ICC_Congreso

Cargo los paquetes (IRR, dentro del pack ICC)

library("irr")
library("tidyverse")
library("ggsignif")

Abro datos de mi df2

ICC <- IC <- read.csv("datos2.csv")

summary

summary(ICC)

 A..os.de.ejercicio.profesional Sexo      Nivel.de.Estudios Promedio.implantes.mensuales
 Min.   : 2.00                  F: 5   Curso       : 1      Min.   : 1.000              
 1st Qu.: 7.00                  M:27   Diplomado   :12      1st Qu.: 4.000              
 Median :10.50                         Especialidad:19      Median : 6.000              
 Mean   :11.16                                              Mean   : 7.156              
 3rd Qu.:14.00                                              3rd Qu.: 8.000              
 Max.   :24.00                                              Max.   :30.000              
      BP1             BP2             MD1              MD2              H1              H2       
 Min.   :3.500   Min.   :4.150   Min.   : 8.280   Min.   : 7.81   Min.   :11.06   Min.   :11.14  
 1st Qu.:4.378   1st Qu.:4.388   1st Qu.: 9.848   1st Qu.:10.04   1st Qu.:13.96   1st Qu.:13.54  
 Median :4.565   Median :4.540   Median :10.955   Median :11.13   Median :14.53   Median :14.52  
 Mean   :4.578   Mean   :4.575   Mean   :10.959   Mean   :11.09   Mean   :14.55   Mean   :14.35  
 3rd Qu.:4.827   3rd Qu.:4.670   3rd Qu.:12.098   3rd Qu.:12.24   3rd Qu.:15.22   3rd Qu.:15.07  
 Max.   :5.650   Max.   :5.600   Max.   :13.860   Max.   :13.78   Max.   :17.25   Max.   :17.31  

Seleciono columnas para mediciones Buco-palatino

long_BP <- ICC %>% 
  select(BP1, BP2)

Calculo ICC para BP

icc(long_BP, model = "twoway", type = "agreement")

 Single Score Intraclass Correlation

   Model: twoway 
   Type : agreement 

   Subjects = 32 
     Raters = 2 
   ICC(A,1) = 0.713

 F-Test, H0: r0 = 0 ; H1: r0 > 0 
   F(31,31) = 5.82 , p = 2.15e-06 

 95%-Confidence Interval for ICC Population Values:
  0.487 < ICC < 0.849

Seleciono columnas para mediciones mesio-distal

long_MD <- ICC %>% 
  select(MD1, MD2)

Calculo ICC para MD

icc(long_MD, model = "twoway", type = "agreement")

 Single Score Intraclass Correlation

   Model: twoway 
   Type : agreement 

   Subjects = 32 
     Raters = 2 
   ICC(A,1) = 0.798

 F-Test, H0: r0 = 0 ; H1: r0 > 0 
 F(31,31.8) = 8.79 , p = 1.13e-08 

 95%-Confidence Interval for ICC Population Values:
  0.627 < ICC < 0.896

Selecciono columnas para mediciones de altura (h)

long_h <- ICC %>% 
  select(H1, H2)

Calculo ICC para h

icc(long_h, model = "twoway", type = "agreement")

 Single Score Intraclass Correlation

   Model: twoway 
   Type : agreement 

   Subjects = 32 
     Raters = 2 
   ICC(A,1) = 0.623

 F-Test, H0: r0 = 0 ; H1: r0 > 0 
 F(31,31.9) = 4.29 , p = 4.65e-05 

 95%-Confidence Interval for ICC Population Values:
  0.358 < ICC < 0.796

Reformateo el df para obtener el ICC general

long_general1 <- ICC %>% 
  select(BP1,MD1,H1)%>% 
  gather("Grupo1","Valor1")

long_general2 <- ICC %>% 
  select(BP2,MD2,H2)%>% 
  gather("Grupo2","Valor2")

General3 <- bind_cols(long_general1, long_general2)

general4 <- General3 %>% 
  select(Valor1,Valor2)

Calculo ICC genral

icc(general4, model = "twoway", type = "agreement")

 Single Score Intraclass Correlation

   Model: twoway 
   Type : agreement 

   Subjects = 96 
     Raters = 2 
   ICC(A,1) = 0.979

 F-Test, H0: r0 = 0 ; H1: r0 > 0 
 F(95,95.1) = 94.2 , p = 8.36e-68 

 95%-Confidence Interval for ICC Population Values:
  0.969 < ICC < 0.986

Promedios y DS, agrupados por nivel de estudio en cada una de las mediciones

Promedio y sd para mediciones

Le doy estrucuta a mi df para graficar

General3 %>% 
  select(BP1,MD1,H1,BP2,MD2,H2) %>% 
  gather("Grupo","Valor")

Error in overscope_eval_next(overscope, expr) : 
  objeto 'BP1' no encontrado

Grafico

ANOVA para mediciones

summary(aov)

                    Df Sum Sq Mean Sq F value Pr(>F)    
long_grafico$Grupo   5   3217   643.5     480 <2e-16 ***
Residuals          186    249     1.3                   
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

TukeyHSD(aov)

  Tukey multiple comparisons of means
    95% family-wise confidence level

Fit: aov(formula = long_grafico$Valor ~ long_grafico$Grupo)

$`long_grafico$Grupo`
              diff        lwr        upr     p adj
BP2-BP1 -0.0031250 -0.8366721  0.8304221 1.0000000
H1-BP1   9.9700000  9.1364529 10.8035471 0.0000000
H2-BP1   9.7734375  8.9398904 10.6069846 0.0000000
MD1-BP1  6.3806250  5.5470779  7.2141721 0.0000000
MD2-BP1  6.5065625  5.6730154  7.3401096 0.0000000
H1-BP2   9.9731250  9.1395779 10.8066721 0.0000000
H2-BP2   9.7765625  8.9430154 10.6101096 0.0000000
MD1-BP2  6.3837500  5.5502029  7.2172971 0.0000000
MD2-BP2  6.5096875  5.6761404  7.3432346 0.0000000
H2-H1   -0.1965625 -1.0301096  0.6369846 0.9840688
MD1-H1  -3.5893750 -4.4229221 -2.7558279 0.0000000
MD2-H1  -3.4634375 -4.2969846 -2.6298904 0.0000000
MD1-H2  -3.3928125 -4.2263596 -2.5592654 0.0000000
MD2-H2  -3.2668750 -4.1004221 -2.4333279 0.0000000
MD2-MD1  0.1259375 -0.7076096  0.9594846 0.9979993