Uncertainty in Science

Another Kind of Shades of Gray: 

Uncertainty in Earth Science and how we teach it

You may remember the Italian geoscientists who were convicted of manslaughter in October of 2012 for failing to warn the public about an earthquake that struck L’Aquila, Italy in 2009.  I remember the dismay I felt, given the complexity of Earth’s systems, that people would expect such certainty from science.  I wondered if the human need for certainty is part of our nature, or if the way we teach science contributes to the public’s misperception that science is never ambiguous.

 This week, I read Naomi Lubick’s article “Be Prepared: Navigating the risks of hazards research” in the January 2014 edition of Earth (http://www.earthmagazine.org/article/be-prepared-navigating-risks-hazards-research).  She explores the thicket of misunderstanding that exists between scientists’ understanding of “uncertainty” and the public’s need to have answers in black and white.  The article reminded me of those questions and made me think about the difficulty my students have in understanding uncertainty.

Recently, I had students read eyewitness accounts of earthquake damage and estimate the intensity of shaking based on the qualitative Modified Mercalli scale.  To help them evaluate their judgments, we reviewed the “answer key”, but, because data from some locations was inadequate for determining the Mercalli ranking, a question mark replaced some answers.  A student told the class “I think I’m uncomfortable with the answer key having question marks.”  Students don’t expect uncertainty in science.  Students think that in science there is right, and there is wrong.  There are no question marks.

Some things we don’t know about earthquakes.  Other things we do know, and with a high degree of certainty.  But some things, we know only with a significant measure of uncertainty.  It is the last of these that are hardest to teach or to convey to the public.  Perhaps one approach is to point out that scientists don’t claim to be able to predict earthquakes exactly, but they do have some ability to predict earthquakes within uncertainty limits.  For example, the legend to the 2008 United States National Seismic Hazard Maps ( USGS) states “Colors on this map show the levels of horizontal shaking that have a 2-in-100 chance of being exceeded in a 50-year period.” However, just stating probabilities doesn’t necessarily convey meaning, especially for middle and high school students.  How exactly can teachers help students understand what this statement means?

Maybe one way to help students understand uncertainty is through a variation of the common math exercise of flipping coins.  Instead of flipping one penny 10 times, consider flipping 10 pennies at the same time.  How many of them will come down heads?  Well, on average, 5.  But in practice, it will only be 5 about a quarter of the time.  It will be either 4 or 6 about 41% of the time (66% chance of being within “1” of 5).  And it will be 3 or 7 about 23% of the time (an 89% chance of being within “2” of 5).  Even though our answer of “five on average” is quite true, there is an uncertainty in our prediction.  Likewise with earthquakes, we may know what’s going to happen on average, but predicting exactly what will happen and when is uncertain.

What have you tried that helps students understand uncertainty?  How did it work?  Does our responsibility to teach “correct information” get in the way of exploring how science works? Please share!



Share Button