Good data analysis allows you to make smarter decisions faster. Here at Minitab, we are constantly striving to make it easier for you to analyze data and communicate those results so you can keep your projects moving forward. Case in point: the Variability Chart in Minitab.

The Variability Chart makes it easy to identify sources of variability in your data; whether you are using it as a preliminary analysis tool or in a final report to demonstrate the primary sources of variation.

# Example 1: Communicating the Results of a Clinical Study

Consider a clinical study where food manufacturers investigated flavor differences across six products. The study was repeated on Day 2. Here is a sample showing how the data are arranged:

We start by selecting the Variability Chart under Stat > Quality Tools > Variability Chart

The vertical axis represents the flavor score and the horizontal axis contains both the product codes and days.

It’s easy to see that the average flavor scores, connected by the lines, are higher for LHA .50% w/ DRN 4 product across both days. Additionally, the Variability Chart makes it easy to see that the variation in flavor scores for the Current product was much smaller compared to all others.

If the goal is to maximize flavor, one reasonable conclusion is that while the Current product delivers a more consistent flavor, the LHA .50% w/ DRN 4 product will require some work to reduce the variation in those flavor scores.  The Variability Chart is a great choice to demonstrate the results of this study because it shows both the higher average scores for the LHA .50% w/ DRN 4 product and the small amount of variation for the Current product in a clear, concise view.

# Example 2: Visualizing Data Collected from a Medical Device Manufacturing Process

Now consider that we have collected data from a process where we measure the infusion amount for a medical pump. We have data collected at four different pump speeds across two delivery lines using two different material types. Here is a sample showing how the data are arranged:

As in the first example, we found the Variability Chart dialog at Stat > Quality Tools > Variability Chart… Here we also selected the Standard deviation chart:

Visualizing these three factors in a single graph is easy with the Variability Chart:

We can see that the overall average is about 21.5, noted with horizontal line across all levels.  The variability within the pump speed of 1300 shows that the results are below the overall average. We can also see that the infusion measurements were consistently lower for Material Code 9978 across lines One and Two. It is also easy to see that the variability within Pump Speed 1400 is much higher compared to the other pump speeds.

The Variability Chart also provides an option to create a standard deviation chart of the data – here it is easy to see that the variation when the pump speed is at 1400 is higher across the levels of Line and Material Code.

### Get an Overview and More In-Depth Information

The next steps in this data analysis might be to perform an Analysis of Variance to quantify the observed variation, or to design an experiment to better understand the process. In this example, we had 3 factors. The Variability Chart allows you to visualize up to 8 factors at a time.

So, if you have data collected across multiple factors, try the new Variability Chart in the Quality Tools menu. Visual analytics just got a little easier.