Dashboards and Data Viz

On the road, I often see some amazing dashboard designs. Sometimes when we make charts and graphs though, we combine too much information into one chart. I regularly see charts that have measurements of many items against one X axis. See the chart below for an example of this – the number of people in a city and the amount of rain (in centimeters) the city has had in the last year.

We can see that the number of people in the city greatly exceeds the number of centimeters of rain.  When we put this into a chart, it looks like this:

Rain population bar chart

Left out of the rain, it makes it look like these cities are all in arid climates and didn’t receive any rain at all.

The above chart renders the data as it should, and displays the information; however, this chart really does not do a decent job showing if there is a correlation between the amount of rain a city receives and its population. This makes a classic case for a need of the iDashboards Interactive Intelligence feature. I like to tell my course participants that if a scale bias is introduced into the dashboard decision-making process, the consequential choice might be a poor one if the decision-maker sees a chart like this. What would happen if emergency water rationing went into place in a city that was actually experiencing a flood year as a result of this chart?

Instead, we apply chart object recycling, and we can now see the same chart, but with one “Y.”

Read next: Fighting Cognitive Bias with Data

We took the chart above, and in the middle frame created these charts:

Left: X of City, Y1 of Population (Hidden), Y2 of CM of Rain (Hidden) Y3: an expression to compute population numbers to rain
Middle: X of City, Y1 of Population
Right: X of City, Y1 of Rain in CM

Now this dashboard tells a story! We can see that even though Denver received the least amount of rain, it also has the highest population, indicating that this city is in real trouble. We can make that correlation somewhat easily by looking at the chart at the right to show rainfall, and the chart at the left that shows us how many people are counting on a single centimeter of rain. A single chart tells facts, but a whole dashboard tells a story.
Dashboard on!

Brad Hines – Sr. Technical Consultant, iDashboards


Your email address will not be published. Required fields are marked *