dos valores observados, na série (p < 0,05, existe tendencia).
library(randtests)##
## Attaching package: 'randtests'## The following object is masked from 'package:tseries':
##
## runs.testcox.stuart.test(Timeline$DAs)##
## Cox Stuart test
##
## data: Timeline$DAs
## statistic = 9, n = 78, p-value = 1.381e-12
## alternative hypothesis: non randomnesscox.stuart.test(Timeline$DA)##
## Cox Stuart test
##
## data: Timeline$DA
## statistic = 10, n = 77, p-value = 1.697e-11
## alternative hypothesis: non randomnesscox.stuart.test(Timeline$Dislexia)##
## Cox Stuart test
##
## data: Timeline$Dislexia
## statistic = 7, n = 76, p-value = 6.418e-14
## alternative hypothesis: non randomnesscox.stuart.test(Timeline$FE)##
## Cox Stuart test
##
## data: Timeline$FE
## statistic = 0, n = 78, p-value < 2.2e-16
## alternative hypothesis: non randomnesscox.stuart.test(Timeline$DE)##
## Cox Stuart test
##
## data: Timeline$DE
## statistic = 18, n = 69, p-value = 8.769e-05
## alternative hypothesis: non randomnesscox.stuart.test(Timeline$PA)##
## Cox Stuart test
##
## data: Timeline$PA
## statistic = 4, n = 75, p-value < 2.2e-16
## alternative hypothesis: non randomnesscox.stuart.test(Timeline$Discalculia)##
## Cox Stuart test
##
## data: Timeline$Discalculia
## statistic = 16, n = 55, p-value = 0.002667
## alternative hypothesis: non randomnesscox.stuart.test(Timeline$Disgrafia)##
## Cox Stuart test
##
## data: Timeline$Disgrafia
## statistic = 2, n = 58, p-value = 1.188e-14
## alternative hypothesis: non randomnesscox.stuart.test(Timeline$TA)##
## Cox Stuart test
##
## data: Timeline$TA
## statistic = 14, n = 53, p-value = 0.0008023
## alternative hypothesis: non randomness#TESTE DE WALD-WOLFOWITS: teste não-paramétrico para verificar a aleatoriedade de uma sequência de dados, Em série temporal avalia tendencia.
library(adehabitat)## Warning: package 'adehabitat' was built under R version 3.3.3## Loading required package: ade4## Warning: package 'ade4' was built under R version 3.3.3## Loading required package: tkrplot## Loading required package: tcltk## Loading required package: shapefiles## Warning: package 'shapefiles' was built under R version 3.3.3## Loading required package: foreign##
## Attaching package: 'shapefiles'## The following objects are masked from 'package:foreign':
##
## read.dbf, write.dbf## Loading required package: sp## Warning: package 'sp' was built under R version 3.3.3##
## ************************************************
## ************************************************
## THE PACKAGE adehabitat IS NOW DEPRECATED!!!!!!!
## It is dangerous to use it, as bugs will no longer be corrected.
## It is now recommended to use the packages adehabitatMA, adehabitatLT, adehabitatHR, and adehabitatHS.
## These 4 packages are the future of adehabitat.
## They have a vignette explaining in detail how they can be used.
## They implement more methods than adehabitat
## They are based on the more common and more clever spatial classes implemented in sp.
## Bugs are corrected frequently.
## Really, avoid to use the classical adehabitat, unless you have a very good reason for it.
##
## *****THIS IS THE VERY LAST WARNING*****
## This is the last version of adehabitat submitted to CRAN (at the time of writing: 2015-03-27).
## THE NEXT VERSION OF adehabitat WILL JUST BE A VIRTUAL PACKAGE LOADING ALL THE PACKAGES DESCRIBED ABOVE.##
## Attaching package: 'adehabitat'## The following object is masked from 'package:zoo':
##
## is.regularwawotest(Timeline$DAs)## a ea va za p
## 109.306108 -1.000000 149.880596 9.010043 0.000000wawotest(Timeline$DA)## a ea va za p
## 107.161307 -1.000000 149.260260 8.853191 0.000000wawotest(Timeline$Dislexia)## a ea va za p
## 130.02448 -1.00000 148.68584 10.74528 0.00000wawotest(Timeline$FE)## a ea va za p
## 136.88907 -1.00000 150.70706 11.23215 0.00000wawotest(Timeline$DE)## a ea va za p
## 5.613169e+01 -1.000000e+00 1.461242e+02 4.726243e+00 1.143558e-06wawotest(Timeline$PA)## a ea va za p
## 118.602500 -1.000000 147.777092 9.838677 0.000000wawotest(Timeline$Discalculia)## a ea va za p
## 6.440443e+01 -1.000000e+00 1.499614e+02 5.340937e+00 4.623381e-08wawotest(Timeline$Disgrafia)## a ea va za p
## 8.985648e+01 -1.000000e+00 1.422142e+02 7.618762e+00 1.276756e-14wawotest(Timeline$TA)## a ea va za p
## 9.273653e+01 -1.000000e+00 1.503839e+02 7.643781e+00 1.054712e-14#teste de Mann-Kendall para tendencia, avalia se a serie é serialmente independente e igualmente distribuidas
library(Kendall)## Warning: package 'Kendall' was built under R version 3.3.3MannKendall(Timeline$DAs)## tau = -0.573, 2-sided pvalue =< 2.22e-16MannKendall(Timeline$DA)## tau = -0.494, 2-sided pvalue =< 2.22e-16MannKendall(Timeline$Dislexia)## tau = -0.539, 2-sided pvalue =< 2.22e-16MannKendall(Timeline$FE)## tau = -0.73, 2-sided pvalue =< 2.22e-16MannKendall(Timeline$DE)## tau = -0.335, 2-sided pvalue =3.4537e-09MannKendall(Timeline$PA)## tau = -0.6, 2-sided pvalue =< 2.22e-16MannKendall(Timeline$Discalculia)## tau = -0.238, 2-sided pvalue =9.1823e-05MannKendall(Timeline$Disgrafia)## tau = -0.53, 2-sided pvalue =< 2.22e-16MannKendall(Timeline$TA)## tau = -0.297, 2-sided pvalue =2.2074e-06#últimos cinco anos
