DataVis Colour Basics

Summary

This video script offers a comprehensive guide on color usage in data visualization, emphasizing the importance of color scales for representing numerical data and categories. It discusses the psychological impact of color, suggesting dark colors for higher values and light for lower ones. The script covers various color scales, including sequential and diverging, and provides tips for choosing categorical colors that are both aesthetically pleasing and functionally distinct. It also addresses accessibility, such as color blindness considerations, and stresses the need for high contrast and intuitive color choices to ensure clarity and comprehension in data visualization.

Takeaways

• 🎨 **Use Color Scales Wisely**: When representing numerical data with color, create a scale that maps larger numbers to darker colors and smaller numbers to lighter ones for better visual comprehension.
• 🌈 **Sequential Color Palettes**: Utilize sequential color palettes that transition from one hue to another, such as a gradient from blue to green, to effectively represent changes in data.
• 🔄 **Divergent Color Scales**: For comparing two opposing categories, use divergent color scales that meet at a midpoint, ensuring that hues are distinct to emphasize the difference.
• 📊 **Categorical Color Choices**: When mapping data categories, choose colors that are distinct from each other to ensure readability, especially for audiences with varying visual capabilities.
• 🚫 **Avoid Adjacent Hues**: For categorical data, avoid using colors that are too close on the color wheel; instead, maximize differences by selecting hues with at least 35° of separation.
• 🔢 **Limit Color Palette**: Restrict the use of colors in data visualization to five or six to maintain ease of understanding and memorability for the audience.
• 📈 **Use Colors to Highlight Important Data**: Employ colors to emphasize key data points, using gradients for patterns and bars or positions for specific values to enhance readability.
• 🔗 **Consistency in Color Usage**: Maintain consistency by using the same color for the same variables across different charts to avoid confusion.
• 👀 **Ensure High Contrast**: Pay attention to color contrast for accessibility and readability, especially important for text and small elements in a visualization.
• 🌐 **Consider Cultural Color Meanings**: Be aware of the cultural significance of colors and use intuitive colors that resonate with the audience's preconceived associations.
• 👓 **Check for Color Blindness Compatibility**: Utilize tools to verify that color choices are distinguishable by individuals with color vision deficiencies to ensure inclusivity.

Q & A

What is the primary purpose of using color in data visualization?

The primary purpose of using color in data visualization is to make shapes visible, encode data or categories, and enhance the visual representation of numerical data.

Why is it important to create a scale when using color to represent numerical data?

Creating a scale is important because it helps in mapping numerical data effectively. It uses a gradient that starts with a larger number on one end and a smaller number on the other, leveraging visual psychology to guide color choices.

How does the association of darker colors with density influence color choices in data visualization?

Darker colors are perceived as having higher value or representing more of something due to their association with density. Therefore, they are typically mapped to the larger end of a scale, while lighter colors represent less or smaller values.

What is a sequential color scale and how is it developed?

A sequential color scale is a gradient that progresses from one color to another, usually from a darker color representing higher values to a lighter color for lower values. It is developed by selecting hues that are adjacent on the color wheel and maintaining a consistent lightness.

What is the significance of using cool and warm colors in data visualization?

Using cool colors like blue or purple for high contrast and warm colors for low contrast on the opposite end helps to accentuate the transition from dark to light, making it easier for viewers to understand the data.

How does the choice of colors affect the differentiation between categories in a data visualization?

The choice of colors should maximize differences between categories to ensure they are easily distinguishable. Colors that are too similar or close on the color wheel can be confusing, especially for those with impaired vision or on low-quality screens.

What is a diverging color scale and how should it be used effectively?

A diverging color scale uses two different scales to represent two opposing ends of a spectrum, such as comparing Democrats and Republicans. It should be used effectively by ensuring that the hues of each scale are distinct and do not get too close to each other, accentuating the distance from the midpoint.

Why is it recommended to limit the color palette to five or six colors in data visualization?

Limiting the color palette helps to maintain readability and ease of understanding. More than six colors can overwhelm the viewer, making it difficult to remember the meaning behind each color and potentially confusing the visualization.

