shrink plot width to make more room for ggrepel labels

Question

I want to shrink the plot area so there is more room for ggrepel labels that currently get cut off. I can't seem to offset the labels any more via nudge_x(), and I do not want to shrink the text size.

I'm trying to find a way to compress the chart so that the groups all move closer to the center, leaving more room for labels at the extremes of the x-axis.

Specifically, I am trying to knit this figure into a portrait PDF. I tried controlling fig.width in the chunk options, but this just made the entire chart smaller.

I want to maximize the width on the portrait page, but shrink the plot area relative to the area for labels.

---
title             : "The title"
shorttitle        : "Title"

author: 
  - name          : "Me"
    affiliation   : "1"
    corresponding : yes    # Define only one corresponding author
    address       : "Address"
    email         : "email"

affiliation:
  - id            : "1"
    institution   : "Company"

authornote: |
  Note here

abstract: |
  Abstract here.


floatsintext      : yes
figurelist        : no
tablelist         : no
footnotelist      : no
linenumbers       : no
mask              : no
draft             : no
note              : "\\clearpage"

documentclass     : "apa6"
classoption       : "man,noextraspace"
header-includes:
  - \usepackage{pdfpages}
  - \usepackage{setspace}
  - \AtBeginEnvironment{tabular}{\singlespacing}
  - \makeatletter\let\expandableinput\@@input\makeatother
  - \interfootnotelinepenalty=10000
  - \usepackage{float} #use the 'float' package
  - \floatplacement{figure}{H} #make every figure with caption = h
  - \raggedbottom
output            : papaja::apa6_pdf
---


```{r test, fig.cap="Caption.", fig.height=8, include=TRUE, echo=FALSE}
library("papaja")
library(tidyverse)
library(ggrepel)

ageGenderF <- structure(list(genAge = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 
3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 
3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 
3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 
3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 
4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 
4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 
4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 
4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L), .Label = c("Women, 15-19", 
"Women, 20-24", "Women, 25-35", "Women, 36+"), class = "factor"), 
    word_ = c("this is label 2", "this is label 3", "this is label 4", 
    "this is label 1", "this is label 7", "this is label 5", 
    "this is label 8", "this is label 10", "this is label 11", 
    "this is label 20", "this is label 12", "this is label 6", 
    "this is label 17", "this is label 9", "this is label 15", 
    "this is label 21", "this is label 31", "this is label 25", 
    "this is label 26", "this is label 19", "this is label 24", 
    "this is label 28", "this is label 29", "this is label 30", 
    "this is label 14", "this is label 22", "this is label 18", 
    "this is label 54", "this is label 32", "this is label 44", 
    "this is label 52", "this is label 34", "this is label 59", 
    "this is label 48", "this is label 23", "this is label 47", 
    "this is label 38", "this is label 35", "this is label 61", 
    "this is label 56", "this is label 39", "this is label 72", 
    "this is label 42", "this is label 16", "this is label 66", 
    "this is label 37", "this is label 51", "this is label 27", 
    "this is label 40", "this is label 73", "this is label 60", 
    "this is label 113", "this is label 50", "this is label 45", 
    "this is label 81", "this is label 84", "this is label 53", 
    "this is label 49", "this is label 67", "this is label 68", 
    "this is label 46", "this is label 65", "this is label 41", 
    "this is label 57", "this is label 1", "this is label 2", 
    "this is label 3", "this is label 4", "this is label 5", 
    "this is label 6", "this is label 7", "this is label 8", 
    "this is label 9", "this is label 10", "this is label 11", 
    "this is label 12", "this is label 13", "this is label 14", 
    "this is label 15", "this is label 16", "this is label 17", 
    "this is label 18", "this is label 19", "this is label 20", 
    "this is label 21", "this is label 22", "this is label 23", 
    "this is label 24", "this is label 25", "this is label 26", 
    "this is label 27", "this is label 28", "this is label 29", 
    "this is label 30", "this is label 31", "this is label 32", 
