CSc47306730 Scientific Visualization

1
CSc4730/6730Scientific Visualization

• Lecture 05
• Basic Visualization Techniques
• Ying Zhu

2
Graphical Forms

• Bar chart
• Time series
• Table
• Pie chart
• Graph
• Unconventional chart

3
Bar graphs

• Used for
• compare magnitudes among several categories
• the evolution in time (the change) of a
  particular magnitude. 
• the comparison of the evolution in time of
  several categories
• Example http//www.gradeinflation.com

4
Data types

• Data can be classified into three types
• Categorical or Nominal
• The order of the categories is arbitrary
• E.g. birds, mammals, tall, small, green or blue
• Ordinal or Sequential
• Categorical data with logical ordering
• E.g. the months of the year, the alphabet
• Numerical or Quantitative
• The data variables mainly used in bar graphs are
  those of categorical and ordinal types
• Bar graph is not very suitable for representing
  quantitative data.

5
Components

• Quantitative axis, Categorical or ordinal axis,
  and a set of rectangles

6
Main types of bar graphs

• Simple
• Contains only one data series
• Grouped
• Contains several data series

7
Main types of bar graphs

• Stacked

8
Main types of bar graphs

• Floating, bi-directional or paired
• Pictorial
• Range

9
Create effective bar graphs

• Consistency
• The same scale, color, or texture should mean the
  same across the representation

10
Create effective bar graphs

• In stacked bar charts, place in the lower part
  the values that have a lower variation

11
Create effective bar graphs

• Use bar graphs to compare
• Bar graphs are suitable for a categorical axis
• Dont use bar graph if you have a temporal axis
  and you would like to see the trends
• A line chart can show the trend in a better way.
• A line better illustrates the continuity of its
  values.
• Bar graph is appropriate if you have a temporal
  axis and you are interested in comparing specific
  values
• Bars are very useful to allow the user to compare
  particular values.

12
Create effective bar graphs

• Vertical or horizontal bars?
• The properties of the visual pattern shouldnt be
  in contradiction with the real properties of what
  is being represented
• E.g. heights are better represented with vertical
  bars than with horizontal ones, while lengths are
  better shown with horizontal ones.

13
Create effective bar graphs

• When in doubt use vertical bars
• Most cultures recognize height as a bigger value,
  while this does not apply for horizontal
  extension.

14
Create effective bar graphs

• Use paired or bi-directional bar graphs to show
  differentiated trends amongst levels of a single
  variable.
• Dont use paired graphs for more than two
  independent variables
• Use stacked bar charts to represent components of
  a whole in a categorical scale.

15
Create effective bar graphs
16
Case study
17
Case study

• Source gradeinflation.com

18
Case study

• Source gradeinflation.com

19
Visualize time series data

• Maximizing data density
• Maximizing data-ink ratio
• Do not show data out of context
• Avoid chartjunk
• Use a consistent scale or interval
• Be careful when using multiple dimensional
  objects to represent one dimensional data
• Be careful with data aggregation

20
Case study

• Keele and Luke, Social Capital and the Dynamics
  of Trust in Government, American Journal of
  Political Science, 2007, 512, pp. 241-254

21
Case study

• Lebo, et al. Strategic Party Government Party
  Influence in Congress, 1789-2000. American
  Journal of Political Science, July 2007, 513,
  pp. 464-481.

22
Case study
23
Case study
24
Tabular Presentation

• Table is for communication, not data storage
• Three rules for tabular presentation
• Order the rows and columns in a way that makes
  sense
• Dont simply go with the alphabetic orders
• Round numbers for easier comparison
• All is different and important
• Summaries or rows and columns are important

25
When to use graphics?

• Use sentence for relations between 2 or 3 data at
  most
• Table is suitable for between 3 and 20 data
• Graphics is good from 2 on, and its especially
  useful for large amounts of data

26
Pie chart

• Pie graphs are used to show percentages and
  proportions
• The number of items compared should be kept to
  five or fewer
• Order the segments from large to small, beginning
  the largest segment at 12 oclock
• A good way to highlight differences is to shade
  the segments from light to dark
• Use patterns or lines and dots to shade the
  segments (?)(source Manual of the American
  Psychological Ass.)

27
Case study

• Lessons learned
• Bar chart is often better than pie chart

28
Graph

• A graph is an abstract object consisting of
  vertices (or nodes) and edges (or arcs) between
  pairs of vertices.

29
Rules for graph

• Basic rules

30
Rules for graph

• Semantic rules

31
Rules for graph

• Structure rules

32
Rules for graph

• Structure rules

33
Summary of principles of visualization

• Maximize data density
• Maximize data/ink ratio
• The magnitudes must correspond with their
  representation
• Minimize the lie factor
• Dont quote data out of context
• Dont change scale in mid-axis
• Encourage comparison

34
Summary of principles of visualization

• Order your data in a way that makes the most
  sense
• Clearly label your data
• Avoid chartjunk