What is the role of gray in data visualization and how should it be used?

Gray is an important color in data visualization as it can be used for less important parts of the chart, allowing color to be used as a highlight for important data points. It's also useful for general context data and less important annotations.

Why is high color contrast important in data visualization, and what are the recommended contrast ratios for text?

High color contrast is important for accessibility and readability, especially on screens in low light conditions. The recommended contrast ratio is 2.5 for large text and at least 4.5 for small text.

How can color choices be improved when considering the target audience's cultural associations with colors?

Color choices can be improved by using intuitive colors that the target audience will associate with the data. This includes considering cultural meanings, avoiding stereotypes, and using colors that are naturally or learnedly associated with certain concepts, like red for alert or green for go.

What is the impact of using gradient color palettes for categories instead of values?

Using gradient color palettes for categories can lead to confusion as viewers may misinterpret the gradient as representing different amounts or values of the categories, rather than distinct categories.

How can color blindness be considered when choosing color palettes for data visualization?

Color blindness can be considered by using online tools to check color combinations for visibility to those with color vision deficiencies. Ensuring that colors used can be distinguished by all viewers is crucial for inclusive data visualization.

Outlines

00:00

🎨 Color Basics in Data Visualization

This paragraph introduces the fundamental role of color in data visualization, emphasizing its use for making shapes visible and encoding data categories. It discusses the importance of color scales, particularly in representing numerical data, and the psychological impact of color intensity and brightness. Dark colors are associated with higher values due to their perceived density. The paragraph also explores the concept of sequential color scales, both single-hue and multi-hue, and their development from a color wheel, highlighting the effectiveness of color differentiation in conveying data insights.

05:00

🌈 Choosing Colors for Categorical Data

The second paragraph delves into the challenges of selecting colors for categorical data representation, where color differentiation is crucial for distinguishing between data objects. It warns against the temptation of choosing adjacent hues on the color wheel, which may appear harmonious but can be difficult to differentiate, especially on lower-quality screens or for those with impaired vision. The paragraph suggests maximizing differences between hues and maintaining consistent lightness to improve readability. It also advises limiting the color palette to five or six colors to enhance memorability and usability.

10:03

📊 Gradient Colors and Data Patterns

This paragraph discusses the use of gradient colors to illustrate patterns in data, such as in a choropleth map. It advises against relying solely on gradient colors to convey precise values, suggesting the use of bars or positional indicators for clarity. The paragraph also touches on the importance of using colors to denote categories rather than numerical values and the potential confusion that can arise from using too many colors in a single chart, recommending a maximum of seven colors for ease of interpretation.

15:04

🖌️ Enhancing Data Visualization with Color Consistency

The fourth paragraph focuses on the importance of color consistency in data visualization. It stresses the need to use the same color for the same variables across different charts to avoid confusion and maintain clarity. The paragraph also highlights the necessity of explaining what each color represents, providing examples of effective color keys that include both numerical ranges and textual explanations to ensure that readers can easily interpret the data presented.

20:06

🔍 Improving Color Choices for Accessibility and Clarity

This paragraph offers guidance on making better color choices to enhance data visualization accessibility and clarity. It suggests using gray for less important parts of a chart to allow color to highlight key data points. The paragraph also emphasizes the importance of high contrast ratios for readability, especially for text, and advises against using complementary hues that can be difficult to distinguish. Tools for testing color contrast compliance are mentioned, along with the suggestion to consider the placement of colors in relation to each other for effective comparison.

25:07

🌍 Cultural and Psychological Considerations in Color Selection

The sixth paragraph examines the cultural and psychological aspects of color selection in data visualization. It encourages the use of intuitive colors that align with the expectations and associations of the target audience, such as using natural colors for environmental data or culturally recognized colors for political affiliations. The paragraph also addresses the issue of color blindness, emphasizing the need to ensure that color choices are distinguishable for those with color vision deficiencies, and provides a resource for checking color combinations for accessibility.