cox.stuart.test(Timeline_5anos$DAs)##
## Cox Stuart test
##
## data: Timeline_5anos$DAs
## statistic = 7, n = 27, p-value = 0.01916
## alternative hypothesis: non randomnesscox.stuart.test(Timeline_5anos$DA)##
## Cox Stuart test
##
## data: Timeline_5anos$DA
## statistic = 15, n = 27, p-value = 0.7011
## alternative hypothesis: non randomnesscox.stuart.test(Timeline_5anos$Dislexia)##
## Cox Stuart test
##
## data: Timeline_5anos$Dislexia
## statistic = 11, n = 27, p-value = 0.4421
## alternative hypothesis: non randomnesscox.stuart.test(Timeline_5anos$FE)##
## Cox Stuart test
##
## data: Timeline_5anos$FE
## statistic = 7, n = 24, p-value = 0.06391
## alternative hypothesis: non randomnesscox.stuart.test(Timeline_5anos$DE)##
## Cox Stuart test
##
## data: Timeline_5anos$DE
## statistic = 11, n = 24, p-value = 0.8388
## alternative hypothesis: non randomnesscox.stuart.test(Timeline_5anos$PA)##
## Cox Stuart test
##
## data: Timeline_5anos$PA
## statistic = 8, n = 22, p-value = 0.2863
## alternative hypothesis: non randomnesscox.stuart.test(Timeline_5anos$Discalculia)##
## Cox Stuart test
##
## data: Timeline_5anos$Discalculia
## statistic = 11, n = 14, p-value = 0.05737
## alternative hypothesis: non randomnesscox.stuart.test(Timeline_5anos$Disgrafia)##
## Cox Stuart test
##
## data: Timeline_5anos$Disgrafia
## statistic = 6, n = 13, p-value = 1
## alternative hypothesis: non randomnesscox.stuart.test(Timeline_5anos$TA)##
## Cox Stuart test
##
## data: Timeline_5anos$TA
## statistic = 14, n = 17, p-value = 0.01273
## alternative hypothesis: non randomnessAnalise de series temporais, por mes ##borrowed from claudio ##https://gustibuseconomia.com/2014/03/26/momento-r-do-dia-furacoes-ou-uma-imagem-vale-mais-do-que-mil-palavras-mas-qual-imagem-e-esta-a-pergunta-de-um-milhao-de-imagens/
Timeline$mes = as.factor(c(rep(month.name, 13), "January"))
Timeline_5anos$mes = as.factor(c(rep(month.name, 5)))
#equivalencia dos dados com o mes
Timeline$ordem = as.factor(c(rep(1:12, 13), 1))
Timeline_5anos$ordem = as.factor(c(rep(1:12, 5)))
library(ggplot2)
library(lattice)
library(latticeExtra) ## Loading required package: RColorBrewer##
## Attaching package: 'latticeExtra'## The following object is masked from 'package:ggplot2':
##
## layer#If the notches of two plots do not overlap this is 'strong evidence' that the two #medians differ (Chambers et al, 1983, p. 62).
# os graficos por mes desde 2004
#Dislexia
op <- par(mfrow = c(1,2))
boxplot(Dislexia~ordem,data=Timeline)
monthplot(Timeline$Dislexia, col = "purple",ylim=c(min(Timeline$Dislexia),max(Timeline$Dislexia)),main="Dislexia",xlab="meses",ylab="media de frequencia")
par(op)
#Dificuldades de Aprendizagem
op <- par(mfrow = c(1,2))
boxplot(DAs~ordem,data=Timeline)
monthplot(Timeline$DAs, col="darkblue",ylim=c(min(Timeline$DAs),max(Timeline$DAs)),main="Dif. de Aprendizagem",xlab="meses",ylab="media de frequencia")
par(op)
#Organizacao por mes
op <- par(mfrow = c(2,2))
monthplot(Timeline$Dislexia, xlab = "Dislexia", ylab = "", col="purple", cex.axis = 0.8)
monthplot(Timeline$DAs, xlab = "Dif.s de Aprendizagem", ylab = "", col="darkblue", cex.axis = 0.8)
monthplot(Timeline$Dislexia, ylab = "", type = "h", col="purple", cex.axis = 0.8)
monthplot(Timeline$DAs, ylab = "", type = "h", col="darkblue", cex.axis = 0.8)
par(op)
#Organizacao por mes nos utimos cinco anos
op <- par(mfrow = c(2,2))
monthplot(Timeline_5anos$Dislexia, xlab = "Dislexia", ylab = "", col="purple", cex.axis = 0.8)
monthplot(Timeline_5anos$DAs, xlab = "Dif.s de Aprendizagem", ylab = "", col="darkblue", cex.axis = 0.8)
monthplot(Timeline_5anos$Dislexia, ylab = "", type = "h", col="purple", cex.axis = 0.8)
monthplot(Timeline_5anos$DAs, ylab = "", type = "h", col="darkblue", cex.axis = 0.8)
par(op)
op <- par(mfrow = c(2,2))
boxplot(Timeline_5anos$Dislexia~Timeline_5anos$ordem, main="Dislexia", xlab="mes", ylab="frequencia", outline = F, col="purple")
boxplot(Timeline_5anos$DAs~Timeline_5anos$ordem, main="Dif.s de Aprendizagem", xlab="mes", ylab="frequencia", outline = F, col="darkblue")
par(op)
Analise de variancias, por mes ##borrowed from Salvatore S. Mangiafico ##http://rcompanion.org/rcompanion/d_06.html
#install these packages if they are not already installed: ("dplyr"); ("FSA"); ("DescTools"); ("rcompanion"), ("multcompView")
##Dificuldade de Aprendizagem
### Specify the order of factor levels
library(dplyr)## Warning: package 'dplyr' was built under R version 3.3.3##
## Attaching package: 'dplyr'## The following object is masked from 'package:adehabitat':
##
## id## The following object is masked from 'package:car':
##
## recode## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionTimeline = mutate(Timeline_5anos,
mes = factor(mes, levels=unique(mes)))
# Medians and descriptive statistics
library(FSA)## Warning: package 'FSA' was built under R version 3.3.3## ## FSA v0.8.13. See citation('FSA') if used in publication.