    "this is label 33", "this is label 34", "this is label 35", 
    "this is label 36", "this is label 37", "this is label 38", 
    "this is label 39", "this is label 40", "this is label 41", 
    "this is label 42", "this is label 43", "this is label 44", 
    "this is label 45", "this is label 46", "this is label 47", 
    "this is label 48", "this is label 49", "this is label 50", 
    "this is label 51", "this is label 52", "this is label 53", 
    "this is label 54", "this is label 55", "this is label 56", 
    "this is label 57", "this is label 58", "this is label 59", 
    "this is label 60", "this is label 61", "this is label 62", 
    "this is label 63", "this is label 64", "this is label 1", 
    "this is label 2", "this is label 3", "this is label 6", 
    "this is label 4", "this is label 5", "this is label 12", 
    "this is label 7", "this is label 8", "this is label 9", 
    "this is label 10", "this is label 14", "this is label 11", 
    "this is label 18", "this is label 29", "this is label 45", 
    "this is label 27", "this is label 15", "this is label 26", 
    "this is label 71", "this is label 37", "this is label 13", 
    "this is label 25", "this is label 23", "this is label 22", 
    "this is label 41", "this is label 42", "this is label 55", 
    "this is label 52", "this is label 36", "this is label 34", 
    "this is label 17", "this is label 63", "this is label 24", 
    "this is label 19", "this is label 28", "this is label 38", 
    "this is label 32", "this is label 21", "this is label 30", 
    "this is label 35", "this is label 16", "this is label 64", 
    "this is label 20", "this is label 31", "this is label 53", 
    "this is label 77", "this is label 39", "this is label 70", 
    "this is label 57", "this is label 48", "this is label 43", 
    "this is label 132", "this is label 51", "this is label 66", 
    "this is label 58", "this is label 85", "this is label 120", 
    "this is label 65", "this is label 40", "this is label 121", 
    "this is label 78", "this is label 59", "this is label 141", 
    "this is label 1", "this is label 12", "this is label 6", 
    "this is label 2", "this is label 3", "this is label 5", 
    "this is label 4", "this is label 45", "this is label 52", 
    "this is label 26", "this is label 77", "this is label 8", 
    "this is label 7", "this is label 10", "this is label 14", 
    "this is label 31", "this is label 59", "this is label 178", 
    "this is label 18", "this is label 27", "this is label 42", 
    "this is label 70", "this is label 29", "this is label 37", 
    "this is label 330", "this is label 78", "this is label 25", 
    "this is label 34", "this is label 21", "this is label 450", 
    "this is label 83", "this is label 185", "this is label 57", 
    "this is label 16", "this is label 50", "this is label 126", 
    "this is label 895", "this is label 63", "this is label 402", 
    "this is label 19", "this is label 724", "this is label 40", 
    "this is label 11", "this is label 43", "this is label 758", 
    "this is label 1099", "this is label 73", "this is label 62", 
    "this is label 46", "this is label 183", "this is label 819", 
    "this is label 295", "this is label 1100", "this is label 17", 
    "this is label 282", "this is label 153", "this is label 1101", 
    "this is label 41", "this is label 1102", "this is label 446", 
    "this is label 216", "this is label 13", "this is label 109", 
    "this is label 20"), n = c(774L, 635L, 618L, 495L, 329L, 
    284L, 259L, 217L, 197L, 181L, 163L, 163L, 162L, 160L, 138L, 
    124L, 114L, 112L, 110L, 107L, 99L, 98L, 97L, 92L, 85L, 84L, 
    84L, 78L, 74L, 72L, 68L, 67L, 66L, 66L, 65L, 60L, 60L, 60L, 
    58L, 57L, 55L, 51L, 51L, 51L, 50L, 50L, 48L, 47L, 47L, 46L, 
    46L, 44L, 44L, 44L, 43L, 43L, 43L, 43L, 42L, 41L, 41L, 41L, 
    41L, 41L, 1568L, 1366L, 1220L, 1012L, 687L, 682L, 633L, 516L, 
    464L, 374L, 372L, 326L, 326L, 304L, 293L, 292L, 274L, 261L, 
    259L, 257L, 236L, 232L, 229L, 223L, 223L, 221L, 221L, 213L, 
    210L, 205L, 198L, 191L, 189L, 167L, 165L, 164L, 146L, 142L, 
    140L, 140L, 139L, 136L, 134L, 129L, 122L, 121L, 115L, 115L, 
    115L, 113L, 112L, 110L, 110L, 109L, 107L, 104L, 103L, 102L, 
    99L, 99L, 99L, 97L, 96L, 93L, 426L, 332L, 310L, 290L, 197L, 
    166L, 147L, 134L, 125L, 113L, 105L, 104L, 97L, 83L, 78L, 
    77L, 77L, 74L, 69L, 69L, 69L, 69L, 68L, 61L, 61L, 59L, 59L, 
    58L, 58L, 58L, 57L, 57L, 56L, 54L, 51L, 48L, 47L, 46L, 43L, 
    42L, 38L, 38L, 36L, 34L, 34L, 33L, 32L, 32L, 32L, 32L, 31L, 