Keywords

💡Data Visualization

Data Visualization refers to the graphical representation of information and data. It is a key theme in the video, which discusses how to effectively represent numerical data using shapes and colors. The script emphasizes the importance of using color to encode data or define categories, as it can enhance the visibility of data and help audiences understand complex information more intuitively.

💡Color Scale

A color scale is a range of colors used to represent numerical data, with one color at one end of the scale and another at the opposite end. In the script, it is explained that darker colors are associated with higher values and lighter colors with lower values, which is a concept used to create effective visual representations of data.

💡Sequential Color Palette

A sequential color palette is a type of color scale where colors change gradually from light to dark, representing an increase in data values. The script uses this concept to illustrate how a color scale can be developed from a single hue or multiple hues, providing examples of how these scales can be used to represent data effectively.

💡Categorical Data

Categorical data refers to variables that are divided into groups or categories. The script discusses the use of color to represent these categories, emphasizing the need to choose colors that are distinct from one another to ensure readability and comprehension.

💡Diverging Color Scale

A diverging color scale is used to represent two opposing ends of a spectrum, often to compare two different sets of data. The script explains that this type of scale should have a clear midpoint and that the hues chosen should be distinct to highlight the contrast between the two data sets.

💡Color Contrast

Color contrast is the difference in color intensity between elements in a visual representation. The script stresses the importance of high contrast for readability and accessibility, especially for text and for distinguishing between different data points or categories.

💡Color Psychology

Color psychology is the study of how colors can influence emotions and behavior. The script touches on this concept by discussing how darker colors are perceived as denser or representing higher values, which is a psychological phenomenon that can be leveraged in data visualization to guide the viewer's interpretation.

💡Categorical Color

Categorical color refers to the use of color to differentiate between distinct categories in data. The script warns against using colors that are too similar, as this can lead to confusion and difficulty in distinguishing between categories, especially in less-than-ideal viewing conditions.

💡Color Accessibility

Color accessibility is the practice of ensuring that color choices in visual representations are usable by people with various visual impairments, including color blindness. The script advises checking color combinations for color blindness to ensure that all viewers can effectively interpret the data presented.

💡Intuitive Colors

Intuitive colors are those that are easily associated with certain concepts or data by the audience. The script suggests using colors that are culturally or contextually associated with the data being represented, such as using blue for Democrats and red for Republicans, to make the visualization more immediately understandable.

💡Color Gradient

A color gradient is a gradual transition between two or more colors. The script discusses the use of gradients to represent a range of values in data, cautioning against using gradients for categorical data and recommending the use of lightness to build gradients rather than hue alone.

Highlights

Color in data visualization is used to make shapes visible and encode data or categories.

A color scale represents numerical data with a gradient from a larger number to a smaller one.

Darker colors are associated with higher values due to visual psychology.

A sequential color palette can be developed using a single hue or multiple hues for better differentiation.

Categorical color choices should be distinct to ensure readability, especially for those with impaired vision or on lower quality screens.

Categorical colors should be chosen from different areas of the color wheel to maximize differences.

Limiting the color palette to five or six colors improves readability and memorability.

Gradient colors can show patterns but may not be ideal for deciphering exact values.

Using bars or position in addition to color can help in distinguishing segments and values.

Consistency in color usage across different charts for the same variables is crucial for clarity.

Gray is an important color in data visualization, used to highlight important data points by contrasting with less important information.

High color contrast is essential for accessibility and readability, especially for text.

The proximity and size of colored areas on a chart affect the ease of comparison.

Cultural and intuitive meanings of colors should be considered when choosing a color palette.

Avoid using gradient color palettes for categorical data to prevent confusion about the quantity or value.

Lightness should be used to build gradients, not just hue, for better readability.

Divergent color gradients can effectively represent comparisons between two opposing categories.

Color blindness should be considered to ensure that all viewers can distinguish colors in a visualization.

