Corporate Design Flashcards
Typography, font, spacing for plots, products, marketing, etc… (39 cards)
Which graphs can you choose for each of the following?
- deviation
- correlation
- ranking
- distribution
- evolution
- part to whole
- magnitude
- deviation: (a) change from baseline, (b) waterfall
- correlation: (a) scatter plot, (b) heat map
- ranking: (a) horizontal bar chart, (b) dot plot
- distribution: (a) boxplot, (b) histogram
- evolution: (a) kaplan meler, (b) line plot
- part to whole: (a) stacked barchart, (b) tree map
- magnitude: (a) verticle bar chart, (b) forest plot
which is right graph for plotting: deviation?

which is right graph for plotting: correlation?

which is right graph for plotting: ranking?

which is right graph for plotting: distribution?

which is right graph for plotting: evolution?

which is right graph for plotting: part-to-whole?

which is right graph for plotting: magnitude?

where to place labels?
try next to data before attenpting legend

how to compare elements?
group on the same plot

how to order?
rank or alphabetize

how to compare 1-d judgements?
common vertical scale

how to plot means?
mean difference (not two means)
reference lines
user them as anchors

differentiating between categories of same variable
do not use different colors

how to compare polar opposites?
red for positive, blue for negative

how to emphasize details (like baseline)?
color: alpha (bold) or contrasting colors

how to emphasize data?
lighten the ink (soften gridelines with light color)

how to reduce plot clutter?
- remove background noise and excessive colors
- use dots over a bar
- repeat same plot excentuating only one plot at a time (only applies for undistinguishable lines - i.e. too dense)
plotting cause vs. effect
cause on the x-axis and effect on the y-axis

slope
use 45 degree angle to avoid over-interpretation of slope

log-normally distributed variables (e.g. AUC)
do NOT plot on linear scale

space for plots
use space to denote scalar distance
(for instance spread out non-linear weighting times non-linearly)

storytelling using plots
plot inferences to support stories






