In BrookingsInstitution/ggbrookings: Grammar of Graphics in Brookings Style

knitr::opts_chunk$set(
  collapse = TRUE,
  warning = FALSE, 
  message = FALSE,
  comment = "#>"
)

library(ggbrookings)

You can check out all the colors for each palette by running brookings_view_palette('name_of_palette'). The colors in some of these palettes may still change from feedback and further experimentation. Some examples below:

Brand

The brand palette is the default for the scale_brookings_ functions. Use colors from the brand palette whenever possible.

brookings_view_palette('brand1')
brookings_view_palette('brand2')

Analogous

Different shades of the same hue, or of similar hues can be used when the associated values are related.

brookings_view_palette('analogous1')
brookings_view_palette('analogous2')

Contrasting

Colors on the opposite ends of the spectrum. Use Brookings Blue with Secondary colors.

brookings_view_palette('contrasting1')
brookings_view_palette('contrasting2')

Semantic

Where applicable, use colors that are associated with certain concepts. For example, semantic3 could show subsets of gender data (female, male and other).

brookings_view_palette('semantic1')
brookings_view_palette('semantic2')
brookings_view_palette('semantic3')

Positive and negative

Shows pros, cons and neutral, or positive, negative and neutral data.

brookings_view_palette('pos_neg1')
brookings_view_palette('pos_neg2')

Political

Use red and blue of similar intensity to represent data related to political parties in the US. Yellow in political3 and political4 represents ‘Independent’ category

brookings_view_palette('political1')
brookings_view_palette('political2')
brookings_view_palette('political3')
brookings_view_palette('political4')

Categorical

Distinguish discrete categories of data that do not have an inherent ordering. This palette uses distinct hues of visually similar brightness and saturation to avoid creating a sense of order or priority. A maximum of 6 categorical colors are provided as it becomes difficult for people to distinguish between hues after 5-8 colors. If more than 6 colors are needed, consider alternative solutions such as grouping some categories or using alternative visual cues.

brookings_view_palette('categorical')

Sequential

Sequential palettes are used to show an inherent order or variations in numeric values. Light colors typically indicate low values and darker or brighter colors indicate high values.

brookings_view_palette('sequential1')

In addition to variations in lightness, we can also use two distinct hues to give more contrast.

brookings_view_palette('sequential2')

Diverging

Diverging palettes are useful when dealing with negative and positive values or a range of values that have two extremes with a baseline central value, like zero. The Brookings diverging palette should uses two distinct hues of similar brightness and saturation with a neutral color in the middle. Using a discrete set of colors with evenly distributed gradation can improve clarity of values relative to a continuous palette.

brookings_view_palette('diverging')

Miscellaneous

A pleasing option using Brookings Blue and accent yellow.

brookings_view_palette('misc')

BrookingsInstitution/ggbrookings documentation built on Jan. 28, 2024, 9:29 p.m.