Transcripts

hi there and welcome back to data viz

today's short PO is on color Basics and

as we know data Vis is can be defined as

representing numbers with shapes and no

matter I guess how what those shapes

look like whether they're area areas

lines or dots um they need to have a

color sometimes colors just make the

shapes visible sometimes they encode the

data or the categories themselves

um but today we're mainly going to focus

on the latter in this talk you know

defining categories and giving them

colors and how to make some sensible

choices there we'll take a general look

at colors and we'll consider um what to

look at when choosing them okay so let's

keep

going all right so let's just look at

this for a moment when you use color to

represent a number you need to

create some sort of scale um this is

important um don't use a scale for

mapping categorical data and we'll talk

about that a little bit more why later

think of your scale as a gradient with a

larger number on one end and a smaller

number on the other end and we can take

this advantage of the visual psychology

of

this to sort of help us with our color

choices we associate darker colors with

density and density with greater numbers

because of this dark colors are

perceived as being higher in value or

there's more of it than lighter ones so

make sure you map the large end of your

scale or the thing with the biggest

numbers um to a dark color and the small

end to a lighter color and the bigger

difference between these two extremes um

the more effective your use of color so

the difference in the in the numbers but

also um the scale of you know the

intensity of the colors your choosing we

can start with a simple black to white

scale for example by swap swapping black

for another dark color we can make

things a little bit easier to look at um

and instead of white we could make the

other end of the scale yellow for

example and as you shift towards darker

blue your scale slowly changes and

becomes a SE Green in the middle you can

see that I've put that in the multi hu

sequential color palette there um so

this is an example of a multi Hu

sequential scale and you can see from

the color wheel how that's been

developed and it's the one just above my

head here you can see the scale going

from five up to zero but you can also

see the color wheel and how it's mapped

through there to make it a little bit

more um buied in

multicolors um if your charts on a light

background for example it's best to

start with a cool high contrast color

like blue or purple and use a warm low

contrast on the other end um this will

accentuate the dark to light transition

moving in from the opposite direction

fights against this natural Trend and

it'll be more difficult to read and

won't look great um so you can see

they're the two types we've got here of

sequential color scales is the single

Hue so you just go through one color

there it is with the blue and then the

multi- Hue going through the color wheel

a little bit there um multi Hue

sometimes can be much more effective

because for us you can see oh yeah it's

going to Green that's going to be easier

to understand than going just to light

blue sometimes we will have trouble

differentiated between the Hues of blue

but you can see by using that second

green color palette it's kind of helped

there so the main takeaways from this is

that um when using a color scale make

sure you know the dark colors are the

things with the most numbers and the

light obviously going to less so dark

equals more light equals less okay so

let's move on uh diverging color scales

our third one in this group so we had

multicolor we had a single one diverging

ones we use two different scales to um

represent the numbers and we can see

here because they're comparing two

things and in this case um it's between

the Democrats and the Republicans um we

can

see that there's two things and we're

measuring how how much Republican it is

so the darker the more Republican the

lighter red there the less and for the

Democrats the same sort of thing except

with blue but we can see where they

converge in the middle somewhere in a

light tone there um so to be effective

in using this kind of diverging color

scale uh we need to think of our scale

as two sequential scales that share a

low value and that's that white one in

the middle this would be our midpoint

um be sure that the Hues of each scale

don't get too close to one another when

you're using this technique because we

want to accentuate the distance from a

midpoint not hide it so hopefully that

makes sense to most of you there all

right moving on why okay still on our

categorical color and we're using Color

to represent categories when we're

mapping data uh without numerical

meaning so categories or a text name for

category we want readers to be able to

tell um our data objects apart so

whether they're slices on a pie CH the

lines or bars Etc so as designers we

want these colors to look good next to

each other and because of that we often

choose adjacent viewes of things that

are close in color they look nice and

we're tempted to do this um and with our

fancy laptops or with my fantastic

screen here it's very easy for me to

tell the St dates apart because I've got

great color separation you know it's

represented quite nicely but if you look

at the colors like we do sometimes in

the classroom on a projector or in fact

on an old computer screen monitor or

you've got impaired Vision or you you're

older than 30 these colors start to look

very similar and it's a big problem

because the entire reason for using

Color here is to help our readers

understand objects that are different

from one another um not NE their

relationships but they are different

categories so if we see here I've got um

three types of categorical

color and so categorical colors

referring to the categories that we're

using to Define things not how much of a

thing that they are so we can see You'

