Packages

Dataset

(just in case)

clean dataset

## # A tibble: 6 × 691
##      id first_name        x1_grado_0 x1_grado_1 x1_grado_2 x1_confianza_2
##   <int> <chr>                  <dbl>      <dbl>      <dbl>          <dbl>
## 1     1 FELIPE GARFIAS            NA         NA          2             NA
## 2     2 VIVIANA SEPULVEDA         NA         NA          2             NA
## 3     3 IGNACIO ASPE              NA         NA          2             NA
## 4     4 SOFIA MELLA               NA          1         NA             NA
## 5     5 BEATRIZ GOMEZ             NA         NA          2             NA
## 6     6 CLAUDIA VILDOSOLA         NA          1         NA             NA
## # … with 685 more variables: x1_confianza_3 <dbl>, x1_confianza_4 <dbl>,
## #   x1_confianza_5 <dbl>, x2_grado_1 <dbl>, x2_grado_2 <dbl>, x2_grado_3 <dbl>,
## #   x2_confianza_1 <dbl>, x2_confianza_2 <dbl>, x2_confianza_3 <dbl>,
## #   x2_confianza_4 <dbl>, x2_confianza_5 <dbl>, x3_grado_0 <dbl>,
## #   x3_grado_1 <dbl>, x3_grado_2 <dbl>, x3_grado_3 <dbl>, x3_confianza_1 <dbl>,
## #   x3_confianza_2 <dbl>, x3_confianza_3 <dbl>, x3_confianza_4 <dbl>,
## #   x3_confianza_5 <dbl>, x4_grado_0 <dbl>, x4_grado_1 <dbl>, …

Data cleaning

Merge and reformat dataset

Create a grado dataset

Create a confianza dataset

Join grado and confianza

## # A tibble: 6 × 5
##      id first_name     imagen grado_value confianza_value
##   <int> <chr>          <fct>        <dbl>           <dbl>
## 1     1 FELIPE GARFIAS 1                2               3
## 2     1 FELIPE GARFIAS 2                2               4
## 3     1 FELIPE GARFIAS 3                1               3
## 4     1 FELIPE GARFIAS 4                1               2
## 5     1 FELIPE GARFIAS 5                2               4
## 6     1 FELIPE GARFIAS 6                1               2

Delete unused datasets

Ojo, hay algunos que faltan

Change all names to lower

Change from dbl to int

df_merged <- df_merged %>% mutate(grade = as.integer(grade)) %>% mutate(confidence = as.integer(confidence))

Dataset ready for analysis

EDA

## Rows: 31,548
## Columns: 5
## $ id         <int> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,…
## $ name       <chr> "FELIPE GARFIAS", "FELIPE GARFIAS", "FELIPE GARFIAS", "FELI…
## $ image      <fct> 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, …
## $ grade      <dbl> 2, 2, 1, 1, 2, 1, 2, 1, 3, 1, 3, 1, 2, 2, 2, 2, 2, 2, 1, 2,…
## $ confidence <dbl> 3, 4, 3, 2, 4, 2, 5, 3, 5, 4, 3, 2, 4, 4, 5, 4, 4, 2, 4, 5,…

Check unique participants

## [1] 39

Check if any nombre has all the images

Check by participant

## # A tibble: 2 × 2
## # Groups:   name [2]
##   name               n
##   <chr>          <int>
## 1 IGNACIO GLARIA    76
## 2 ANDREA DIAZ       80

Delete two raters with less than 88 images

plot grade vs image

plot name vs grade

plot grade vs confidence

Agreement calculations

Reformat the dataset as: one column per examiner, filled with the grades per row, as follows

Agreement TOTAL

##  Single Score Intraclass Correlation
## 
##    Model: twoway 
##    Type : agreement 
## 
##    Subjects = 88 
##      Raters = 36 
##    ICC(A,1) = 0.566
## 
##  F-Test, H0: r0 = 0 ; H1: r0 > 0 
##  F(87,1614) = 53.3 , p = 0 
## 
##  95%-Confidence Interval for ICC Population Values:
##   0.493 < ICC < 0.645

Consistency TOTAL

##  Single Score Intraclass Correlation
## 
##    Model: twoway 
##    Type : consistency 
## 
##    Subjects = 88 
##      Raters = 36 
##    ICC(C,1) = 0.592
## 
##  F-Test, H0: r0 = 0 ; H1: r0 > 0 
##  F(87,3045) = 53.3 , p = 0 
## 
##  95%-Confidence Interval for ICC Population Values:
##   0.522 < ICC < 0.669

New dataset with identification to merge

Identifiy demographic data to compare ttm vs rx