── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag() masks stats::lag()
ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
Style a plot with one of the built-in themes:
p <-ggplot(data = mtcars, mapping =aes(x = wt, y = mpg, color =factor(cyl))) +geom_point(size =rel(3)) +labs(x ="Weight", y ="MPG", color ="Cylinders", title ="mtcars Plot", subtitle ="The subtitle", caption ="The caption")p +theme_gray()
p +theme_bw()
p +theme_dark()
With a dark background we also need to think about the palettes we’re using on the color and fill scales.
p +scale_color_brewer(type ="qual", palette ="Pastel1") +theme_dark()
The theme() function
The many pieces of the theme function are covered in the slides. We can make adjustments to them either by setting a value (for options like legend.position) or by asking, of other elements, whether it’s text, a line, or a rectangle and adjusting the element with element_text(), element_line(), or element_rect(). Remember that a number of elements inherit from a parent, like axis.
p +theme(plot.title =element_text(face ="bold", color ="red", size =rel(3)), panel.grid.major =element_line(color ="black"), panel.grid.minor =element_line(color ="green", linewidth =rel(1.4),linetype ="dotted"), plot.background =element_rect(fill ="steelblue1"))
Themes are just functions that make a bunch of these adjustments
E.g. the built-in theme_bw() modifies theme_grey() with a bunch of additional options, expressed in a theme() call:
but with the special %+replace% operator. You can use + if you like. The difference (read the help for theme_get()) is that %+replace% swaps out the entire element being updated, whereas + does not:
+ updates the elements of e1 that differ from elements specified (not NULL) in e2. Thus this operator can be used to incrementally add or modify attributes of a ggplot theme.
In contrast, %+replace% replaces the entire element; any element of a theme not specified ine2 will not be present in the resulting theme (i.e. NULL). Thus this operator can be used to overwrite an entire theme.
To make a theme yourself, you can begin entirely from scratch or you can adapt an existing theme (like theme_bw() does. For example:
theme_duke <-function(base_size =14, base_family ="", base_line_size = base_size/22, base_rect_size = base_size/22) {## See https://brand.duke.edu/colors/ dukeblue1 <-"#012169"# Really too dark to be of use for much tbh dukeblue2 <-"#00539B" textcolor <-"#FFFFFF" half_line <- base_size /2 ggplot2::theme_minimal(base_size = base_size, base_family = base_family, base_line_size = base_line_size, base_rect_size = base_rect_size) %+replace%theme(axis.text =element_text(color = textcolor, face ="bold"), axis.ticks =element_line(color = textcolor),axis.title =element_text(color = textcolor),plot.title =element_text(color = textcolor, face ="bold",size =rel(2),# left-justified texthjust =0, vjust =1),plot.subtitle =element_text(color = textcolor, size =rel(1.2), hjust =0, vjust =1, margin =margin(t = half_line, b = half_line)),plot.caption =element_text(color = textcolor, hjust =1), # right-justifiedplot.background =element_rect(fill = dukeblue2, color ="white"),panel.background =element_rect(fill ="white", color =NA),panel.border =element_rect(fill =NA, color = dukeblue2),panel.grid =element_line(color = dukeblue2, linetype ="dotted", linewidth =rel(0.4)), panel.grid.minor =element_blank(), strip.background =element_rect(fill ="#FCF7E5"), legend.text =element_text(color = textcolor), legend.title =element_text(color = textcolor),legend.position ="top", legend.key.size =unit(2,"line"),legend.title.position ="top" )}
This is ugly! But you see how much control you have over how things look. Also it serves as a reminder that you will likely have to do quite a bit of detail-oriented work if you want a properly good custom theme. Choose a theme and stick with it.