got alpha bravo charlie Eko and Delta in

that first one they're quite beautiful

together you know they they they don't

look like a candy strip they they're

quite harmonious but I I would say that

they're not the most usable

uh color palette that you could have in

the center we've got ones which are

separated a little bit more on the color

wheel there with the black dots you can

see that uh from the going from the blue

down to the yellow for Delta I think

they're usable usable but kind of starts

to look like you've gone to the Lolly

shop um not so beautiful or harmonious

together and then the last one also

beautiful but still not usable where

we've

got similar tones or like too dark alpha

bravo and Charlie all looks very similar

so categorical colors we're going to

sort of look at that a little bit more

let's keep

going okay so my last one is trying to

show The Best of Both Worlds we can see

that there's a it's still got the candy

effect but we can see that um at least

we can see the separation between the

colors there's no way I could confuse

the orange of echo with the yellow of

delta or the Bravo or Al with each other

um but they look slightly more

harmonious together

um hence the thing so maximizing the

differences from bouncing to and from

opposite ends of the color wheel

um you know is a good thing form or

function is something you always need to

toss up in data VI and we usually go for

function over form

um um but let's always recall the py

psychological phenomenon where we

associate darker colors with more of

something and lighter colors with less

of something and make sure that you're

not unwittingly or unknowingly telling

out your readers that the darker colors

represent higher values and you can see

hence this choices for the colors for

the categories um are trying not to

confuse our brains with that there's no

sort of one size fits all answer to

these these questions it's always going

to be a compromise when you're choosing

um you'll need to always relinquish some

sort of aesthetic um choice for some

readability ones um if you are choosing

adjacent Hues or Hues that are close to

each other on the color wheel um you can

but take bigger steps between them at

least 35° so leave a a segment in

between of H change um and maintain a

consistent lightness you can see the

lightness is on the uh second ring out

from the center there and that they're

separated at least by one of the

sections there so more than 35° of H

change um the other thing would be to

limit your color palette to five or six

colors um more than that um your readers

or viewers or whatever you want to call

them will have a quite a hard time

remembering the meaning behind any more

than that anyway all right let's move on

um when to use colors in data

Vis

um gradient colors as we've just started

with today uh can be great to show a

pattern you know something like a corle

map that's like a heat map but it's hard

to decipher actual values from them and

to see differences between values

consider showing your most important

values using

bars or position like in a DOT graph or

even area

and use colors to only show the

categories um it mean that your readers

will be able to decipher your values

faster and if we look at this we've got

a gradient value here CH in color change

from poor to rich and Shore we go richer

the darker ones that's fine um but you

can see that the shape by

adding circles makes it easier for for

us to differentiate the segments if I

just gave you that sliding value change

there would be very difficult to sort of

kind of guess at where one segment fits

to another we can see that's not ideal

but if we look at the next section here

you can see that they sort of start

working with the colors but you can see

with the addition of some bars here when

we're comparing people in group a which

are the rich people obviously and b poor

people we can see how they are

differentiated using the bars and it

makes it very easy to see how group A

and B are segmented across these

particular

categories

okay if you need more than seven colors

in a chart I'm going to ask you to

consider using another chart type or to

group the categories together and with

the MyTime data you'll notice that we

have overarching categories and then

we've got activities within them um so I

think that's important uh try to avoid

using more than seven of them that's

kind of a rule of thumb thing the more

colors in the chart rep to represent

your data the hard it becomes to read it

very quickly your readers will often

need to consult back to the legend or

the color key to understand what is

shown in your chart so if that's the

case you might want to choose using

another chart so we can see there we'

