Humans process visual info with pre-attentive perception and Gestalt principles.
– White spaces create groupings (group into rows? columns? separate? bunch similar?). Anything that moves together is also grouped as a unit.
Definition – making sense data, facilitate consummation in an easy way; display abstract info; shift the balance between perception and cognition.
How to make successful – quantities need to be selected accurately, clear, allow comparison, “this number is ___ compared to what?”. Support decisions.
- Define goals.
- Fix the data. (overlap on what the data represents?)
- Highlight the important stuff
- Eliminate distractions
- Provide clear information hierarchy
- Provide filters
- Assess info users really need
- Understand their goals
- Select and aggregate the data properly
Anatomy of Quantitative Data
- Measures – raw data
- Dimensions – organize the data
- Categorical – divisions
- Relationship/Filters – sorting, cluster data
What is the story?
- Variations within measures (SD, mean, normal distribution)
- Relationship between measures (correlation, i.e.)
- Variations in time matter
- Variations within dimensions
- Variations across Time
- Variations across Space
Line and bar charts – show the zero point
- Raw measures – typically money and time
- Normalized measures (designed) (like sales per sales person, sales per advertising dollar spent, sales per square foot of store, etc._
- Summarized measures (statistics like subtotals, averages, etc)
Dimensions Organize the Data
- Hierarchical or Categorical
- Exact schema (IA)
- Natural dimensions (scalar values)
- Synthetic dimensions (designed, dynamic)
Ways to slice up the data
- Normalization (comparison or baseline)
- reduce noise
- filter on measures or dimensions
Big three Data Design tasks
Note that comparison to business goals (subjective target) is not a signal.
Visualization -relationships within charts (slide 42)
-Dimension relationships (nominal, ordinal, interval, hierarchical).
You show the measure relationships:
Bar Charts – tell a story about differences. good to show rank
Lines – for trends