Search code examples
rggplot2fillfacet-grid

Fill pattern not expanded to all cells

I am filling the cells of a matrix with a color according to the value of the number in the cell.

Some example data:

plotData <- structure(list(Dimension = structure(c(1L, 2L, 1L, 2L, 1L, 2L, 
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 
1L, 2L, 1L, 2L, 1L, 2L), .Label = c("Dim 1", "Dim 2", "Dim 3", 
"Dim 4", "Dim 5", "Dim 6", "Dim 7"), class = "factor"), R1 = structure(c(3L, 
3L, 3L, 3L, 3L, 3L, 2L, 2L, 3L, 3L, 1L, 1L, 4L, 4L, 4L, 4L, 2L, 
2L, 4L, 4L, 1L, 1L, 5L, 5L, 2L, 2L, 5L, 5L, 1L, 1L, 2L, 2L, 6L, 
6L, 1L, 1L, 2L, 2L, 1L, 1L, 1L, 1L, 4L, 4L, 5L, 5L, 6L, 6L, 3L, 
3L, 7L, 7L, 3L, 3L, 5L, 5L, 6L, 6L, 4L, 4L, 7L, 7L, 4L, 4L, 6L, 
6L, 5L, 5L, 7L, 7L, 5L, 5L, 6L, 6L, 7L, 7L, 6L, 6L, 7L, 7L, 2L, 
2L, 7L, 7L, 3L, 3L, 2L, 2L, 1L, 1L, 4L, 4L, 5L, 5L, 6L, 6L, 3L, 
3L, 2L, 2L, 1L, 1L, 4L, 4L, 5L, 5L, 6L, 6L, 3L, 3L, 2L, 2L, 1L, 
1L, 4L, 4L, 5L, 5L, 6L, 6L, 3L, 3L, 2L, 2L, 1L, 1L, 4L, 4L, 5L, 
5L, 6L, 6L, 3L, 3L, 2L, 2L, 1L, 1L, 4L, 4L, 5L, 5L, 6L, 6L, 3L, 
3L, 2L, 2L, 1L, 1L, 4L, 4L, 5L, 5L, 6L, 6L), .Label = c("Rater 1", 
"Rater 2", "Rater 3", "Rater 4", "Rater 5", "Rater 6", "Rater 7"
), class = "factor"), R2 = structure(c(4L, 4L, 5L, 5L, 6L, 6L, 
3L, 3L, 7L, 7L, 3L, 3L, 5L, 5L, 6L, 6L, 4L, 4L, 7L, 7L, 4L, 4L, 
6L, 6L, 5L, 5L, 7L, 7L, 5L, 5L, 6L, 6L, 7L, 7L, 6L, 6L, 7L, 7L, 
2L, 2L, 7L, 7L, 3L, 3L, 3L, 3L, 3L, 3L, 2L, 2L, 3L, 3L, 1L, 1L, 
4L, 4L, 4L, 4L, 2L, 2L, 4L, 4L, 1L, 1L, 5L, 5L, 2L, 2L, 5L, 5L, 
1L, 1L, 2L, 2L, 6L, 6L, 1L, 1L, 2L, 2L, 1L, 1L, 1L, 1L, 3L, 3L, 
3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 5L, 5L, 
5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 6L, 6L, 6L, 6L, 6L, 6L, 
6L, 6L, 6L, 6L, 6L, 6L), .Label = c("Rater 1", "Rater 2", "Rater 3", 
"Rater 4", "Rater 5", "Rater 6", "Rater 7"), class = "factor"), 
    kappa = c(0.607540983606557, 0.437185929648241, 0.731739272846381, 
    0.534769230769231, 0.498230088495575, 0.503184713375796, 
    0.189421015010722, 0.24248496993988, 0.196913787319691, 0.188524590163934, 
    0.71434460016488, 0.583883751651255, 0.676277476448758, 0.597444089456869, 
    0.630104083266613, 0.390125847047435, 0.265364008824456, 
    0.269453768690876, 0.12396449704142, 0.378727400798309, 0.567901234567901, 
    0.591351351351351, 0.630104083266613, 0.471698113207547, 
    0.185430463576159, 0.094765651727677, 0.348789131718842, 
    0.167259786476868, 0.784320438206094, 0.616555082166768, 
    0.130634774609016, 0.207547169811321, 0.160714285714286, 
    0.135011441647597, 0.533908754623921, 0.505882352941176, 
    0.198526950117174, 0.199491740787802, 0.121372031662269, 
    0.446559297218155, 0.313291139240506, 0.488774682174736, 
    0.607540983606557, 0.437185929648241, 0.731739272846381, 
    0.534769230769231, 0.498230088495575, 0.503184713375796, 
    0.189421015010722, 0.24248496993988, 0.196913787319691, 0.188524590163934, 
    0.71434460016488, 0.583883751651255, 0.676277476448758, 0.597444089456869, 
    0.630104083266613, 0.390125847047435, 0.265364008824456, 
    0.269453768690876, 0.12396449704142, 0.378727400798309, 0.567901234567901, 
    0.591351351351351, 0.630104083266613, 0.471698113207547, 
    0.185430463576159, 0.094765651727677, 0.348789131718842, 
    0.167259786476868, 0.784320438206094, 0.616555082166768, 
    0.130634774609016, 0.207547169811321, 0.160714285714286, 
    0.135011441647597, 0.533908754623921, 0.505882352941176, 
    0.198526950117174, 0.199491740787802, 0.121372031662269, 
    0.446559297218155, 0.313291139240506, 0.488774682174736, 
    NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
    NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
    NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
    NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
    NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA)), row.names = c(1L, 
2L, 6L, 7L, 11L, 12L, 16L, 17L, 21L, 22L, 26L, 27L, 31L, 32L, 
36L, 37L, 41L, 42L, 46L, 47L, 51L, 52L, 56L, 57L, 61L, 62L, 66L, 
67L, 71L, 72L, 76L, 77L, 81L, 82L, 86L, 87L, 91L, 92L, 96L, 97L, 
101L, 102L, 106L, 107L, 111L, 112L, 116L, 117L, 121L, 122L, 126L, 
127L, 131L, 132L, 136L, 137L, 141L, 142L, 146L, 147L, 151L, 152L, 
156L, 157L, 161L, 162L, 166L, 167L, 171L, 172L, 176L, 177L, 181L, 
182L, 186L, 187L, 191L, 192L, 196L, 197L, 201L, 202L, 206L, 207L, 
211L, 212L, 216L, 217L, 221L, 222L, 226L, 227L, 231L, 232L, 236L, 
237L, 241L, 242L, 246L, 247L, 251L, 252L, 256L, 257L, 261L, 262L, 
266L, 267L, 271L, 272L, 276L, 277L, 281L, 282L, 286L, 287L, 291L, 
292L, 296L, 297L, 301L, 302L, 306L, 307L, 311L, 312L, 316L, 317L, 
321L, 322L, 326L, 327L, 331L, 332L, 336L, 337L, 341L, 342L, 346L, 
347L, 351L, 352L, 356L, 357L, 361L, 362L, 366L, 367L, 371L, 372L, 
376L, 377L, 381L, 382L, 386L, 387L), class = "data.frame")

