Nicole contacted me with an email subject line: Thank you for the Data Viz Checklist! and she said she’d been using the checklist to overhaul some of her organization’s data visualizations. I’ve invited her here to showcase her before and afters.

I’m Nicole Huggett and I recently transitioned into a healthcare position where I was introduced to a new lingo of data visualizations – Pareto Charts, Run Charts, and Control or Shewhart Charts. It didn’t take long for me to realize that Pareto charts are essentially just compound column and line charts, a run chart is just a line chart with a median line, and control charts are line charts that also show mean and special standard deviations. Once I got used to these new (and super useful!) tools, there was something I noticed…

They were all really ugly! No color, lots of vertical text, plenty of unnecessary tick marks, you get the picture. Naturally, I set about ensuring that any data our organization presented was done so in a visually appealing, easy-to-read way. So if you work in healthcare quality (shout out to CPHQs! I’ll be testing this year), please help me revolutionize the way the healthcare industry displays data! If you’re familiar with Stephanie’s work and methods, then none of the Excel Ninjery below should be new. So here’s what I’m doing within our organization.

Pareto Charts

Pareto charts are generally used when planning an intervention or addressing common causes of issues. The columns show the occurrence of an event, while the line shows cumulative percentage. The biggest change I make in Pareto charts is not using a secondary axis. Because we deal with such a large number of clients in our organization, percentages are generally more helpful than raw numbers, and it also eliminates the need for the second y-axis. Of course, in meetings I always have raw numbers on hand in case we need those for planning purposes. Since these tools are new to our organization, I also spend time at the beginning of meetings to introduce the tool, what it is used for, and how to interpret it.

Here is how my organization used to show Pareto Charts

ParetoBefore
Source: http://asq.org/learn-about-quality/cause-analysis-tools/overview/pareto.html

and now we show something better

ParetoAfter

  • No unnecessary lines (axis lines, graph border)
  • Emphasized with color, all other data are muted with gray
  • No secondary y-axis
  • Helpful title and subtitle
  • I opted not to directly label the values of the columns here, as the takeaway message is about the fact that the majority of reports were related to 3 categories.

Run Charts

Run charts (aka our old friend line charts) are very useful tools for trending data over longer periods of time. There are a set of pretty easy rules that go along with run charts to tell when data are significant; these are called signals. Although I normally hate using data markers, they are helpful in run charts. I made them a touch smaller so the graph appears more polished. Aside from the normal formatting I would apply to a line graph, the main thing I always do is emphasize where a signal occurs in the data. Sometimes you’ll see these circled, which also works. I prefer color because it keeps unnecessary bulk out of the chart. If applicable, you might also indicate where a certain intervention initiated.

Before

Source: Provost, L. & Murray, S.K. (2011). The Health Care Data Guide: Learning from Data for Improvement.
Source: Provost, L. & Murray, S.K. (2011). The Health Care Data Guide: Learning from Data for Improvement.

After

RunAfter

  • No unnecessary lines (grid lines, tick marks, graph border)
  • Minimized data markers
  • Emphasized important points with color, all other data are muted
  • Deleted y-axis label
  • Helpful title and subtitle

Control (Shewhart) Charts

Control Charts are the most challenging charts for us. There are several different kinds, and the one used depends on the format of data. Special software is generally recommended to avoid using complicated formulas. Our organization uses SPSS, which includes several quality control visuals, including Control Charts as well as Run and Pareto Charts. This is wonderful because I don’t have to worry about lengthy Excel formulas. Although SPSS graphs are more challenging to customize than Excel graphs, once we have the calculations from SPSS, it doesn’t take much effort to recreate the charts in Excel.

To be fair, not all Control Charts are as busy as the example shown here; this is the example used on the American Society for Quality website.

Since the example Control Chart has little information, the remake is from a presentation that I recently developed. These graphs can contain a lot of data, so I tend to limit what I present depending on who is in the audience to prevent information and visual overloads.

Before

Source: http://asq.org/learn-about-quality/data-collection-analysis-tools/overview/control-chart.html
Source: http://asq.org/learn-about-quality/data-collection-analysis-tools/overview/control-chart.html

After

ControlAfter

  • No unnecessary lines (grid lines, tick marks, graph border)
  • Showed unit of measurement on y-axis
  • Meaningful data labels
  • Showed patterns in color, all other data are muted
  • Helpful title and subtitle

Though these were new names, they were familiar charts and I was able to help my organization show our data more effectively. Have another great example of a Control, Run, or Pareto chart? Link to it in the comments!

Learn something new?

Share this helpful info with a friend who needs an extra perk today or post it to your social where your third cousin can benefit, too.