got the Candy Camera if I look at just

that stacked bar chart at the top with

the colors it's kind of okay I will

probably put them in order of um how

much they are but you know there's some

segments there I'd need to know what a

was and then I'd look through it would

be quite difficult but you can see in

the second one there the Stacked bar

chart here we can compare them easily

across even though they are

gray okay let's move on to how to make

better color choices again um always

consider using and this is another rule

um the same color for the same

variables you can see here um if you use

one color for all of your charts using

the same color every time is the best

option um for for displaying your data

for example it's okay to show the

unemployment rate uh in blue in the

first chart that's sort of tearly dark

blue color

um but it's not okay to kind of switch

it around or use it for something else

in another chart so we can see the UK

and China here are compared um UK's the

red one going down China's going up and

then underneath there yeah we can see

that

China is now red and Germany's in blue

so it's the same color palette which

you'd go oh yeah that's great I'm a

designer I've got the color palette no

let's make sure that if you've chosen a

color for one thing it always remains on

the next thing so we can see there where

in the next chart Germany and China

Germany now becomes yellow China is

still going to be the teal color so

we've been consistent and added in

another color to show that so again the

rule is use the same color for the same

variables or the same categories if

you'd like to think of them like that

okay next one

[Music]

um make sure that you explain and some

part in your data VI what your colors

actually are encoding don't just put

them there um so we can see there's a

color several versions of how to do a

color key here uh we've got a color key

of Z to 100 so less to more of something

we haven't told them what that scale

refers to it's not to 100 what of we

don't know and again we've got another

one where it's segmented which I think

is a lot easier to read so not 2550 75

100 and then the third one is quite nice

and that it's in a text version so

you've managed to label how the data is

being represented but also to color code

and provide um some indication about

what they mean so China's going to be in

red and Germany's going to be in that

teal color so make sure you explain to

your readers what your colors are

encoding next one

um how to make better color choices I

would always consider gray as the most

important color in data views um using

gray for less important parts in your

chart allows you to use color as a

highlight to draw attention to that as

being the important data point in your

your

visualization um it's helpful for

General context data um it's less or for

less important

annotations um to show what's not

important to the user gray is great

since gray can sometimes seem a bit cold

uh you could consider using with a hint

of color so a warm gray or something

like that um or as a mattive or a super

light yellow could work as well so we

can see here with this chart we don't

even need to know what it is but it

would be very difficult to read um

what's going on there go what's which

what's what's the important one I should

be looking at and it's the teal line so

you can see just by making the others a

thinner line and grayed out we can see

the context of it our teal data but

we're not distracted by having to look

at all of the other colors there all

right let's move on

um make sure your contrasts are high

enough and color contrast is one of

those things that we need to pay

attention to for accessibility but also

just generally as designers it's

something that's

always you always need to pay attention

to um it it means that we need to care

about how our readers can read charts on

a screen even in low light um and it's

especially important for text the

smaller the text the higher the contrast

needs to be the contrast ratio between

the background and foreground needs to

be quite high it's 2.5 for big text and

at least four for small text um in

addition to having high contrast ratios

you should avoid complimentary hues so

things like red and green orange and

blue are no nose or bright colors for

backgrounds you can use tools to test

your color contrast um and make sure

that they're compliant we'll put a list

of those at the bottom of the Pod

today um so we can see some really bad

examples of contrast there the one: one

1.5 is okay in some cases 2.5 is a good

choice 4.5 is a is a safe a choice and

you can see with the colors horrible my

eyes that's bad non ideal red and green

and you can see how gray use again is

being useful to us all right let's move

on consider where your colors appear in

relation to each other so this is an

obvious one but something to think about

the smaller the areas are on your chart

the bigger or and the bigger the

distance between them the harder it is

to compare them so if we look at those

little circles there we can see that you

know we might have a great scale of that

that green color there but because

they're far apart and they're quite

small circles it's it's going to be

really hard we're asking our brain to

remember colors and then try and compare

them it's almost meaningless that first

one but if they are in fact next to each

other or the area of them is bigger we

can see that it's a better one and you

can see the colored in map there is much

better than just the spots it's the same