## ## Run fishR() for related website and fishR('IFAR') for related book.Summarize(DAs ~ mes,data = Timeline_5anos)## mes n mean sd min Q1 median Q3 max
## 1 April 5 27.2 3.701351 23 24 28 29 32
## 2 August 5 23.0 3.674235 18 21 23 26 27
## 3 December 5 16.4 2.073644 15 15 16 16 20
## 4 February 5 17.0 2.828427 14 14 18 19 20
## 5 January 5 12.8 3.768289 9 11 12 13 19
## 6 July 5 20.0 4.358899 15 17 19 24 25
## 7 June 5 25.6 3.781534 22 23 24 28 31
## 8 March 5 25.6 2.701851 23 24 25 26 30
## 9 May 5 27.2 4.764452 23 24 26 28 35
## 10 November 5 25.8 2.588436 23 24 25 28 29
## 11 October 5 22.4 3.507136 19 20 22 23 28
## 12 September 5 26.4 3.646917 22 24 26 29 31# Histograms for each group
library(lattice)
histogram(~ DAs | ordem, data=Timeline_5anos, layout=c(12,1))
#Analise de variancia para dados nao-parametricos (p-value menos de 0,05, deve-se rejeitar a hipotese nula), interpreted as a difference in medians. applied to one-way data with more than two groups
kruskal.test(DAs ~ mes, data = Timeline_5anos)##
## Kruskal-Wallis rank sum test
##
## data: DAs by mes
## Kruskal-Wallis chi-squared = 37.858, df = 11, p-value = 8.269e-05#TESTES POS HOC
#Nemenyi test for multiple comparisons, not appropriate for groups with unequal numbers of observations. Utilizando qui quadrado.
library(DescTools)## Warning: package 'DescTools' was built under R version 3.3.3##
## Attaching package: 'DescTools'## The following object is masked from 'package:forecast':
##
## BoxCox## The following object is masked from 'package:car':
##
## RecodeNemenyiTest(x = Timeline_5anos$DAs, g = Timeline_5anos$mes, dist="tukey")##
## Nemenyi's test of multiple comparisons for independent samples (tukey)
##
## mean.rank.diff pval
## August-April -14.4 0.9789
## December-April -34.2 0.0831 .
## February-April -33.4 0.1016
## January-April -40.3 0.0140 *
## July-April -22.9 0.6421
## June-April -6.3 1.0000
## March-April -4.3 1.0000
## May-April -1.7 1.0000
## November-April -3.6 1.0000
## October-April -17.7 0.9087
## September-April -2.4 1.0000
## December-August -19.8 0.8225
## February-August -19.0 0.8593
## January-August -25.9 0.4443
## July-August -8.5 0.9998
## June-August 8.1 0.9999
## March-August 10.1 0.9990
## May-August 12.7 0.9923
## November-August 10.8 0.9981
## October-August -3.3 1.0000
## September-August 12.0 0.9953
## February-December 0.8 1.0000
## January-December -6.1 1.0000
## July-December 11.3 0.9972
## June-December 27.9 0.3234
## March-December 29.9 0.2225
## May-December 32.5 0.1262
## November-December 30.6 0.1927
## October-December 16.5 0.9427
## September-December 31.8 0.1483
## January-February -6.9 1.0000
## July-February 10.5 0.9986
## June-February 27.1 0.3698
## March-February 29.1 0.2601
## May-February 31.7 0.1517
## November-February 29.8 0.2270
## October-February 15.7 0.9596
## September-February 31.0 0.1769
## July-January 17.4 0.9182
## June-January 34.0 0.0874 .
## March-January 36.0 0.0514 .
## May-January 38.6 0.0240 *
## November-January 36.7 0.0422 *
## October-January 22.6 0.6615
## September-January 37.9 0.0297 *
## June-July 16.6 0.9402
## March-July 18.6 0.8759
## May-July 21.2 0.7473
## November-July 19.3 0.8461
## October-July 5.2 1.0000
## September-July 20.5 0.7865
## March-June 2.0 1.0000
## May-June 4.6 1.0000
## November-June 2.7 1.0000
## October-June -11.4 0.9970
## September-June 3.9 1.0000
## May-March 2.6 1.0000
## November-March 0.7 1.0000
## October-March -13.4 0.9881
## September-March 1.9 1.0000
## November-May -1.9 1.0000
## October-May -16.0 0.9538
## September-May -0.7 1.0000
## October-November -14.1 0.9821
## September-November 1.2 1.0000
## September-October 15.3 0.9666
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1require(PMCMR)## Loading required package: PMCMR## Warning: package 'PMCMR' was built under R version 3.3.3posthoc.kruskal.nemenyi.test(x = Timeline_5anos$DAs, g = Timeline_5anos$mes, dist="Chisquare")## Warning in posthoc.kruskal.nemenyi.test.default(x = Timeline_5anos$DAs, :
## Ties are present. Chi-sq was corrected for ties.##
## Pairwise comparisons using Nemenyi-test with Chi-squared
## approximation for independent samples
##
## data: Timeline_5anos$DAs and Timeline_5anos$mes
##
## April August December February January July June March May
## August 1.00 - - - - - - - -
## December 0.56 0.99 - - - - - - -
## February 0.60 0.99 1.00 - - - - - -
## January 0.27 0.90 1.00 1.00 - - - - -
## July 0.96 1.00 1.00 1.00 1.00 - - - -
## June 1.00 1.00 0.84 0.87 0.57 1.00 - - -
## March 1.00 1.00 0.77 0.80 0.47 0.99 1.00 - -
## May 1.00 1.00 0.65 0.69 0.34 0.98 1.00 1.00 -
## November 1.00 1.00 0.74 0.77 0.44 0.99 1.00 1.00 1.00
## October 1.00 1.00 1.00 1.00 0.96 1.00 1.00 1.00 1.00
## September 1.00 1.00 0.68 0.72 0.38 0.98 1.00 1.00 1.00
## November October
## August - -
## December - -
## February - -
## January - -
## July - -
## June - -
## March - -
## May - -
## November - -
## October 1.00 -
## September 1.00 1.00
##
## P value adjustment method: none##Dunn test for multiple comparisons; If the Kruskal-Wallis test is significant, a post-hoc analysis can be performed to determine which levels of the independent variable differ from each other level. Dunn's Kruskal-Wallis Multiple Comparisons
library(FSA)
dunnTest(DAs ~ mes, data=Timeline_5anos)## Dunn (1964) Kruskal-Wallis multiple comparison## p-values adjusted with the Holm method.## Comparison Z P.unadj P.adj
