Embedded Files
“There are three kinds of lies: Lies, Damned Lies, and Statistics” Mark Twain
Representing data
Representing data
Qualitative data
Qualitative data
Pro: Provides context and nuance
Con: Makes aggregation or comparison difficult
Quantitative data
Quantitative data
Pro: Allows for patterns to emerge across a large set of data
Con: Can be misused to reduce a child to a label or number
Data Reduction
Data Reduction
Virtually all data analysis and representation involves reducing the total data set down into a few comprehensible pieces of info.
Jagged Data Profiles
Jagged Data Profiles
As Todd Rose points out, most people have 'jagged profiles' where we are strong in some areas and weak in others. Although there may be an arithmetic average, no person fits into an 'average' mold.
Finding the Right colors
Finding the Right colors
Rag Scale
Rag Scale
(B)RAG Scale
(B)RAG Scale
Non-RAG Scale
Non-RAG Scale
Palette Pickers
Palette Pickers
Matching Visualizations to the Data
Matching Visualizations to the Data
Pie Chart
Pie Chart
- When to use:
To show a part of a whole
Not making a comparison
Fewer than 7 slices
What to watch out for:
What to watch out for:
Comparing pie charts to each other
Too many slices
Example: a set of pie charts to give users an overall picture plus active filters.
Column/Bar Chart
Column/Bar Chart
When to use:
When to use:
To compare in a group, over time
Only a few periods
Use Bar chart if many items
What to watch out for:
What to watch out for:
Too many periods
Tracking multiple series over time
Example: a student profile page on attendance that shows how many days in each month that child had been marked Absent or Late.
Stacked Column/Bar chart
Stacked Column/Bar chart
- When to use:
Looking at changes in composition over time
What to watch out for:
What to watch out for:
Significant changes in sample size
Example: results from an annual survey to students asking them to what extent they agree with different types of statements (Likert scale).
Line Chart
Line Chart
- When to use:
Tracking data points over time
What to watch out for:
What to watch out for:
Gaps in data cause distortions
Not having enough measurements
Example: data showing the number of students that are enrolled in each grade level on each day of that academic year.
Scatterplot
Scatterplot
- When to use:
Distribution or relationship between two variables
What to watch out for:
What to watch out for:
If there are too many data points, they will overlap and hide points
Example: data that compares the end of year term grades for Grade 1 Math (1-4) against their MAP Math percentile.
Finding the right chart
Finding the right chart
For additional advice in choosing between charts, see this article by InfoGram, or consult the infograph below.
Source: Dr. Andrew Abela's Chart Chooser infograph, via ExtremePresentation.com
Page updated
Report abuse