19 Critique

Hadley Wickham

Stat405Graphical theory & critique

Monday, 2 November 2009

Exploratory graphics

Are for you (not others). Need to be able to create rapidly because your first attempt will never be the most revealing.

Iteration is crucial for developing the best display of your data.

Gives rise to two key questions:

Monday, 2 November 2009

What should I plot?How can I plot it?

Monday, 2 November 2009

Two general tools

Plot critique toolkit: “graphics are like pumpkin pie”

Theory behind ggplot2:“A layered grammar of graphics”

plus lots of practice...

Monday, 2 November 2009

Graphics are like pumpkin pie

The four C’s of critiquing a graphic

Monday, 2 November 2009

Content

Monday, 2 November 2009

Construction

Monday, 2 November 2009

ContextMonday, 2 November 2009

ConsumptionMonday, 2 November 2009

Content

What data (variables) does the graph display?What non-data is present?What is pumpkin (essence of the graphic) vs what is spice (useful additional info)?

Monday, 2 November 2009

Your turn

Identify the data and non-data on “Napoleon's march” and “Building an electoral victory”. Which features are the most important? Which are just useful background information?

Monday, 2 November 2009

Results

Minard’s march: (top) latitude, longitude, number of troops, direction, branch, city name (bottom) latitude, temperature, dateBuilding an electoral victory: state, number of electoral college votes, winner, margin of victory

Monday, 2 November 2009

Construction

How many layers are on the plot?What data does each layer display? What sort of geometric object does it use? Is it a summary of the raw data? How are variables mapped to aesthetics?

Monday, 2 November 2009

Perceptual mapping1. Position along a common scale 2. Position along nonaligned scale 3. Length4. Angle/slope5. Area 6. Volume7. Colour

Best

Worst

For continuous

variables only!

Monday, 2 November 2009

Your turn

Answer the following questions for “Napoleon's march” and “Flight delays”:How many layers are on the plot?What data does the layer display? How does it display it?

Monday, 2 November 2009

ResultsNapoleon’s march: (top) (1) path plot with width mapped to number of troops, colour to direction, separate group for each branch (2) labels giving city names (bottom) (1) line plot with longitude on x-axis and temperature on y-axis (2) text labels giving datesFlight delays: (1) white circles showing 100% cancellation, (2) outline of states, (3) points with size proportional to percent cancellations at each airport.

Monday, 2 November 2009

Can the explain composition of a graphic in words, but how do we

create it?

Monday, 2 November 2009

“If any number of magnitudes are each the same multiple of the same number of other magnitudes, then the sum is that multiple of the sum.” Euclid, ~300 BC

Monday, 2 November 2009

“If any number of magnitudes are each the same multiple of the same number of other magnitudes, then the sum is that multiple of the sum.” Euclid, ~300 BC

m(Σx) = Σ(mx)Monday, 2 November 2009

The grammar of graphics

An abstraction which makes thinking about, reasoning about and communicating graphics easier.

Developed by Leland Wilkinson, particularly in “The Grammar of Graphics” 1999/2005

You’ve been using it in ggplot2 without knowing it! But to do more, you need to learn more about the theory.

Monday, 2 November 2009

What is a layer?

• Data

• Mappings from variables to aesthetics (aes)

• A geometric object (geom)

• A statistical transformation (stat)

• A position adjustment (position)

Monday, 2 November 2009

layer(geom, stat, position, data, mapping, ...)

layer( data = mpg, mapping = aes(x = displ, y = hwy), geom = "point", stat = "identity", position = "identity")

layer( data = diamonds, mapping = aes(x = carat), geom = "bar", stat = "bin", position = "stack")

Monday, 2 November 2009

# A lot of typing!

layer( data = mpg, mapping = aes(x = displ, y = hwy), geom = "point", stat = "identity", position = "identity")

# Every geom has an associated default statistic# (and vice versa), and position adjustment.

geom_point(aes(displ, hwy), data = mpg)geom_histogram(aes(displ), data = mpg)

Monday, 2 November 2009

# To actually create the plotggplot() + geom_point(aes(displ, hwy), data = mpg) ggplot() + geom_histogram(aes(displ), data = mpg)

Monday, 2 November 2009

# Multiple layersggplot() + geom_point(aes(displ, hwy), data = mpg) + geom_smooth(aes(displ, hwy), data = mpg)

# Avoid redundancy:ggplot(mpg, aes(displ, hwy)) + geom_point() + geom_smooth()

Monday, 2 November 2009

# Different layers can have different aestheticsggplot(mpg, aes(displ, hwy)) + geom_point(aes(colour = class)) + geom_smooth()

ggplot(mpg, aes(displ, hwy)) + geom_point(aes(colour = class)) + geom_smooth(aes(group = class), method = "lm", se = F)

Monday, 2 November 2009

Your turnFor each of the following plots created with qplot, recreate the equivalent ggplot code.

qplot(price, carat, data = diamonds)

qplot(hwy, cty, data = mpg, geom = "jitter")

qplot(reorder(class, hwy), hwy, data = mpg, geom = c("jitter", "boxplot"))

qplot(log10(price), log10(carat), data = diamonds), colour = color) + geom_smooth(method = "lm")

Monday, 2 November 2009

ggplot(diamonds, aes(price, data)) + geom_smooth()

gglot(mpg, aes(hwy, cty)) + geom_jitter()

ggplot(mpg, aes(reorder(class, hwy), hwy)) + geom_jitter() + geom_boxplot()

ggplot(diamonds, aes(log10(price), log10(carat), colour = color)) + geom_point() + geom_smooth(method = "lm")

Monday, 2 November 2009

More geoms & stats

See http://had.co.nz/ggplot2 for complete list with helpful icons:

Geoms: (0d) point, (1d) line, path, (2d) boxplot, bar, tile, text, polygon

Stats: bin, summary, sum

Monday, 2 November 2009

Go back to the descriptions of “Minard’s march” and “Flight delays” that you created before. Start converting your textual description in to working ggplot2 code. Use the data sets available on line to produce the graphics.

Your turn

Monday, 2 November 2009

Other featuresScales. Used to override default perceptual mappings. Mainly useful for polishing plot for communication.

Coordinate system. Rarely useful, but when needed are critical.

Facetting. Have seen in use already. Only other feature of importance is interaction with scales.

Themes: control presentation of non-data elements.

Monday, 2 November 2009

Project 3 preview

Choose your own dataset.

Results will be presented as poster during a formal poster session in the last week of class.

On November 16, Tracey Volz (a communications expert) will come and talk about how to create a good poster

Monday, 2 November 2009

Data suggestions

http://delicious.com/hadley/data

http://github.com/hadley/data-housing-crisis

Monday, 2 November 2009