Understanding Statistical Process Charts


How do we know a change is an improvement? Our star Data Analyst Laura Jackman guest blogs on SPC charts.

Statistical Process Charts (SPC) charts are a tool used in improvement science to give an indication of variation within a process or a system over time. They help to see what effect (if any) did your efforts have on the process performance and ultimately to answer the second question of the Model for Improvement “How will we know a change is an improvement?”

SPC charts go by several different names but they are all essentially the same thing, a standard run chart, showing data over time, but with the inclusion of Control Limits. There are several different types of SPC charts depending on the type of data set you are looking at, but they all contain a few key components; control limits, a mean centre line, and at least 15–20 data points.

What do SPC Charts show?

Control Limits

The control limits allow us to say, with more certainty than in a run chart, what constitutes common cause or special cause variation ie, what is normal performance and what is not.

The limits are based on sigma limits from the centre line (similar to standard deviation) & this varies for each type of data set.

Control limits are extremely useful when trying to predict how the system may perform in the future. If the system remained stable (no special cause) we would expect data points to fall within the control limits in the future.

Mean Centre Line

All SPC charts have a centre line which is based on the mean of the data. This is different to a run chart which uses the median formula to calculate the centre line. By using the mean and not the median, all data points are taken into account and used when calculating the centre line and control limits.

The Data

Most importantly control charts help to show data over time. Each point represents one data point and is displayed by time or sequence from left to right.

There are 5 key rules, based on patterns within the data that indicate Special Cause. If one of the rules is spotted within your data, it means the chance of this happening is so small that it is statistically unlikely to have occurred without outside influence.

Top Tips For SPC

  • Make sure you understand the measure you are plotting. Being about the centre line isn’t always a good thing, understanding which direction the data will go if there is an improvement.
  • Always Investigate – When you spot Special Cause or anything unusual in your data, always investigate. An astronomical point might be a yearly occurrence in your system (ie winter pressures), but it might indicate that something unusual has happened. You won’t know until you investigate; don’t make decisions based on your data without doing so.
  • Annotations – Once you’ve investigated your data, the next step is to add annotations. It makes it the chart quick and clear to understand and adds another level of knowledge to your data.

Want to learn more? Join us at our How to Understand Your Data Workshop! Find out more and book here.

What do you think?

Leave a comment below

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