    29L, 29L, 28L, 28L, 27L, 27L, 27L, 27L, 27L, 26L, 26L, 25L, 
    24L, 37L, 26L, 26L, 20L, 19L, 18L, 17L, 15L, 14L, 12L, 12L, 
    12L, 12L, 12L, 11L, 10L, 9L, 9L, 9L, 9L, 8L, 7L, 7L, 7L, 
    7L, 7L, 7L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 5L, 5L, 5L, 5L, 5L, 
    5L, 5L, 5L, 5L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 
    4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 3L), rank = c(1L, 2L, 
    3L, 4L, 5L, 6L, 7L, 8L, 9L, 10L, 11L, 12L, 13L, 14L, 15L, 
    16L, 17L, 18L, 19L, 20L, 21L, 22L, 23L, 24L, 25L, 26L, 27L, 
    28L, 29L, 30L, 31L, 32L, 33L, 34L, 35L, 36L, 37L, 38L, 39L, 
    40L, 41L, 42L, 43L, 44L, 45L, 46L, 47L, 48L, 49L, 50L, 51L, 
    52L, 53L, 54L, 55L, 56L, 57L, 58L, 59L, 60L, 61L, 62L, 63L, 
    64L, 1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L, 10L, 11L, 12L, 13L, 
    14L, 15L, 16L, 17L, 18L, 19L, 20L, 21L, 22L, 23L, 24L, 25L, 
    26L, 27L, 28L, 29L, 30L, 31L, 32L, 33L, 34L, 35L, 36L, 37L, 
    38L, 39L, 40L, 41L, 42L, 43L, 44L, 45L, 46L, 47L, 48L, 49L, 
    50L, 51L, 52L, 53L, 54L, 55L, 56L, 57L, 58L, 59L, 60L, 61L, 
    62L, 63L, 64L, 1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L, 10L, 11L, 
    12L, 13L, 14L, 15L, 16L, 17L, 18L, 19L, 20L, 21L, 22L, 23L, 
    24L, 25L, 26L, 27L, 28L, 29L, 30L, 31L, 32L, 33L, 34L, 35L, 
    36L, 37L, 38L, 39L, 40L, 41L, 42L, 43L, 44L, 45L, 46L, 47L, 
    48L, 49L, 50L, 51L, 52L, 53L, 54L, 55L, 56L, 57L, 58L, 59L, 
    60L, 61L, 62L, 63L, 64L, 1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 
    9L, 10L, 11L, 12L, 13L, 14L, 15L, 16L, 17L, 18L, 19L, 20L, 
    21L, 22L, 23L, 24L, 25L, 26L, 27L, 28L, 29L, 30L, 31L, 32L, 
    33L, 34L, 35L, 36L, 37L, 38L, 39L, 40L, 41L, 42L, 43L, 44L, 
    45L, 46L, 47L, 48L, 49L, 50L, 51L, 52L, 53L, 54L, 55L, 56L, 
    57L, 58L, 59L, 60L, 61L, 62L, 63L, 64L)), class = c("grouped_df", 
"tbl_df", "tbl", "data.frame"), row.names = c(NA, -256L), groups = structure(list(
    genAge = structure(1:4, .Label = c("Women, 15-19", "Women, 20-24", 
    "Women, 25-35", "Women, 36+"), class = "factor"), .rows = list(
        1:64, 65:128, 129:192, 193:256)), row.names = c(NA, -4L
), class = c("tbl_df", "tbl", "data.frame"), .drop = TRUE))

ageGenderFLow <- 
  ageGenderF %>%
  filter(genAge=="Women, 15-19") %>%
  filter(rank<=10)

ageGenderFHigh <- 
  ageGenderF %>%
  filter(genAge=="Women, 36+") %>%
  filter(rank<=10)

ageGenderF_ <-
  ageGenderF %>%
  filter(word_ %in% ageGenderFLow$word_ |
         word_ %in% ageGenderFHigh$word_)

# get rank order of words for low set
ageGenderFLowRank <- 
  ageGenderF_ %>%
  filter(genAge=="Women, 15-19") %>%
  arrange(rank) %>%
  mutate(order = 1:n()) 

ageGenderF_ %>%
  mutate(word = factor(word_, ordered=TRUE, levels=ageGenderFLowRank$word_)) %>%
  # https://ibecav.github.io/slopegraph/
  ggplot(., aes(x = genAge, y = reorder(rank, -rank), group = word_)) +
  geom_line(aes(color = word_, alpha = 1), size = 1.5) +
  #geom_line(size = 0.5, color="lightgrey") +
  geom_text_repel(data = . %>% filter(genAge == "Women, 15-19"), 
                  aes(label = word) , 
                  hjust = "left", 
                  #fontface = "bold", 
                  size = 3, 
                  nudge_x = -3, 
                  direction = "y") +
  geom_text_repel(data = . %>% filter(genAge == "Women, 36+"), 
                  aes(label = word) , 
                  hjust = "right", 
                  #fontface = "bold", 
                  size = 3, 
                  nudge_x = 3, 
                  direction = "y") +
  geom_label(aes(label = rank), 
             size = 2.5, 
             label.padding = unit(0.15, "lines"), 
             label.size = 0.0) +
  scale_x_discrete(position = "top") +
  theme_bw() +
  # Remove the legend
  theme(legend.position = "none") +
  # Remove the panel border
  theme(panel.border     = element_blank()) +
  # Remove just about everything from the y axis
  theme(axis.title.y     = element_blank()) +
  theme(axis.text.y      = element_blank()) +
  theme(panel.grid.major.y = element_blank()) +
  theme(panel.grid.minor.y = element_blank()) +
  # Remove a few things from the x axis and increase font size
  theme(axis.title.x     = element_blank()) +
  theme(panel.grid.major.x = element_blank()) +
  theme(axis.text.x.top      = element_text(size=10)) +
  # Remove x & y tick marks
  theme(axis.ticks       = element_blank()) +
  # Format title & subtitle
  theme(plot.title       = element_text(size=10, face = "bold", hjust = 0.5)) +
  theme(plot.subtitle    = element_text(hjust = 0.5))
```

