The importance of understanding probability

The disconnect between the ubiquity of probability in our culture, especially in relation to health and illness, and the poverty of understanding of this area of mathematics strikes me as highly problematic. If one goal of eduction is to prepare our children (and ourselves) to participate in the world, then a solid understanding of probability should definitely be included.

But, as my opening statement suggests, while the world offers many opportunities to delve into the topic in relevant ways, it offers few guides to understanding. There is a lot of misinterpretation of statistics out there that a good education should equip us to recognize and counter.

Take, for example, the relationship between a particular illness and various factors that seem to be correlated with the incidence of this illness. If we exercise regularly, eat well, scorn smoking, and drink moderately should we feel cheated if we develop heart disease? Does the fact that we might die of heart disease anyway mean that none of these behaviours are worthwhile? Or, in the famous example, if one might smoke all ones life and still live to 97, does that mean that the relationship between smoking and lung cancer is all smoke and mirrors?

No. Most of this health information, and there is lots of it out there, is based on epidemiological studies. Epidemiology is the study of population health. It uses statistics to calculate the correlation between various diseases and various possible contributing factors. It does not determine cause. It determines probabilities, which can also be called risks. Just as individuals do win the lottery, despite the long odds, some smokers will live to 100. And some people who “do everything right” will get cancer.

What most of this health reporting is telling us is that there are correlations, sometimes strong correlations, between certain behaviours or characteristics and certain illnesses. Correlation is not cause. Human bodies are incredibly complex. Despite the enormous advances in scientific knowledge in the past 100 years or so, we still have very little idea of how that complexity works. Because of that, we know little of cause.

We do know that viruses cause the flu. And we know how viruses are transmitted. We only know in the most general terms why some people exposed to a virus will get sick and others will not, and why some will get sicker than others. We know that those things are correlated with factors like age, general state of health, diet, etc. And we can calculate the relative risks. But we don’t know exactly why.

So much of the current health debates seem to misrepresent these basic facts. There is a tendency to try to allocate blame to individuals for their illnesses. There is also a tendency to think that we should be able to eradicate illness. If only we all ate properly, no one would have heart attacks. The fact is that the probability you will die sometime is 100%. The part that is difficult to predict is when and how.

For those who want to learn more about probability, I recommend Chapter 2 of Nathalie Angier’s The Canon or Chapter 7 of Keith Devlin’s The Language of Mathematics. The archived radio programme on the history of probability from the BBC’s In Our Time series, is also informative. Devlin’s chapter also has a lot of history in it. There is also good list of readers for various ages at Living Math.


7 thoughts on “The importance of understanding probability

  1. Excellent post! I do think people mindlessly follow (or ignore) statistical information because they don’t really understand it, even on a basic level. They also don’t understand the basic caveat we were taught in grad. school, “correlation doesn’t prove causation.” 🙂 Hence the prevalence of people manipulating statistics to “prove” anything they like. (I am remembering Mark Twain: “There are lies, damn lies and statistics.”)

    I include myself in this generalization, of course. I understand very little about probability and statistics. Thanks for reminding me to check out Living Math for resources. 😉


  2. ‘ear ‘ear. I get driven bonkers by the misreporting of risk factors in the media. The publicity that accompanied the publication of a study on HRT a few years ago was an excellent example. No-one seemed to be able to report it correctly, and the resulting panic meant that several excellent and useful drugs were withdrawn from the market here in Aus because women were to afraid to use them even for short periods (no-one had suggested they were harmful in healthy women for short periods) and the company decided that it wasn’t worth continuing to import them.


  3. Probability and statistics is a subject that I really wish was addressed in greater detail in cross-curricula studies. I have always had difficulty articulating this to my students, in a understandable way for their level, yet the terms are thrown around so loosely when studying social sciences, science, and health as well. I am going to be trying to address this in the INtro to Psychology class I am teaching at the co-op this year, with some student created experiments and data analysis studies, hoping that they will “get” it a bit more through the doing. Thanks for the references..and pass on any others!


  4. I couldn’t agree with you more. I wrote a picture book on probability (A Very Improbable Story) partly because my friend turned to me and said, “what society needs most is children who learn probability at a young age.” I am reading now a book called The Drunkard’s Walk, which I think is one of the best books on probability written. I highly recommend it.


  5. Oh, Jo, you are right – this is such an issue! Trying to discuss risks and possible benefits with people is so difficult.

    I try to get them to think about what sort a person they are – someone who dislikes the very idea that they are regularly taking medication, for example, and for whom therefore a really significant and relatively sure benefit would be required before taking any medication; versus someone who will worry terribly about eg their cholesterol killing them, and cannot accept doing nothing who might choose to take preventive medication at a fairly low level of risk. And benefits are of different values, too – I have a patient taking long term HRT because for *her* it stopped male pattern balding, and that felt catastrophic to her. Very reasonable, I’d say.

    Just complicates it, but it is important! I find we value familiar risks that we feel we have some control over (such as crossing busy roads) much less highly than those we read about in the media that relate to active interventions such as M-H’s example of HRT, too.

    Rant over (well, continuing in my head!)


  6. Pingback: Erratic … I Mean Eclectic Home Schooling Notes » Post from 8/1/08 — More Math

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s