How predictable are you?

Data science is capable of amazing predictions, but communicating the power of a great model can be tough. All the hard work, insight and importance can be lost on someone who just doesn’t think about these things on a regular basis. Not only can big accomplishments go under-appreciated, but people’s everyday alienation from data science can also make them question its validity.

For example, if I told you quantum electrodynamics can predict how electrically charged particles will interact and share photons with an accuracy of up to 0.0038 parts per million, a lot of people would say “so what? I don’t get it, and even if I did, it’s hard to believe it really works.”

That’s why it’s so critical to break down and communicate data science research in ways non-data scientists can appreciate. It isn’t that non-data-heads don’t care or aren’t smart. Of course they are. Rather, as data scientists it’s on us to connect the dots for them. We need to help them digest what’s being done and, most importantly, why it matters. This is hard, we know. It’s especially tough when empirical knowledge conflicts with people’s intuitive sense of how the world works (as it usually does). The science may be right, but if it doesn’t feel right, people can become disengaged.

That’s what we love about this video. While it wasn’t made (we’re guessing) explicitly with data science in mind, it’s a fantastic example of how a little empirical knowledge can lead to surprisingly accurate predictions. In a simple, elegant way, it connects the power of quantitative prediction with something anyone can really feel.

This Post Has 5 Comments

  1. Deborah

    This is awesome – tried the experiment 2x! I am passing this link everywhere. Thank you, Thomas!

  2. Mark

    I couldn’t agree more with the first paragraph of this article. I also like the terminology Data Scientist. It is an accurate description of mathematical modeling of physical entities.

  3. Robert

    Everybody is predictable once you understand their psychographic profile.

  4. Gordon Bergelson

    The art of taking complex data and relating it in a concise, understandable and relevant way has always been important no matter
    what the field of endeavor.

  5. Eula Blevins

    The use of logic and the close examination of evidence are necessary but not usually sufficient for the advancement of science. Scientific concepts do not emerge automatically from data or from any amount of analysis alone. Inventing hypotheses or theories to imagine how the world works and then figuring out how they can be put to the test of reality is as creative as writing poetry, composing music, or designing skyscrapers. Sometimes discoveries in science are made unexpectedly, even by accident. But knowledge and creative insight are usually required to recognize the meaning of the unexpected. Aspects of data that have been ignored by one scientist may lead to new discoveries by another.

Leave a Reply