To me it looks like adding `plot.margin = margin(0, 0, 0, 0)` in `theme()` just squeaks your labels in your example. Does that seem to help? — aosmith, Mar 27 '19 at 19:42
Thanks, @aosmith. It does in this toy example, but it's not enough in my actual example where labels are even longer. — Eric Green, Mar 27 '19 at 19:46
What do you think about changing your approach and using the current text labels as the y axis tick labels instead (with a secondary axis on the right)? That would keep the space around the plot that you need for the text without cutting anything off. (This brings up another question about how you are treating `rank` as a factor, with equal spacing among levels; is that would it should be instead of numeric?) — aosmith, Mar 27 '19 at 20:27
so the labels for the y-axis would be the words by rank for the Women 15-19 group and Women 36+ for the secondary, right? i'll need to look into how to do that, but might be a decent solution. i go back and forth on factor vs numeric on the rank, and i think i favor the factor tradeoff. — Eric Green, Mar 27 '19 at 20:34
Well, after all that I think what you needed to solve your original problem is to increase the default value of `expand` in `scale_x_discrete()`. This "widens" the edges of the plot in the x direction/shrinks the center of the plot, leaving more space for your labels. See, e.g., `expand = expand_scale(add = .7)` — aosmith, Mar 28 '19 at 13:26