## 1 April - August 1.30691423 0.1912418440 1.00000000
## 2 April - December 3.10392129 0.0019097413 0.11649422
## 3 August - December 1.79700706 0.0723344988 1.00000000
## 4 April - February 3.03131494 0.0024349113 0.14365977
## 5 August - February 1.72440072 0.0846355286 1.00000000
## 6 December - February -0.07260635 0.9421193768 1.00000000
## 7 April - January 3.65754467 0.0002546429 0.01680643
## 8 August - January 2.35063045 0.0187416370 0.89959858
## 9 December - January 0.55362339 0.5798366168 1.00000000
## 10 February - January 0.62622973 0.5311642674 1.00000000
## 11 April - July 2.07835665 0.0376765222 1.00000000
## 12 August - July 0.77144243 0.4404447365 1.00000000
## 13 December - July -1.02556464 0.3050968439 1.00000000
## 14 February - July -0.95295829 0.3406112043 1.00000000
## 15 January - July -1.57918802 0.1142929362 1.00000000
## 16 April - June 0.57177497 0.5674744339 1.00000000
## 17 August - June -0.73513925 0.4622546910 1.00000000
## 18 December - June -2.53214631 0.0113366675 0.56683338
## 19 February - June -2.45953997 0.0139115207 0.68166451
## 20 January - June -3.08576970 0.0020302601 0.12181560
## 21 July - June -1.50658168 0.1319179168 1.00000000
## 22 April - March 0.39025911 0.6963449569 1.00000000
## 23 August - March -0.91665512 0.3593233924 1.00000000
## 24 December - March -2.71366218 0.0066543987 0.35268313
## 25 February - March -2.64105583 0.0082648098 0.42150530
## 26 January - March -3.26728557 0.0010858409 0.06732214
## 27 July - March -1.68809754 0.0913925062 1.00000000
## 28 June - March -0.18151586 0.8559626817 1.00000000
## 29 April - May 0.15428849 0.8773822762 1.00000000
## 30 August - May -1.15262574 0.2490640330 1.00000000
## 31 December - May -2.94963280 0.0031815181 0.18452805
## 32 February - May -2.87702646 0.0040144194 0.22480748
## 33 January - May -3.50325619 0.0004596072 0.02987447
## 34 July - May -1.92406817 0.0543460397 1.00000000
## 35 June - May -0.41748649 0.6763226048 1.00000000
## 36 March - May -0.23597062 0.8134554651 1.00000000
## 37 April - November 0.32672856 0.7438731996 1.00000000
## 38 August - November -0.98018567 0.3269944769 1.00000000
## 39 December - November -2.77719273 0.0054830656 0.29608554
## 40 February - November -2.70458638 0.0068389485 0.35562532
## 41 January - November -3.33081612 0.0008659178 0.05455282
## 42 July - November -1.75162809 0.0798377784 1.00000000
## 43 June - November -0.24504642 0.8064204908 1.00000000
## 44 March - November -0.06353055 0.9493440309 1.00000000
## 45 May - November 0.17244007 0.8630915715 1.00000000
## 46 April - October 1.60641540 0.1081826688 1.00000000
## 47 August - October 0.29950118 0.7645576742 1.00000000
## 48 December - October -1.49750588 0.1342616756 1.00000000
## 49 February - October -1.42489954 0.1541862431 1.00000000
## 50 January - October -2.05112927 0.0402543594 1.00000000
## 51 July - October -0.47194125 0.6369687248 1.00000000
## 52 June - October 1.03464043 0.3008368658 1.00000000
## 53 March - October 1.21615629 0.2239253894 1.00000000
## 54 May - October 1.45212692 0.1464663185 1.00000000
## 55 November - October 1.27968685 0.2006552924 1.00000000
## 56 April - September 0.21781904 0.8275701107 1.00000000
## 57 August - September -1.08909519 0.2761119115 1.00000000
## 58 December - September -2.88610225 0.0039004542 0.22232589
## 59 February - September -2.81349590 0.0049006011 0.26953306
## 60 January - September -3.43972564 0.0005823042 0.03726747
## 61 July - September -1.86053761 0.0628095020 1.00000000
## 62 June - September -0.35395594 0.7233719059 1.00000000
## 63 March - September -0.17244007 0.8630915715 1.00000000
## 64 May - September 0.06353055 0.9493440309 0.94934403
## 65 November - September -0.10890952 0.9132742566 1.00000000
## 66 October - September -1.38859637 0.1649555232 1.00000000require(PMCMR)
posthoc.kruskal.dunn.test(DAs ~ mes, data=Timeline_5anos, p.adjust.method="bonferroni") ## Warning in posthoc.kruskal.dunn.test.default(c(19L, 19L, 30L, 29L, 35L, :
## Ties are present. z-quantiles were corrected for ties.##
## Pairwise comparisons using Dunn's-test for multiple
## comparisons of independent samples
##
## data: DAs by mes
##
## April August December February January July June March May
## August 1.000 - - - - - - - -
## December 0.126 1.000 - - - - - - -
## February 0.161 1.000 1.000 - - - - - -
## January 0.017 1.000 1.000 1.000 - - - - -
## July 1.000 1.000 1.000 1.000 1.000 - - - -
## June 1.000 1.000 0.748 0.918 0.134 1.000 - - -
## March 1.000 1.000 0.439 0.545 0.072 1.000 1.000 - -
## May 1.000 1.000 0.210 0.265 0.030 1.000 1.000 1.000 -
## November 1.000 1.000 0.362 0.451 0.057 1.000 1.000 1.000 1.000
## October 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
## September 1.000 1.000 0.257 0.323 0.038 1.000 1.000 1.000 1.000
## November October
## August - -
## December - -
## February - -
## January - -
## July - -
## June - -
## March - -
## May - -
## November - -
## October 1.000 -
## September 1.000 1.000
##
## P value adjustment method: bonferroniposthoc.kruskal.dunn.test(DAs ~ mes, data=Timeline_5anos, p.adjust.method="none")## Warning in posthoc.kruskal.dunn.test.default(c(19L, 19L, 30L, 29L, 35L, :
## Ties are present. z-quantiles were corrected for ties.##
## Pairwise comparisons using Dunn's-test for multiple
## comparisons of independent samples
##
## data: DAs by mes
##
## April August December February January July June
## August 0.19124 - - - - - -
## December 0.00191 0.07233 - - - - -
## February 0.00243 0.08464 0.94212 - - - -
## January 0.00025 0.01874 0.57984 0.53116 - - -
## July 0.03768 0.44044 0.30510 0.34061 0.11429 - -
## June 0.56747 0.46225 0.01134 0.01391 0.00203 0.13192 -
## March 0.69634 0.35932 0.00665 0.00826 0.00109 0.09139 0.85596
## May 0.87738 0.24906 0.00318 0.00401 0.00046 0.05435 0.67632
## November 0.74387 0.32699 0.00548 0.00684 0.00087 0.07984 0.80642
## October 0.10818 0.76456 0.13426 0.15419 0.04025 0.63697 0.30084
## September 0.82757 0.27611 0.00390 0.00490 0.00058 0.06281 0.72337
## March May November October
## August - - - -
## December - - - -
## February - - - -
## January - - - -
## July - - - -
## June - - - -
## March - - - -
## May 0.81346 - - -
## November 0.94934 0.86309 - -
## October 0.22393 0.14647 0.20066 -
## September 0.86309 0.94934 0.91327 0.16496
##
## P value adjustment method: none#Teste de normalidade
library(stats)
#Teste de Bartlett com a hipótese nula de que as variâncias dos grupos são iguais (p > 0,05).
