Three Things You Need to Know About Data Analysis
Sports fan or not, you’ve probably watched, at least once in your life, some type of game or match (think Super Bowl, World Cup, Wimbledon etc.) where the ruling on the court or field was questioned. The play then is reviewed, usually from many different camera angles. You might see it from the overhead view followed by the field view and finally a zoomed view of the exact same play or volley. After reviewing all the angles, the situation is clear. The ground caused the fumble or the player stepped out of bounds. Think of data analysis in the same way. It provides you with various camera angles into the situation so you can determine what is going on. Do we have more defects on Mondays than other days? Is the XYZ region doing something different than the ABC region? Does it take longer to onboard a residential customer than a commercial customer? The list of questions that can be answered with data is as extensive as the comfort level of the leader asking the questions.
Data collection and analysis has never been easier. Technology has afforded the collection of information in ways few predicted. There is no excuse for failing to raise your level of awareness when the situation dictates. With three basic data analytic concepts, you can make well informed decisions to tip the scales towards success.
1. Accurately collect the data.
I think we all know that data can be skewed to support any view or bias of the situation. If that is all you are trying to do, I would not waste the time. If I set my bathroom scale to something less than zero only to claim I’ve lost weight when I stand on it, I’m just fooling myself. But if I really want to know if I’ve lost weight, I’ll make sure the scale is accurately calibrated to give me the true weight. This same process is important for any collection of data or measurements. Performing a Measurement System Analysis helps determine if the data is accurate, repeatable, reproducible and stable over time. If there is an issue with the integrity of the data, you need to look at the Operational Definition which is a clear and concise instruction of how something should be measured, recorded and analyzed. The goal is to make sure data is collected and interpreted in the same way so conclusions are accurate.
2. Collect both inputs and outputs.
There is a statistical expression that you probably learned at some point on your educational journey signaling that y = f (x). If you were like me, you didn’t realize at the time how important this equation is to problem solving and goal fulfillment. So allow me to convey that this simple statistical expression is a key part of improving results. It is now a part of my daily thought process.
As a quick review, y is the statistical symbol for an output. It is something measurable that you want to change. In our weight example, the “Y” would represent weight loss. “X” is the statistical symbol for inputs or factors that generate the “Y” output. Our equation tells us that Y is a function of all the factors. In this example, these factors are going to be things like caloric intake, exercise and water intake, just to name a few. So the first step is to identify all the factors that impact the result. Think of these factors as the seeds that must be properly planted in order to grow the harvest or result you are looking for.
3. Look for the “vital view” versus the “trivial many.”
Once you collect both the outputs and inputs, you’ll want to group them and look for something known as a Pareto Principle. This shows you which factors have the largest impact to the results. You can see these factors by displaying your data in a Pareto Chart (see Figure 1). For example, if you are trying to reduce the number of technical support issues the IT team deals with, you will want to look at those that occur most frequently. It might be password issues, or permissions to load new software to a company computer. Focusing on those issues that occur most often will take you a long way to the goal of reducing the calls.
Looking at accurate data through different types of charts and graphs will provide the awareness you need to solve problems and make good decisions.
About the Author
Anne Foley, PMP, MBB, CSSBB has been a quality leader for 20 years. She has relied on data analytics to solve problems and make decisions in various leadership roles. Anne is a senior trainer and consultant for Corporate Education Group. She is also the author of The Passages to Peace (a novel), Vitabook: Healthy Supplements for the Mind of a Graduate, and a frequent contributor to Project Management, Lean Six Sigma and other various publications. Anne has a Bachelor’s of Science degree in Journalism and Mass Communications from Kansas State University.