Source Code
---title: "Example 07: Themes"---## Setup```{r}library(here) # manage file pathslibrary(socviz) # data and some useful functionslibrary(tidyverse) # your friend and mine```Style a plot with one of the built-in themes:```{r}p <-ggplot(data = mtcars, mapping =aes(x = wt, y = mpg, color =factor(cyl))) +geom_point(size =rel(3)) +labs(x ="Weight", y ="MPG", color ="Cylinders", title ="mtcars Plot", subtitle ="The subtitle", caption ="The caption")p +theme_gray()``````{r}p +theme_bw()``````{r}p +theme_dark()```With a dark background we also need to think about the palettes we're using on the color and fill scales. ```{r}p +scale_color_brewer(type ="qual", palette ="Pastel1") +theme_dark() ```## The `theme()` functionThe many pieces of the theme function are covered in the slides. We can make adjustments to them either by setting a value (for options like `legend.position`) or by asking, of other elements, whether it's text, a line, or a rectangle and adjusting the element with `element_text()`, `element_line()`, or `element_rect()`. Remember that a number of elements inherit from a parent, like `axis`.```{r}p +theme(plot.title =element_text(face ="bold", color ="red", size =rel(3)), panel.grid.major =element_line(color ="black"), panel.grid.minor =element_line(color ="green", linewidth =rel(1.4),linetype ="dotted"), plot.background =element_rect(fill ="steelblue1"))```## Themes are just functions that make a bunch of these adjustmentsE.g. the built-in `theme_bw()` modifies `theme_grey()` with a bunch of additional options, expressed in a `theme()` call: ```{r}theme_bw```but with the special `%+replace%` operator. You can use `+` if you like. The difference (read the help for `theme_get()`) is that `%+replace%` swaps out the entire element being updated, whereas `+` does not:> `+` updates the elements of `e1` that differ from elements specified (not NULL) in `e2`. Thus this operator can be used to incrementally add or modify attributes of a ggplot theme.> In contrast, `%+replace%` replaces the entire element; any element of a theme not specified i`n `e2 will not be present in the resulting theme (i.e. `NULL`). Thus this operator can be used to overwrite an entire theme.To make a theme yourself, you can begin entirely from scratch or you can adapt an existing theme (like `theme_bw()` does. For example:```{r}theme_duke <-function(base_size =14, base_family ="", base_line_size = base_size/22, base_rect_size = base_size/22) {## See https://brand.duke.edu/colors/ dukeblue1 <-"#012169"# Really too dark to be of use for much tbh dukeblue2 <-"#00539B" textcolor <-"#FFFFFF" half_line <- base_size /2 ggplot2::theme_minimal(base_size = base_size, base_family = base_family, base_line_size = base_line_size, base_rect_size = base_rect_size) %+replace%theme(axis.text =element_text(color = textcolor, face ="bold"), axis.ticks =element_line(color = textcolor),axis.title =element_text(color = textcolor),plot.title =element_text(color = textcolor, face ="bold",size =rel(2),# left-justified texthjust =0, vjust =1),plot.subtitle =element_text(color = textcolor, size =rel(1.2), hjust =0, vjust =1, margin =margin(t = half_line, b = half_line)),plot.caption =element_text(color = textcolor, hjust =1), # right-justifiedplot.background =element_rect(fill = dukeblue2, color ="white"),panel.background =element_rect(fill ="white", color =NA),panel.border =element_rect(fill =NA, color = dukeblue2),panel.grid =element_line(color = dukeblue2, linetype ="dotted", linewidth =rel(0.4)), panel.grid.minor =element_blank(), strip.background =element_rect(fill ="#FCF7E5"), legend.text =element_text(color = textcolor), legend.title =element_text(color = textcolor),legend.position ="top", legend.key.size =unit(2,"line"),legend.title.position ="top" )}```Now we can do, e.g., ```{r}#| fig-width: 8#| fig-height: 6#| out-width: "100%"bd_colors <-c("#C84E00", "#A1B70D", "#E89923")p +scale_color_manual(values = bd_colors) +theme_duke(base_family ="Open Sans") # See https://brand.duke.edu```This is ugly! But you see how much control you have over how things look. Also it serves as a reminder that you will likely have to do quite a bit of detail-oriented work if you want a properly good custom theme. Choose a theme and stick with it.