bartlett.test(DAs~ordem, Timeline)##
## Bartlett test of homogeneity of variances
##
## data: DAs by ordem
## Bartlett's K-squared = 4.1341, df = 11, p-value = 0.9658library(car)
#Teste da homogeneidade das variancias (a homogeneidade sera atingida com p > 0,05, hipótese de igualdade de variâncias)
leveneTest(DAs~ordem, Timeline)## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 11 0.3663 0.963
## 48##Dislexia
### Specify the order of factor levels
library(dplyr)
Timeline = mutate(Timeline_5anos,
mes = factor(mes, levels=unique(mes)))
# Medians and descriptive statistics
library(FSA)
Summarize(Dislexia ~ mes,data = Timeline_5anos)## mes n mean sd min Q1 median Q3 max
## 1 April 5 25.6 5.856620 20 22 24 27 35
## 2 August 5 20.8 2.588436 18 19 20 23 24
## 3 December 5 16.0 2.000000 14 15 15 17 19
## 4 February 5 19.0 1.870829 17 17 20 20 21
## 5 January 5 16.6 3.781534 13 14 15 19 22
## 6 July 5 19.2 2.774887 16 17 19 22 22
## 7 June 5 22.2 3.962323 17 21 22 23 28
## 8 March 5 23.2 3.271085 19 22 22 26 27
## 9 May 5 23.0 1.224745 21 23 23 24 24
## 10 November 5 22.6 3.781534 19 20 21 25 28
## 11 October 5 21.2 2.588436 18 20 21 22 25
## 12 September 5 24.0 5.338539 18 21 22 28 31# Histograms for each group
library(lattice)
histogram(~ Dislexia | ordem, data=Timeline_5anos, layout=c(12,1))
#Analise de variancia para dados nao-parametricos (p-value menos de 0,05, deve-se rejeitar a hipotese nula).Kruskal-Wallis test be interpreted as a difference in medians. applied to one-way data with more than two groups
kruskal.test(Dislexia ~ mes, data = Timeline_5anos)##
## Kruskal-Wallis rank sum test
##
## data: Dislexia by mes
## Kruskal-Wallis chi-squared = 26.593, df = 11, p-value = 0.005292#TESTES POS HOC
#Nemenyi test for multiple comparisons, not appropriate for groups with unequal numbers of observations. Utilizando qui quadrado.
library(DescTools)
NemenyiTest(x = Timeline_5anos$Dislexia, g = Timeline_5anos$mes, dist="chisq")##
## Nemenyi's test of multiple comparisons for independent samples (chisq)
##
## mean.rank.diff pval
## August-April -15.1 0.9989
## December-April -36.9 0.4371
## February-April -25.2 0.9229
## January-April -32.1 0.6789
## July-April -22.8 0.9628
## June-April -9.1 1.0000
## March-April -4.7 1.0000
## May-April -1.8 1.0000
## November-April -8.7 1.0000
## October-April -13.4 0.9997
## September-April -5.4 1.0000
## December-August -21.8 0.9737
## February-August -10.1 1.0000
## January-August -17.0 0.9968
## July-August -7.7 1.0000
## June-August 6.0 1.0000
## March-August 10.4 1.0000
## May-August 13.3 0.9997
## November-August 6.4 1.0000
## October-August 1.7 1.0000
## September-August 9.7 1.0000
## February-December 11.7 0.9999
## January-December 4.8 1.0000
## July-December 14.1 0.9995
## June-December 27.8 0.8537
## March-December 32.2 0.6741
## May-December 35.1 0.5286
## November-December 28.2 0.8404
## October-December 23.5 0.9533
## September-December 31.5 0.7070
## January-February -6.9 1.0000
## July-February 2.4 1.0000
## June-February 16.1 0.9981
## March-February 20.5 0.9840
## May-February 23.4 0.9547
## November-February 16.5 0.9976
## October-February 11.8 0.9999
## September-February 19.8 0.9880
## July-January 9.3 1.0000
## June-January 23.0 0.9602
## March-January 27.4 0.8662
## May-January 30.3 0.7601
## November-January 23.4 0.9547
## October-January 18.7 0.9926
## September-January 26.7 0.8865
## June-July 13.7 0.9996
## March-July 18.1 0.9945
## May-July 21.0 0.9805
## November-July 14.1 0.9995
## October-July 9.4 1.0000
## September-July 17.4 0.9961
## March-June 4.4 1.0000
## May-June 7.3 1.0000
## November-June 0.4 1.0000
## October-June -4.3 1.0000
## September-June 3.7 1.0000
## May-March 2.9 1.0000
## November-March -4.0 1.0000
## October-March -8.7 1.0000
## September-March -0.7 1.0000
## November-May -6.9 1.0000
## October-May -11.6 0.9999
## September-May -3.6 1.0000
## October-November -4.7 1.0000
## September-November 3.3 1.0000
## September-October 8.0 1.0000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##Dunn test for multiple comparisons; If the Kruskal-Wallis test is significant, a post-hoc analysis can be performed to determine which levels of the independent variable differ from each other level. Dunn's Kruskal-Wallis Multiple Comparisons
library(FSA)
