Get Curious About Your Data

by Barbara Reilly, SVP, Science and Research

As this is my first blog for Morehead, I will take a moment to introduce myself. My name is Barbara Reilly, and I am the senior vice president of Science and Research. I spend a fair amount of time harnessing the power stored in our data and looking at data in different ways. Sometimes I use multivariate statistical models that are powerful beyond measure: regression, factor analysis, cluster analysis, and the like. At other times, it's back to the basics for me. And back to the basics means looking at the data in its purest form.

When I was a graduate student, decades ago, one of my professors had a large banner, printed on continuous feed paper and taped up behind her desk that read, “Have you plotted your data? Have you looked at it in multiple ways? Have you explained it, charted it, thought about it, and made sense out of it? Only after you have done that will I talk to you.” At the time, I thought this message was rather harsh, but over the years, I have been constantly reminded of the value of looking carefully at whatever the data is telling me. The data is telling a story, but sometimes you have to look to find it. For this reason, and many others, when it comes to spotting a trend, there is no substitute for three data points. A quick example will demonstrate my point.

This trend looks interesting and up from last year, and then we add a third data point.

Now our thoughts about the large jump have shifted, just with the addition of one extra data point. I can’t tell you the number of times having the third data point makes all the difference. Imagine if Cedar City only surveyed every other year and they failed to survey in 2011. They would be under the false assumption that engagement was uniformly high across the years.

Consider this pattern.

This hospital is doing something right. The data tells a meaningful story, and the trend is easy to spot.

What does this mean for our clients? Track your data. Look for trends. For the metrics you are most interested in, know what the value has been historically. If you survey regularly with us, we will be able to track this for you. We have clients that survey twice a year. These clients are curious about their data, and they know how to use it. They want the extra data point. Are you thinking about skipping a year of surveying? If so, proceed with caution. Chances are you will miss a vital data point. I have never heard a client regret having too much data, but I have heard the opposite many times over. Are you curious about your data? I am.

Several years ago, I had a summer intern (not at Morehead) complete some analyses for me. When he handed me the output, I asked, “What is the data telling us?” His candid reply was, “I didn’t look at the output. I just ran the analyses.” When he realized his response sounded a bit lame, he added, “Did you ask me to look at the output, too?” His question was so disappointing. It told me I didn’t have a curious intern. If you are going to hang around data, it makes sense to be curious about the data you have.

Don’t you imagine Facebook, valued as a multi-billion dollar empire, has many data analysts who are curious about their data?

A recent article in The New York Times, “The Age of Big Data,” emphasizes the value of having good data and knowing what to do with it. The article highlights research completed last year by Erik Brynjolfsson, an economist at Massachusetts Institute of Technology’s Sloan School of Management, and two colleagues that suggests data-guided management is spreading across corporate America and has started to pay off. They cite a study consisting of 179 large companies that demonstrated adopting “data-driven decision making” achieved productivity gains 5 percent to 6 percent higher than other factors could explain.

So, get good data. Get curious about the data you have. If you are not sure what your data is telling you, just ask us. We have lots of curious people at Morehead.