+ - 0:00:00
Notes for current slide
Notes for next slide

Data Visualization

Dr Thiyanga Talagala

2020 - 02 - 18

1 / 18

Guidelines for effective data visualization

Encoding: translating the data into a visual element on a chart/map/etc.

Image credit: Stuart Hall’s 1973 “Encoding/Decoding” model.

2 / 18

Encoding vs Decoding

Read here

3 / 18

1. Simplicity

  • Colour

  • Increment: Use increments like (0, 2, 4, 6,..) instead of , say (0, 3, 6, 9,...)

  • Scale: Don't plot two unrelated series with one scale on left and one on the right.

  • Style: Flat and simple. No 3D effects, shadows, distracting shadings.

  • Set the baseline to zero.

4 / 18

2. Type of encoding object and attribute used to create a plot

  • Encoding objects: points, lines, bars

  • Value-encoding attributes [to show different pieces of information]: point position, line length, color

  • everyone has different perceptions of visualizations but there are a few simple steps that everyone can follow.

description of the image

Image Credit: Patrik Lundblad

5 / 18

Which visual encoding method is best for you?

description of the image

Image Credit: Patrik Lundblad

6 / 18

Inappropriate encoding

description of the image

Image Credit: Patrik Lundblad

7 / 18

Appropriate encoding

description of the image

Image Credit: Patrik Lundblad

8 / 18

Select meaningful axis ranges

  • When absolute magnitudes are important, the vertical axis should begin at zero.

9 / 18

Data transformation

  • plotting on a logarithmic vertical axis can remove skewness in datasets with ranges that include very large and small values (Cleveland, 1994).

10 / 18

Graph aspect ratio

  • ratio of a graph's height to width

11 / 18

Overlapping points

Plot overlapping points in a way that density differences become clear in scatter plots

Method 1

12 / 18

Overlapping points (cont.)

Method 2

13 / 18

Use lines when connecting sequential data in time-series plot

14 / 18

Aggregate larger datasets in meaningful ways

  • Large quantitative dataset : box-and-whisker plots or through kernel smoothing strategies

  • Combination of quantitative and categorical: dotplots, or linked micromap plots

  • Pie charts should be avoided: difficult to perceive differences in angles

  • long time series: temporal aggregation, averaging values across a large time step ( eg: daily to monthly)

15 / 18

16 / 18

Maintain similar vertical or horizontal axis ranges across subplots

17 / 18

Select an appropriate colour scheme

18 / 18

Guidelines for effective data visualization

Encoding: translating the data into a visual element on a chart/map/etc.

Image credit: Stuart Hall’s 1973 “Encoding/Decoding” model.

2 / 18
Paused

Help

Keyboard shortcuts

, , Pg Up, k Go to previous slide
, , Pg Dn, Space, j Go to next slide
Home Go to first slide
End Go to last slide
Number + Return Go to specific slide
b / m / f Toggle blackout / mirrored / fullscreen mode
c Clone slideshow
p Toggle presenter mode
s Start & Stop the presentation timer
t Reset the presentation timer
?, h Toggle this help
Esc Back to slideshow