A common problem with SPC Charts is knowing the right one to pick! To help you make the right choice and ensure you select the best possible chart for your data, we have created an 'SPC Chart Type' infographic guiding you through the decision-making process.
Lets start at the top ...
The chart you use will always depend on the type of numeric data you have collected, so you need to determine if the data is:
Discrete (Counted) or Continuous (Measured)
What is Discrete (Counted) Data?
Discrete data can only take certain values, so it's pretty black or white. For example No. of people on the hospital ward, you are counting whole, indivisible entities! You can't have 3.5 people!
What is Continuous (Measured) Data?
Continuous Data can take any value (within a range). For example measuring the weight of each person on the ward or the distance a patient is able to walk after an operation. This data can be recorded at many different points.
Are you combining observations from multiple subgroups?
When we refer to subgroups we mean a set of measurements that were obtained under similar condition's, during the same period. An example of this would be running the same test, in multiple wards on the same day.
Are you measuring a rare event?
Conventional charts exhibit problems with rare events due to the infrequency of the data recorded. which is why at this stage of the flow chart we need to identify this, an example could be number of medical errors or number of patient falls. Let's face it nobody decides when the next patient fall is going to happen, it is just a rare occurrence that should absolutely be recorded.
How are you measuring the space between rare events?
Either by time (E.g. days, weeks, months in between rare events) or by number of 'Normal' events (e.g. recording the number of admissions between infections (Infections being the rare event) on a ward).
Are the observations binomial?
Meaning the results will only have two possible outcomes: Yes/No, Female/Male, Low/Normal birth weight, Pass / Fail.
Is your sample size constant?
Are you using the same sample size each week throughout the whole project?
Ok, now you have determined which chart you need, why not get a quick overview of the chart you are about to use! Each chart type will give you typical examples of when they would be used and the type of data you will capture, Simple!