score 6 · Accepted Answer · answered Mar 27 '19 at 20:48

If you are willing to change your approach, you could make a big switch and use the text you are using as labels as the axis labels. You could take advantage of the secondary axis to do separate labels for each side of the plot, so things will look a lot like what you are doing now.

The advantage I see to this is that the text fits because it's now part of the axis.

First here's an example using rank as a factor. You have to make the factor into something numeric via as.numeric() in order to get duplicate axes (so far discrete axes don't have secondary axes). Then there's some work to do be done getting the breaks and labels for the axis on each side, so I move data manipulation into a second step (and make rank2 as the reordered factor for ease of doing the breaks later).

Note also the use of expand in scale_x_discrete() to remove space from around the edges of the panel area.

ageGenderF_ = ageGenderF_ %>%
    ungroup() %>%
    mutate(word = factor(word_, ordered = TRUE, levels = ageGenderFLowRank$word_),
           rank2 = reorder(rank, -rank) )

ageGenderF_ %>%
    # https://ibecav.github.io/slopegraph/
    ggplot(., aes(x = genAge, y = as.numeric(rank2), group = word_)) +
    geom_line(aes(color = word_, alpha = 1), size = 1.5) +
    geom_label(aes(label = rank), 
           size = 2.5, 
           label.padding = unit(0.15, "lines"), 
           label.size = 0.0) +
    scale_x_discrete(position = "top", expand = c(0, .05) ) +
    scale_y_continuous(breaks = filter(ageGenderF_, genAge == "Women, 15-19") %>% pull(rank2) %>% as.numeric(), 
                    labels = filter(ageGenderF_, genAge == "Women, 15-19") %>% pull(word),
                    sec.axis = dup_axis(~., 
                                        breaks = filter(ageGenderF_, genAge == "Women, 36+") %>% pull(rank2) %>% as.numeric(), 
                                        labels = filter(ageGenderF_, genAge == "Women, 36+") %>% pull(word) ) ) +
    theme_bw() +
    # Remove the legend
    theme(legend.position = "none",
          # Remove the panel border
          panel.border     = element_blank(),
          # Remove just about everything from the y axis
          axis.title.y     = element_blank(),
          panel.grid.major.y = element_blank(),
          panel.grid.minor.y = element_blank(),
          # Remove a few things from the x axis and increase font size
          axis.title.x     = element_blank(),
          panel.grid.major.x = element_blank(),
          axis.text.x.top      = element_text(size=10),
          # Remove x & y tick marks
          axis.ticks       = element_blank(),
          axis.ticks.length = unit(0, "cm"),
          # Format title & subtitle
          plot.title       = element_text(size=10, face = "bold", hjust = 0.5),
          plot.subtitle    = element_text(hjust = 0.5) )

From a simple r markdown document this looks similar to your example (although not exact):

You can do the exact same thing with rank as numeric, using scale_y_reverse() to reverse the y axis.

ageGenderF_ = ageGenderF_ %>%
    ungroup() %>%
    mutate(word = factor(word_, ordered = TRUE, levels = ageGenderFLowRank$word_))

ageGenderF_ %>%
    # https://ibecav.github.io/slopegraph/
    ggplot(., aes(x = genAge, y = rank, group = word_)) +
    geom_line(aes(color = word_, alpha = 1), size = 1.5) +
    geom_label(aes(label = rank), 
               size = 2.5, 
               label.padding = unit(0.15, "lines"), 
               label.size = 0.0) +
    scale_x_discrete(position = "top", expand = c(0, .05) ) +
    scale_y_reverse(breaks = filter(ageGenderF_, genAge == "Women, 15-19") %>% pull(rank), 
                    labels = filter(ageGenderF_, genAge == "Women, 15-19") %>% pull(word),
                    sec.axis = dup_axis(~., 
                                        breaks = filter(ageGenderF_, genAge == "Women, 36+") %>% pull(rank), 
                                        labels = filter(ageGenderF_, genAge == "Women, 36+") %>% pull(word) ) ) +
    theme_bw() +
    # Remove the legend
    theme(legend.position = "none",
          # Remove the panel border
          panel.border     = element_blank(),
          # Remove just about everything from the y axis
          axis.title.y     = element_blank(),
          panel.grid.major.y = element_blank(),
          panel.grid.minor.y = element_blank(),
          # Remove a few things from the x axis and increase font size
          axis.title.x     = element_blank(),
          panel.grid.major.x = element_blank(),
          axis.text.x.top      = element_text(size=10),
          # Remove x & y tick marks
          axis.ticks       = element_blank(),
          axis.ticks.length = unit(0, "cm"),
          # Format title & subtitle
          plot.title       = element_text(size=10, face = "bold", hjust = 0.5),
          plot.subtitle    = element_text(hjust = 0.5) )

This looks awesome, @aosmith. Going to explore it with my actual data. Big thanks. — Eric Green, Mar 27 '19 at 21:03

score 0 · Answer 2 · answered Mar 27 '19 at 19:47

0

One option is to save your plot as an object (p) and then use the set_panel_size argument in the egg package to explicitly set height and width of the panel (as done in this answer). Something like this gets you close:

library(egg)
library(grid)

p2 <- set_panel_size(p, width=unit(7,"in"), height=unit(10, "in"))

grid.draw(p2)

answered Mar 27 '19 at 19:47

phalteman

3,442
1
29
46

Thanks, @phalteman. It seems to me like `width` in `set_panel_size()` shrinks the entire plot, including with the labels. They seem to move together. I might not be implementing correctly. Will continue looking into this approach. – Eric Green Mar 27 '19 at 19:57

shrink plot width to make more room for ggrepel labels

2 Answers2