dunnTest(Dislexia ~ mes, data=Timeline_5anos) ## Dunn (1964) Kruskal-Wallis multiple comparison## p-values adjusted with the Holm method.## Comparison Z P.unadj P.adj
## 1 April - August 1.372439 0.1699268364 1.00000000
## 2 April - December 3.353841 0.0007969819 0.05260081
## 3 August - December 1.981402 0.0475462202 1.00000000
## 4 April - February 2.290428 0.0219965242 1.00000000
## 5 August - February 0.917989 0.3586246475 1.00000000
## 6 December - February -1.063413 0.2875947227 1.00000000
## 7 April - January 2.917569 0.0035277174 0.22224620
## 8 August - January 1.545130 0.1223148395 1.00000000
## 9 December - January -0.436272 0.6626394094 1.00000000
## 10 February - January 0.627141 0.5305668331 1.00000000
## 11 April - July 2.072292 0.0382382272 1.00000000
## 12 August - July 0.699853 0.4840191325 1.00000000
## 13 December - July -1.281549 0.2000009218 1.00000000
## 14 February - July -0.218136 0.8273231568 1.00000000
## 15 January - July -0.845277 0.3979562218 1.00000000
## 16 April - June 0.827099 0.4081809731 1.00000000
## 17 August - June -0.545340 0.5855197159 1.00000000
## 18 December - June -2.526742 0.0115126081 0.67924388
## 19 February - June -1.463329 0.1433773950 1.00000000
## 20 January - June -2.090470 0.0365756094 1.00000000
## 21 July - June -1.245193 0.2130608367 1.00000000
## 22 April - March 0.427183 0.6692460572 1.00000000
## 23 August - March -0.945256 0.3445282187 1.00000000
## 24 December - March -2.926658 0.0034262557 0.21928037
## 25 February - March -1.863245 0.0624278167 1.00000000
## 26 January - March -2.490386 0.0127604467 0.74010591
## 27 July - March -1.645109 0.0999473510 1.00000000
## 28 June - March -0.399916 0.6892184011 1.00000000
## 29 April - May 0.163602 0.8700444762 1.00000000
## 30 August - May -1.208837 0.2267254956 1.00000000
## 31 December - May -3.190239 0.0014215525 0.09240091
## 32 February - May -2.126826 0.0334345515 1.00000000
## 33 January - May -2.753967 0.0058877728 0.35915414
## 34 July - May -1.908690 0.0563021056 1.00000000
## 35 June - May -0.663497 0.5070123203 1.00000000
## 36 March - May -0.263581 0.7921028103 1.00000000
## 37 April - November 0.790743 0.4290939999 1.00000000
## 38 August - November -0.581696 0.5607714848 1.00000000
## 39 December - November -2.563098 0.0103742797 0.62245678
## 40 February - November -1.499685 0.1336960363 1.00000000
## 41 January - November -2.126826 0.0334345515 1.00000000
## 42 July - November -1.281549 0.2000009218 1.00000000
## 43 June - November -0.036356 0.9709984992 0.97099850
## 44 March - November 0.363560 0.7161866137 1.00000000
## 45 May - November 0.627141 0.5305668331 1.00000000
## 46 April - October 1.217926 0.2232521168 1.00000000
## 47 August - October -0.154513 0.8772052676 1.00000000
## 48 December - October -2.135915 0.0326863510 1.00000000
## 49 February - October -1.072502 0.2834946389 1.00000000
## 50 January - October -1.699643 0.0891981121 1.00000000
## 51 July - October -0.854366 0.3929022508 1.00000000
## 52 June - October 0.390827 0.6959251297 1.00000000
## 53 March - October 0.790743 0.4290939999 1.00000000
## 54 May - October 1.054324 0.2917346272 1.00000000
## 55 November - October 0.427183 0.6692460572 1.00000000
## 56 April - September 0.490806 0.6235636824 1.00000000
## 57 August - September -0.881633 0.3779753261 1.00000000
## 58 December - September -2.863035 0.0041960439 0.26015472
## 59 February - September -1.799622 0.0719203583 1.00000000
## 60 January - September -2.426763 0.0152342089 0.86834991
## 61 July - September -1.581486 0.1137669746 1.00000000
## 62 June - September -0.336293 0.7366499400 1.00000000
## 63 March - September 0.063623 0.9492704199 1.00000000
## 64 May - September 0.327204 0.7435136074 1.00000000
## 65 November - September -0.299937 0.7642252218 1.00000000
## 66 October - September -0.727120 0.4671524689 1.00000000require(PMCMR)
posthoc.kruskal.nemenyi.test(x = Timeline_5anos$Dislexia, g = Timeline_5anos$mes, dist="Chisquare")## Warning in posthoc.kruskal.nemenyi.test.default(x = Timeline_5anos
