Measurement and Data Collection
Posted on September 10, 2005 Comments (4)
This is my response to the Deming Electronic Network message (the site is dead, so I removed the link) on measurement.
I find it useful, to assure that data collection is a wise use of resources, to ask what will be done with the results. If you don’t have an answer for how you will use the data, once you get it, then you probably shouldn’t waste resources collecting it (and I find there is frequently no plan for using the results).
I have found it helpful to ask: what will you do if the data we collect is 30? What will you do if it is 3? The answer does not need to be some formula, if 30 then x. But rather that the results would be used to help inform a decision process to make improvements (possibly the decision to focus resources in that area). I find, that asking that question often helps reach a better understanding of what data is actually needed, so you then collect better data.
I believe, it is better to focus on less data, really focus on it. My father, Bill Hunter, and Brain Joiner, believed in the value of actually plotting the data yourself by hand. In this day and age that is almost never done (especially in an office environment). I think doing so does add value. For one thing, it makes you select the vital few important measures to your job.
But it is very difficult for anyone to actual suggest plotting data by hand: they must be very secure in their reputation (or maybe a bit crazy), because it seems to be a hopelessly outdated idea that paints you as the same. My appeal, within the Deming context, is that the psychology of plotting the points yourself is qualitatively different from letting the computer do it. Plotting the data yourself serves to lift the data that you plot out of the sea of data that we find ourselves inundated with and gives you a deeper connection to it. You would not plot all the data that you use by hand; just the most important items.
Curious Cat Management Improvement Connections