map that we're using but you can see

using the area color allows us to see oh

yeah More's happening in the wh the top

top right hand left hand corner for you

um there all right so consider where

your colors appear in relation to each

other and then make some

adjustments all right next one use

intuitive colors um when you're choosing

your color palette consider the meaning

or the culture meaning of your target

audience uh if possible use colors that

the readers will associate with your

data anyway so party color colors that

we looked at before like Republicans red

Democrats are blue things like natural

colors for the forest or water Green

Lake blue or learned colors like red is

alert or attention um or

stop is red or green is sort of go um so

when it it comes to uh color and coding

things like gender as well um let's

consider avoiding stereo stereotypical

pink and blue

combinations um but don't confuse your

readers completely you could try you

know a blue for male or teal color uh we

can see the blue for the female red for

the not ideal we can see over here the

females have gone to yellow and the

males have gone to teal so the not ideal

ones

that's oh that's so over there um good

and bad not ideal good is usually green

bad is usually red or you know it's aert

Forest usually green like is the blue

Etc so use intuitive colors where

possible next one um

using using light colors for low values

and dark colors for high values and I've

gone on about this a little bit already

um so when you're using Color gradients

make sure that the bright colors low

values we know this now dark colors High

values or big things um this will make

it inity for most readers and you can

see I flipped it around there in that

first map um low to high low values to

high values not very readable we would

always go that dark area at the top

there means there's a lot of something

but now it's the high Valu and then you

can see it flipped to its proper

Arrangement there so that's a very

intuitive one to take note of okay next

one don't use a gradient color palette

for

categories so we saw it just used a

second ago we use a gradient pal palette

for values or the numbers of things do

not use a gradient color P palette for

categories so we can see Ivan David Anna

and Lisa

here but immediately if I look at that

not ideal chart I would say that this in

some way Ivan has more of something than

everybody else um but you can see by

choosing a different color palette we

can see um we're not going to confuse

our our um readers or viewers so don't

use a gradient color palette so use of

the same color for categories

all right let's go back to some corop

PLS again um use lightness to build

gradients not just the

Hue so you can see there's a hue base

gradient in the first map it looks

lovely I think it looks great to look at

I like the teal and orange it's a great

combination but at some point it becomes

useless to me because I can't really

looking at that tell what the

concentration is I get the red is the

most of

something um but I kind of get fuzzy

somewhere in the middle but you can see

it's better for us just to use a

lightness based gradients as in the

second map there I can quite easily

compare areas or locations across that

map gradients with too much variation

like rain rainbow scale scales will

definitely confuse your viewers or

readers okay cons you could

to help with this consider using two

Hues for a gradient not just one and you

can see the one H it's not so bad we saw

that a moment ago but using two Hues

here going from the green to the dark

blue Works quite nicely as well and

makes it a little less overwhelming to

look at the

map so two or three carefully selected

Hues can work in that

case next uh consider they using

Divergent color diverging color

gradients so here we're going from

Orange to Red so they're not very

Divergent they're quite close to to each

other in the color wheel but you can see

teal and orange they're opposites on the

color wheel and you can see that that is

a better choice um to to

display um the color Center should

always ideally be a light gray rather

than a white as well

you can see both of those they look the

first one looks white second one we've

got a gray happening in

there all right we're almost there

people um next one color blindness it's

something that we all need to pay

attention to and I've put a link there

to data wrapper to check for color blind

combinations um people with color defici

color vision deficiencies will need to

be able to distinguish your colors

there's many types there's many online

tools data wrappers one is a great one

for checking um to make sure they can

consider so you can see that your map

can be read and we can see normal vision

for the first one there on the right

there up there

um you can

see going through how different people

with different blindnesses will go

through that and you can see it becomes

quite muddy in the second and third

versions there on the the left hand side

so always do a do do a quick check

before you commit to your color um

choices for color

blindness um some further reading here

um and I'll put that down below below

today's um meure P all right well that

was it our quick introduction to color

um there's a lot more we could say say

about it but just to wrap up remembering

when you're going to be using gradients

and remembering when you're using scale

remembering some of the rules about

category colors um and hopefully that

was helpful so thanks for listening and

I'll see you in class