## $Dislexia, : Ties are present. Chi-sq was corrected for ties.##
## Pairwise comparisons using Nemenyi-test with Chi-squared
## approximation for independent samples
##
## data: Timeline_5anos$Dislexia and Timeline_5anos$mes
##
## April August December February January July June March May
## August 1.00 - - - - - - - -
## December 0.42 0.97 - - - - - - -
## February 0.92 1.00 1.00 - - - - - -
## January 0.67 1.00 1.00 1.00 - - - - -
## July 0.96 1.00 1.00 1.00 1.00 - - - -
## June 1.00 1.00 0.85 1.00 0.96 1.00 - - -
## March 1.00 1.00 0.66 0.98 0.86 0.99 1.00 - -
## May 1.00 1.00 0.51 0.95 0.75 0.98 1.00 1.00 -
## November 1.00 1.00 0.83 1.00 0.95 1.00 1.00 1.00 1.00
## October 1.00 1.00 0.95 1.00 0.99 1.00 1.00 1.00 1.00
## September 1.00 1.00 0.70 0.99 0.88 1.00 1.00 1.00 1.00
## November October
## August - -
## December - -
## February - -
## January - -
## July - -
## June - -
## March - -
## May - -
## November - -
## October 1.00 -
## September 1.00 1.00
##
## P value adjustment method: noneposthoc.kruskal.dunn.test(Dislexia ~ mes, data=Timeline_5anos, p.adjust.method="bonferroni") ## Warning in posthoc.kruskal.dunn.test.default(c(22L, 20L, 27L, 27L, 24L, :
## Ties are present. z-quantiles were corrected for ties.##
## Pairwise comparisons using Dunn's-test for multiple
## comparisons of independent samples
##
## data: Dislexia by mes
##
## April August December February January July June March May
## August 1.000 - - - - - - - -
## December 0.053 1.000 - - - - - - -
## February 1.000 1.000 1.000 - - - - - -
## January 0.233 1.000 1.000 1.000 - - - - -
## July 1.000 1.000 1.000 1.000 1.000 - - - -
## June 1.000 1.000 0.760 1.000 1.000 1.000 - - -
## March 1.000 1.000 0.226 1.000 0.842 1.000 1.000 - -
## May 1.000 1.000 0.094 1.000 0.389 1.000 1.000 1.000 -
## November 1.000 1.000 0.685 1.000 1.000 1.000 1.000 1.000 1.000
## October 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
## September 1.000 1.000 0.277 1.000 1.000 1.000 1.000 1.000 1.000
## November October
## August - -
## December - -
## February - -
## January - -
## July - -
## June - -
## March - -
## May - -
## November - -
## October 1.000 -
## September 1.000 1.000
##
## P value adjustment method: bonferroniposthoc.kruskal.dunn.test(Dislexia ~ mes, data=Timeline_5anos, p.adjust.method="none")## Warning in posthoc.kruskal.dunn.test.default(c(22L, 20L, 27L, 27L, 24L, :
## Ties are present. z-quantiles were corrected for ties.##
## Pairwise comparisons using Dunn's-test for multiple
## comparisons of independent samples
##
## data: Dislexia by mes
##
## April August December February January July June March
## August 0.1699 - - - - - - -
## December 0.0008 0.0475 - - - - - -
## February 0.0220 0.3586 0.2876 - - - - -
## January 0.0035 0.1223 0.6626 0.5306 - - - -
## July 0.0382 0.4840 0.2000 0.8273 0.3980 - - -
## June 0.4082 0.5855 0.0115 0.1434 0.0366 0.2131 - -
## March 0.6692 0.3445 0.0034 0.0624 0.0128 0.0999 0.6892 -
## May 0.8700 0.2267 0.0014 0.0334 0.0059 0.0563 0.5070 0.7921
## November 0.4291 0.5608 0.0104 0.1337 0.0334 0.2000 0.9710 0.7162
## October 0.2233 0.8772 0.0327 0.2835 0.0892 0.3929 0.6959 0.4291
## September 0.6236 0.3780 0.0042 0.0719 0.0152 0.1138 0.7366 0.9493
## May November October
## August - - -
## December - - -
## February - - -
## January - - -
## July - - -
## June - - -
## March - - -
## May - - -
## November 0.5306 - -
## October 0.2917 0.6692 -
## September 0.7435 0.7642 0.4672
##
## P value adjustment method: none#carregar pacote para rodar a funcao do teste para homogeneidade das variancias
#Caso nao tenho instalado: install.packages("car")
library(stats)
#Teste de Bartlett com a hipótese nula de que as variâncias dos grupos são
#iguais.
bartlett.test(Dislexia~ordem, Timeline)##
## Bartlett test of homogeneity of variances
##
## data: Dislexia by ordem
## Bartlett's K-squared = 14.382, df = 11, p-value = 0.2126library(car)
#Teste da homogeneidade das variancias (a homogeneidade sera atingida com valores de p acima de 0,05)
leveneTest(Dislexia~ordem, Timeline)## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 11 0.8041 0.6355
## 48Descricao por estado
GeoMap <-read.csv("C:/Users/Karina/Desktop/UFMG/Mestrado/2015 - 2016/Projetos/Projeto 2016/Instrumentos e Dados/Dados/GeoMap.csv", sep=",", dec=",",fill=TRUE, header=TRUE)
str(GeoMap)## 'data.frame': 27 obs. of 21 variables:
