1. Learn home
  2. Quality Improvement Zone
  3. QI tools
  4. SPC charts

SPC charts

Add to favourites

An SPC chart helps teams monitor process performance and identify variation over time.

What is it?

A Statistical Process Control (SPC) chart is a tool that helps teams understand variation in a process over time. By distinguishing between common cause variation (the natural variation within a process) and special cause variation (variation caused by something changing), SPC charts help determine whether a process is stable, whether changes are leading to improvement and whether performance is likely to be sustained. 

SPC charts can help answer important questions such as:
•    Is this process stable and predictable? 
•    Is performance improving, staying the same or getting worse? 
•    Has something changed within the system? 
•    Is the process capable of achieving the level of performance we need?

An SPC chart includes: 
•    a centre line, which represents the average (mean) of the data 
•    upper and lower control limits, which show the range within which future data points are expected to fall if the process remains stable

The control limits are calculated from the variation within the data. When a process is stable, most future observations are expected to fall within these limits. 

There are different types of SPC chart, depending on the type of data you are working with. The most common ones are: 

•    XmR (or I) chart 
Used for individual measurements and suitable for many improvement measures.

•    P chart 
Used for proportions or percentages. 

•    C chart 
Used for counts of events. 

•    U chart 
Used when counts are expressed as a rate. 

•    X and S chart 
Used for grouped measurement data.

In most improvement work, an XmR chart is a good starting point and will be suitable for many measures. 
If you would like to learn more about different SPC chart types, you can find further detail in The Health Care Data Guide by Provost and Murray.

The image shows an example control chart (X chart) with labels identifying its main features. Data points are plotted over time and connected by a line to show variation in a measured process. The horizontal axis represents time and the vertical axis represents the measured value. A centre line shows the mean or average value. Upper and lower control limits are displayed as dashed lines above and below the mean, indicating the expected range of variation. The chart demonstrates how control charts are used to monitor process performance and identify variation over time.
Why use this tool?

SPC charts provide additional insight beyond a run chart by calculating control limits and helping teams understand process stability and capability.

SPC charts can help you: 

  • understand how your data is changing over time
  • distinguish between common cause or special cause variation (see understanding variation page)
  • understand whether your process is stable and predictable, or unstable and unpredictable
  • understand whether your process is capable of meeting the level of performance you need
  • monitor whether improvements are sustained over time
Where does this tool fit in the improvement journey?
Measurement Icon

 

This tool is relevant at this stage of the quality improvement journey.

The quality improvement journey
How to use it

SPC charts help teams understand whether observed changes are likely to reflect genuine changes in the system or simply normal variation. The process below provides a practical approach to creating and interpreting SPC charts.

1. Define the measure

Choose a measure that will help you understand the performance of the process you are interested in. This may be an outcome, process or balancing measure. Ensure the measure has a clear operational definition so data is collected consistently over time.

2. Select the appropriate chart

Choose the SPC chart that matches the type of data being collected. Different chart types use different calculations and assumptions, so selecting the correct chart is important.

If you are unsure, an XmR chart will often be appropriate.

3. Collect sufficient data

SPC charts require enough data to calculate reliable control limits.

For an XmR chart, at least 20 data points are required. With fewer data points, control limits are unreliable, and conclusions about stability and capability may not be valid.

If fewer than 20 data points are available, consider using a run chart until more data has been collected. Run charts are often a useful starting point, while SPC charts provide additional insight once sufficient data is available.

4. Create the chart

Plot the data in time order and calculate the centre line and control limits. SPC charts can be created using Excel. There are also commercial add‑ins, such as QI Charts or QI Macros. 

5. Apply the SPC rules

SPC rules help identify signals that suggest something has changed within the process. When these patterns are present, they indicate that special cause variation may exist

6. Understand process stability

Check the chart for signals of change using the SPC chart rules below. If one or more SPC rules are present, this suggests special cause variation, meaning something has changed in the process.

If none of the SPC rules are present, the variation observed is likely to be common cause variation, suggesting the process is operating in a stable and predictable way.

In a stable process, future performance is likely to remain within the control limits. The control limits therefore provide an estimate of what the process is currently capable of delivering.

For some chart types, such as P charts, U charts and X and S charts, the control limits may vary over time if denominators or subgroup sizes change.

7. Use the findings to guide improvement

Use the SPC chart to understand how the system is performing and whether changes are leading to improvement. The chart should support learning and decision-making rather than simply reporting performance

SPC chart rules

SPC chart rules (image 1)
SPC chart rules (image 2)

 

Use SPC charts to support improvement 

When creating your first SPC chart for a measure, particularly when historical data is available, calculate the mean and control limits using all available baseline data.

During improvement work, it is often helpful to freeze and extend the baseline mean and control limits. This provides a stable reference point against which future performance can be compared and can make improvement easier to detect.

This should only be done when there is no evidence of special cause variation in the baseline period.

Rather than immediately recalculating the chart when improvement is detected, consider whether enough data has been collected to demonstrate a sustained change in performance. Once there is sufficient evidence that the process has fundamentally changed, the chart can be recalculated using the new phase of data.

If historical baseline data are unavailable, use your judgement to determine whether the first 20 or more data points represent a suitable baseline period, taking into account when changes occurred and what is known about the system.

For more information about common cause and special cause variation, see the Understanding variation page.

Resources to download