In this seventh article in a series on internal quality control, Stephen MacDonald moves on from the monitoring of process with rules, to focus this month on the various ways that control can be charted.
In this article we begin with a quote: “...people do what is inspected, not what is expected”.1 In my opinion, in no other aspect of our work is this more evident than in quality control (QC) charting. There is a wealth of information that could be used. Charts are produced within commercial software solutions and integrated within laboratory hardware. They produce restricted analysis based not on what is useful, but what is easily understood. Limiting QC charting to analysis of individual results as a function of time, or run number, does not do it justice. The art of QC charting is being lost.
Describing all available charts would become very dull, very quickly. We will focus on the common histogram, density plot, boxplot and (arguably) the most important – the simple line chart. How the charts are constructed should be familiar to all. However, the data used, and the impact on interpretation, requires closer inspection. The purpose here is to show that an integrated use of different charting techniques is as relevant today as it was in the days of Levey and Jennings.2 In the next article we will discuss even lesser used techniques from the days of manual charting that remain relevant today, and also include some new players in the field.
Modelling our data
Log in or register FREE to read the rest
This story is Premium Content and is only available to registered users. Please log in at the top of the page to view the full text.
If you don't already have an account, please register with us completely free of charge.