Now, this code should (I think) fill the cells with a color according to the kappa-value in the cell:

library(ggplot2)
library(ggpubr)    
p <- ggplot(plotData,aes(x=R1,y=R2)) + 
      geom_tile(aes(fill = kappa))  +
      geom_text(aes(label = round(kappa, 2))) +
      labs(x="",y="",fill="Kappa") + 
      labs_pubr(base_size = 15) +
      facet_grid(Dimension ~ .)

plot(p)

However, what I get is this:

enter image description here

so not all cells are coloured. I can't see why this would not work actually, but I am probably doing something stupid.

Solution

  • It happens because of wrong data formating - every combination (R1 & R2 is given twice, while one of the entries has NA).

    Here we see that R1 == "Rater 1" & R2 == "Rater 2" combination is given twice with one of them having NA's.

    subset(plotData, R1 == "Rater 1" & R2 == "Rater 2")
   Dimension      R1      R2     kappa
1:     Dim 1 Rater 1 Rater 2 0.1213720
2:     Dim 2 Rater 1 Rater 2 0.4465593
3:     Dim 1 Rater 1 Rater 2        NA
4:     Dim 2 Rater 1 Rater 2        NA

    geom_text uses kappa numeric value, however fill takes NA and that's why you're getting grey background.

    To fix it use na.omit() on your data when plotting (you still won't be able to use NA values):

    ggplot(na.omit(plotData), aes(R1, R2)) + 
      geom_tile(aes(fill = kappa))  +
      geom_text(aes(label = round(kappa, 2))) +
      facet_grid(Dimension ~ .)

    enter image description here