“All Others Must Bring Data”
To produce a quality product--and continue to do so--you need data.
“In God we trust; all others must bring data.”
—W. Edwards Deming
“It is a capital mistake to theorize before one has data.”
—Arthur Conan Doyle
“When human judgment and big data intersect, there are some funny things that happen.”
Quality is based on a series of facts and statistics collected together and analyzed. To produce a quality product—and continue producing a quality product—you need data. Data you can trust doesn’t simply appear; you need to collect it and then analyze it properly. Luckily, there are several tools available to you for this use. For the sake of this article, we will take a look at three you may or may not be familiar with.
Box and Whisker Plot
Imagine you have multiple machines creating the same product. You want to know which are most effective. An applicable, easy and graphical method for displaying data variation is the box and whisker plot. The advantage of the box and whisker plot—over another helpful tool, the histogram—is that multiple sets of data are displayed on the same graph. Before you plot your data set you need to determine the following numbers:
- Minimum value—smallest value in the data set
- Maximum value—largest value
- Median value—middle number in the range (when set has an even number of values, the median is calculated as the average of the two middle values)
- Second quartile—the value at which the lower 25% begins
- Third quartile—the value at which the upper 25% begins
Once you plot the numbers, you find the range for each set (unlike the frequency of each number, as you do in a histogram) and can easily determine which machine is producing the better product. To download a box and whisker plot template, visit the ASQ Learn About Quality page (asq.org/learn-about-quality).
While keeping track of your organization’s equipment and product is important, you need to collect and analyze data for your processes as well. To do that, try a control chart. The control chart graphs the changes within a process over time and can serve many goals. Use a control chart when you want to:
- Find and correct process problems as they occur
- Predict the expected range of outcomes
- Determine if a process is stable or unstable
- Analyze whether patterns of process variation stem from special (non-routine) or common (built into the process) causes
- Determine whether to make fundamental changes or prevent specific problems
To download a template and learn more about the control chart procedure, visit the ASQ Learn About Quality page (asq.org/learn-about-quality/).
Scatter Diagram (Also Scatter Plot and X-Y Graph)
You are trying to solve a problem. Your team has brainstormed cause and effects and you’ve completed a fishbone diagram. But are you sure your causes and effects are related? Use the scatter diagram and objectively determine the relation between a particular cause and effect.
In “The Quality Toolbox,” (Second Edition, ASQ Quality Press, 2005, pp. 471–474), Nancy Tague explains that a scatter diagram is, “a diagram that graphs pairs of numerical data, one variable on each axis, to look for a relationship. The scatter diagram graphs pairs of numerical data, with one variable on each axis, to look for a relationship between them. If the variables are correlated, the points will fall along a line or curve. The better the correlation, the tighter the points will hug the line.”
The relation between the data is extremely important, so use the scatter diagram to determine whether two effects appearing to be related occur with the identical cause.
The box and whisker plot, control chart, and scatter diagram are powerful, versatile tools to make sure your data collection and analysis occur effectively—and productively.