elle
commented
7 months ago 2: Choosing an effective visual
12 visuals most used:
- Text - when you have a number or two that you want to communicate
- Tables - people read tables row by row. So not good for live presentations.
- Interacts with our verbal system
- Suggestion: fade the table itself into the background, so the data pop out
- Heatmap - visual cues. In addition to numbers, we now have leverage coloured cells to convey the relative magnitude of the values → reduce mental processing
- Suggestion: use conditionals in a spreadsheet. Consider including a legend
- Graphs - interacts with our visual system - faster in processing information
- 4 categories: points, lines, bars, and area
- Scatterplot - useful for showing relationships between two things. Used more in scientific contexts.
- Can indicate the average on the graph to draw our eye (and maybe even a dotted line for even better clarification of what is below and above the average
- Lines - used for continuous data. It implies a connection between the dots, usually a connection over time
- Suggestion: you can use multiple lines in different colours to indicate the min, max, and average value (being the average is in a more prominent colour. Or even use a lighter background colour as the area for max and min, with the line indicating the average.
- Slopegraph - when you have two time periods or points of comparison and want to quickly show relative increases and decreases or differences across various categories between the two data points.
- Good use case: employee survey over time
- Might take longer to set up because they are not the standard graphs included in graphing applications
- Suggestion: we can draw attention to one of the relationships by giving it a more prominent colour
- Template:
https://www.storytellingwithdata.com/blog/2013/11/slopegraph-template
- Bars - common so sometimes avoided for this reason. But should be used because the audience is familiar with them.
- Suggestion: use proportional height, as to not mislead the audience. Also start the axis with zero (zero baseline!)
- Graph axis vs data labels: when graphing data, a common decision to make is whether to preserve the axis labels or eliminate the axis and instead label the data points directly. When making the decision, consider the level of specificity needed.
- If the focus is on high level trends → deemphasise the labels
- If numerical data is important → label the data points directly
- Ethics and data visualisations - be careful of misleading the audience and thus damaging your credibility
- Consider width of bars → should be wider than the width of the gaps
- Vertical bar chart - ?
- Stacked vertical bar - meant to compare totals across categories and also see subcomponents within a category.
- Can be visually overwhelming. Hard to compare beyond the the bottom series
- Better if each column summed to 100% → then decide if it makes sense to include the absolute numbers
- Example: what my tax money paid for
- Waterfall chart - pull apart pieces of the stacked vertical bar to focus at each subcomponent at a time.
- Horizontal stacked bar - better than vertical, because it is easier to read, and easier to add category names
- Logical ordering of categories
- Numerical ordering is easy
- But if natural ordering isn’t present, consider which order to present the categories makes sense → to ease the interpretation process
- For example using a decreasing order of importance - from biggest to smallest, or smallest to biggest
- Stacked horizontal bar - useful for scale ranges
- Example: survey data of Likert scale
- Area - generally avoids them with one exception, visualising numbers with vast magnitudes.
Infographics
Do they really inform?
Good data visualization—infographic or otherwise—is not simply a collection of facts on a given topic; good data visualization tells a story.
To be avoided
Pie charts
- Asking the audience to compare angles and areas.
- 3D Pie charts are misleading - parts at the front and bottom seem bigger, and top elements further away seem smaller. Don’t use 3D!
- Even flat pie charts are hard to read and require data labels
- Instead → horizontal bar chart
Donut charts
- Asking the audience to compare arc lengths.
3D
- Never use 3D! And definitely not for single dimension charts!
- Adds unnecessary elements: sides and and floor panels
Secondary Y axis
- Complicated!
- Use two separate graphs instead
Summary
What do you need your audience to know? Then choose a visual display that will enable you to make this clear.
If you’re wondering What is the right graph for my situation?, the answer is always the same: whatever will be easiest for your audience to read.
Book: Storytelling with Data by Cole Nussbaumer Knaflic
Aiming to read:
Chapters: 2 MC: @cjeffries Notes: @elle
See you all at 12pm AEDT, Feb 7th @ https://blackmill.whereby.com/bookclub
As always, if you'd like a calendar invite and/or access to Slack beforehand, get in touch via gday@blackmill.co.