Back when I was a student at RPI, I took numerous courses as part of the requirements for my degrees in Management Engineering. Most of them offered a good foundation in aspects of business which proved very useful in the years to follow. But only one of them had a massive and lasting effect on my worldview: Statistics.
Statistics? Why Statistics? Probably part of the reason was that it took me a while to get the hang of it. My professor, Dr. George Manners, had a wonderful Georgian accent which was quite different than those of us from New England. Then, he gave us a test about mid-semester and virtually everybody did poorly. Then he did a surprising thing. He got up in front of the class and said that obviously he hadn’t done a good job in teaching the material. So he would cover the material again and give a re-test. Wow! That made me take notice. So I paid chose attention and discovered there was magic in the art of choosing a sample size, paying attention to aspects such as normalizing the data and running statistical tests. I did pretty well on the re-test and I also struck up a friendship with the professor which extended beyond the point when I’d graduated. Later, he taught me Marketing, another course which has had a long standing impact.
Getting back to statistics, I never looked at data the same way after taking this course and a successor course in the design of experiments. I learned a valuable lesson that many politicians and other public figures have not learned, which is that a sample size of one, or even a few more data points than that, is basically statistically meaningless. We also learned about the margin of error. If a political poll is conducted and one candidate leads another by 3 percentage points, but the margin of error is six percentage points, it’s basically a statistical tie, not a conclusion one can bank on. So when the press, or politicians tout a particular result, I pay close attention to the statistics behind the conclusion and decide whether the suggested results pass muster. Frankly, this is a skill which can be valuable to anybody who wants to be able to follow any news with statistical information or other uses of data and make sense of it.
Now, we are all living in the Covid-19 era. And for many of us in the public, the people we can trust the most with their statements are the ones who understand data and are able to present conclusions based on the statistical data, rather than on gut feel. This is closely related to looking at news stories or public statements, and determining whether the information is evidence based, using techniques such as the scientific method and peer-based reviews, or anecdotal, and therefore not to be trusted. Intuition can be very powerful, but not nearly as useful unless it is also combined with a data driven outlook for these types of challenges.
I teach an Introduction to Business course at Northeastern University and we have the students conduct surveys as part of their business projects. Using survey and statistical tools such as those offered by Qualtrics, they take surveys and test conclusions on ideas for their business models. The surveys carry much more weight for me if the survey size is at least 30 and ideally much more, and if the respondents have a demographic profile which fits the proposed target market for their products or services. This is hardly rigorous statistical work, but it is much more valid, and convincing for a reader, than just stating opinions and not having any research to back it up.
So statistics and big data are having a moment. It’s a good reminder that certain skills, such as statistics, investment planning and effective written and oral communications, can offer value that goes well beyond the classroom and better equips all of us to be effective and informed citizens and consumers. That’s why I encouraged both of my sons to take college level statistics courses and add this particular skill to their life toolkit. So the next time you hear a news story or listen to a public figure, think about whether the opinions stated are backed up by data. It’s a basic skill from which we can all benefit.