## $ X : int 1 2 3 4 5 6 7 8 9 10 ...
## $ Região : Factor w/ 27 levels "Acre","Alagoas",..: 1 2 3 4 5 6 7 8 9 10 ...
## $ DA_Shop : int NA 0 0 10 3 6 7 0 2 0 ...
## $ Dislexia_Shop: int NA 13 100 10 2 6 7 0 4 11 ...
## $ Cartilha_Shop: int NA 20 99 20 14 12 29 6 5 17 ...
## $ Cartilha : int 61 46 91 50 46 38 50 37 43 52 ...
## $ DA : int NA 77 NA 58 70 64 40 49 49 100 ...
## $ Dislexia : int 59 47 100 54 49 51 46 48 49 55 ...
## $ Pais : int 78 81 98 77 81 77 74 74 70 82 ...
## $ Pais1 : int 73 74 90 70 76 71 69 68 65 74 ...
## $ Fam1 : int 26 29 44 29 27 27 27 23 22 30 ...
## $ EF1 : int 24 28 32 18 29 23 17 27 19 30 ...
## $ Crianca1 : int 83 82 100 72 80 80 85 74 73 81 ...
## $ TA1 : int 0 0 0 0 0 0 0 0 0 0 ...
## $ DA1 : int 0 23 0 15 21 20 11 14 14 28 ...
## $ Dislexia1 : int 61 43 100 55 50 49 44 47 49 54 ...
## $ Disgrafia1 : int 0 0 0 0 4 4 4 0 4 0 ...
## $ Discalculia1 : int 0 4 0 4 4 4 3 4 3 6 ...
## $ Cartilha_sug : int 80 52 95 58 52 44 57 43 49 59 ...
## $ DA_sug : int NA 83 NA 53 75 71 41 51 51 100 ...
## $ Pais_sug : int 78 79 96 75 81 76 73 73 69 80 ...tail(GeoMap)## X Região DA_Shop Dislexia_Shop Cartilha_Shop Cartilha DA
## 22 22 Rond<U+0093>nia 0 0 54 52 NA
## 23 23 Roraima NA NA NA 100 NA
## 24 24 Santa Catarina 3 0 11 32 43
## 25 25 SÆo Paulo 2 5 20 32 31
## 26 26 Sergipe 0 35 18 46 NA
## 27 27 Tocantins 0 30 30 65 NA
## Dislexia Pais Pais1 Fam1 EF1 Crianca1 TA1 DA1 Dislexia1 Disgrafia1
## 22 55 77 71 21 24 77 0 0 55 0
## 23 77 100 94 34 33 95 0 0 79 0
## 24 48 77 71 19 27 63 0 13 48 3
## 25 46 71 64 18 19 65 1 9 45 2
## 26 46 78 72 27 22 77 0 0 42 0
## 27 41 81 72 24 34 81 0 0 41 0
## Discalculia1 Cartilha_sug DA_sug Pais_sug
## 22 0 63 NA 76
## 23 0 100 NA 100
## 24 3 36 48 75
## 25 3 37 33 69
## 26 0 54 NA 77
## 27 0 75 NA 77summary(GeoMap)## X Região DA_Shop Dislexia_Shop
## Min. : 1.0 Acre : 1 Min. : 0.00 Min. : 0.00
## 1st Qu.: 7.5 Alagoas : 1 1st Qu.: 0.00 1st Qu.: 4.00
## Median :14.0 Amap : 1 Median : 2.00 Median : 7.00
## Mean :14.0 Amazonas: 1 Mean : 2.56 Mean : 12.32
## 3rd Qu.:20.5 Bahia : 1 3rd Qu.: 3.00 3rd Qu.: 11.00
## Max. :27.0 Cear : 1 Max. :10.00 Max. :100.00
## (Other) :21 NA's :2 NA's :2
## Cartilha_Shop Cartilha DA Dislexia
## Min. : 5.00 Min. : 29.00 Min. : 28.00 Min. : 41.00
## 1st Qu.: 9.00 1st Qu.: 37.50 1st Qu.: 42.50 1st Qu.: 46.00
## Median :17.00 Median : 46.00 Median : 55.00 Median : 49.00
## Mean :20.84 Mean : 48.63 Mean : 56.53 Mean : 52.11
## 3rd Qu.:20.00 3rd Qu.: 52.50 3rd Qu.: 68.00 3rd Qu.: 54.00
## Max. :99.00 Max. :100.00 Max. :100.00 Max. :100.00
## NA's :2 NA's :8
## Pais Pais1 Fam1 EF1
## Min. : 67.00 Min. :62.00 Min. :18.00 Min. :17.00
## 1st Qu.: 72.00 1st Qu.:67.00 1st Qu.:21.50 1st Qu.:20.50
## Median : 75.00 Median :69.00 Median :26.00 Median :25.00
## Mean : 76.89 Mean :70.74 Mean :25.22 Mean :25.04
## 3rd Qu.: 78.00 3rd Qu.:72.00 3rd Qu.:27.00 3rd Qu.:29.00
## Max. :100.00 Max. :94.00 Max. :44.00 Max. :34.00
##
## Crianca1 TA1 DA1 Dislexia1
## Min. : 60.00 Min. :0.00000 Min. : 0.00 Min. : 41.00
## 1st Qu.: 71.50 1st Qu.:0.00000 1st Qu.: 0.00 1st Qu.: 46.00
## Median : 77.00 Median :0.00000 Median :13.00 Median : 49.00
## Mean : 76.11 Mean :0.03704 Mean :11.48 Mean : 52.07
## 3rd Qu.: 80.50 3rd Qu.:0.00000 3rd Qu.:17.50 3rd Qu.: 54.50
## Max. :100.00 Max. :1.00000 Max. :28.00 Max. :100.00
##
## Disgrafia1 Discalculia1 Cartilha_sug DA_sug
## Min. :0.000 Min. :0.000 Min. : 32.00 Min. : 30.00
## 1st Qu.:0.000 1st Qu.:1.500 1st Qu.: 43.50 1st Qu.: 46.00
## Median :0.000 Median :4.000 Median : 52.00 Median : 53.00
## Mean :1.593 Mean :2.926 Mean : 55.33 Mean : 59.05
## 3rd Qu.:3.500 3rd Qu.:4.000 3rd Qu.: 63.00 3rd Qu.: 73.00
## Max. :5.000 Max. :6.000 Max. :100.00 Max. :100.00
## NA's :8
## Pais_sug
## Min. : 67.00
## 1st Qu.: 71.50
## Median : 74.00
## Mean : 75.52
## 3rd Qu.: 77.00
## Max. :100.00
## Relacao Palavras x Regiao
GRafico de frequencia Comparativa - 2004 a 2017
plot(GeoMap$Dislexia1~GeoMap$Região , type = "l", lty=1, lwd=2, col="red", ylim=c(0,100), xlab="região", ylab="Frequencia relativa", main = "Frequencia relativa por regiao geografica, 2004 a 2017")
lines(GeoMap$Dislexia1~GeoMap$Região, type = "l", lty=1, lwd=2,col= "red")
lines(GeoMap$DA1~GeoMap$Região, type = "l", lty=1, lwd=2,col= "darkblue")
lines(GeoMap$Discalculia1~GeoMap$Região, type = "l", lty=1, lwd=2,col= "green")
lines(GeoMap$Disgrafia1~GeoMap$Região, type = "l", lty=1, lwd=2,col= "orange")
lines(GeoMap$TA1~GeoMap$Região, type = "l", lty=1, lwd=2,col= "purple")
legend("topright", c("Dislexia", "Dif. de Aprendizagem", "Discalculia", "Disgrafia", "Transt. da Aprendizagem"), bty="n", lty=c(1),
col=c("red","darkblue", "green", "orange", "purple"), box.col="white",lwd=c(2,2), cex=0.9)