Here is another great article from our friend Adam Stoehr of the National Quality Institute. The National Quality Institute, http://www.nqi.ca, are a Canadian partner of BPIR.com. Adam’s article describes a personal use for run charts (pardon the pun) and their power in revealing trends.


Graphing Marathon Measures 2 – Run Chart Run


Adam Stoehr, MBA
Vice President, Education, National Quality Institute

“Run Forrest Run”… Run charts can take data over time and tell interesting stories. The chart below is a simple run chart that shows my weight in pounds from 2008 to 2011.  I have lost about 50(ish) pounds since January 2008. My strategy has been the classic “eat less and move more”. The eating less part is always tough for me so I’ve adjusted the strategy to be “move more and eat decently.”  To keep up the move more side, I have run in 7 organized races including a half-marathon over the last 2 years.  In order to finish these races you need to keep a pretty rigorous training schedule. It’s been lots of fun and I still have a long way to go. I’m not looking for a medal by the way (other than the ones you get when you finish the races).  I just thought using my marathon training/weight loss data (in pounds) would be a good way to remind people of the correct use of some simple charts.
 
 
in our work lives charts and graphs are useful to display data in a way that helps identify a gap or a trend if one exists. Charts can also be used to highlight relationships.  Generally a picture of your data says a thousand words or numbers. Over the next few Quest for Excellence newsletters, I’ll cover a bunch of charts and graphs using data from my marathon training.

Before we draw some graphs, let’s set some general ground rules for chart creation.

  • Rule 1: Make sure you have a clear purpose for your graph and that it will convey an important message.
  • Rule 2: Try to use simple pictures to depict complex data.
  • Rule 3: Try to make your data talk and tell interesting stories.
  • Rule 4: Remember to adapt your graph to suit the audience.
  • Rule 5: Don’t be afraid to experiment with various options and graph styles.

In this article we’ll focus on the “run chart” which is one of the simplest and most-used charts available.

Run Chart

A run chart is a graph that shows changes and trends over time. It can help you recognize patterns of performance in a process and document changes over time. It should be used to show trends at a series of progressively increasing or decreasing points and shifts in consecutive points that fall above or below the average. I don’t think my daughter has tried a run chart yet (like she had with the bar chart reviewed in last months newsletter) but it should be kept simple as well.

Basically, run charts have two rules.

  • Rule 1: The lines should represent data over time
  • Rule 2: The lines should represent data in chronological order

Figure 1:

In Figure 1 we have data over time from May 2010 to July 2011 compared to my goal.  I have a monthly goal to run at least 100km. According to the chart, since May 2010 I have achieved my goal in every month except for January 2011.

It’s important to point out that the data does not represent a snapshot in time like we saw with the bar chart in Figure 2 from last month’s newsletter. The best way to explain the over time (run chart) vs. snapshot in time (bar chart) is that the data in the past on a run chart will not change. For example the number of KM’s that I ran in May 2011 will never change because May 2011 is in the past.

Figure 2:

Conversely for a snapshot in time, the categories would vary every time you draw the chart like in Figure 2. If I drew this chart again in 2 months all of the bars would be different (unless of course I quit running for some reason).

Figure 3:

Figure 3 represents my monthly weight in pounds compared to my average weight since January 2008.  Run charts are most useful to see trends over time. They allow you to easily see patterns in data.

The first pattern to look for is if there are 5 or 6 points in a row that steadily increase or decrease. In figure 3 you can see this in both directions. For example from May 2008 to Jan 2009 you can see 7 points in a row that are increasing. Luckily for me (considering what I’m measuring) between July 2010 and March 2010 there is the opposite pattern of weight loss.

The second pattern to look for is if 5 or 6 points in a row are on the same side of the average/goal. This is evident in figure 3 between November 2009 to present. If either of these patterns are present then it’s a pretty good sign that your process is changing (hopefully for you in a positive direction). So my running schedule and my weight loss have been trending in the right direction according to my goals.

Figure 4:

Another pattern to look for is significant shifts, bunching, or cyclical patterns. In Figure 4 I’ve taken the same data as in Figure 3 and cut it up year over year. We can see a couple of patterns emerging. January and February seem to be bad months for my weight as well as July and August each year. May/June and November/December are generally good months. 2008 had the most significant shifts from January to April – if this type of swing kept happening, it would be something to investigate.

With all charts it should be pretty obvious what story you are trying to tell. You should be able to make one or two statements about the data without thinking too much.

At the end of the day you want the charts to tell stories. For example looking at the four run charts we can make the following fact based statements.

  • Figure 1: I have been consistently meeting or exceeding my goal of running 100 km a month since May 2010. I only missed the goal once in January 2011.
  • Figure 2: Sunday is my most active day with 65 total runs and Saturday is my least active day with 33 total runs.
  • Figure 3: My weight has been trending downward and has been consistently below my 3 year average weight over the last 6 months.
  • Figure 4: January/February and July/August have been my most challenging times of year to keep my weight down. April/May and November/December have been my best months for weight loss.

Common uses for bar charts include:

  • Any data over time
  • Revenue trends
  • Stock prices
  • Process performance
  • Etc.

In next month’s Quest for Excellence we’ll look at Scatter Diagrams to see relationships between data.


Receive Our Newsletter

Keep informed on best practice research and business improvement events from around
the globe - be the first to read our Best Practice Reports



Share
Posted by: -
Author: -
Source : -
BPIR Category : -
Latest News Categories: Latest News, Quality Management
X
X