Nicholas Christakis suggests we need to get over our obsession with statistical averages.
Christakis is a physician and social scientist who coauthored the book Connected: The Surprising Power of Our Social Networks and How They Shape Our Lives.
He says we've persuaded ourselves that mathematical averages are the
most important way to compare things-countries, professions, actions and
groups of people. Instead, he says, we need to compare variances, which
capture the range or spread of whatever value we're trying to measure.
instance, he writes, the U.S. and Sweden may have nearly the same
average income, but the variance in income-income inequality--is far
greater in the U.S. So the variance may have more to do with life in
either country than the average. He says a more equal distribution of
income might improve the health of the group, and even of individuals
within the group, so we might wish for more equality at the expense of
wealth. But in some cases, he goes on, inequality might be better.
Gathering a crew of 10 sailors, would it be better if they were all
equally myopic, or if one had perfect vision and the remaining nine
varied in their degree of visual impairment? He says you'd probably
choose more inequality in exchange for one with reliable vision.
Christakis is one of the scholars, scientists, thinkers, artist and authors who responded to this year's Edge.org question: What scientific idea is ready for retirement? Cultural impresario John Brockman,
who founded Edge, an online salon for provocative ideas and
intellectual debate, has been posing an intriguing open-ended annual
question since 1998. Last year's question was "What should we be
worried about?" Earlier queries have included such challenges as "what
would change everything?" and "what do you believe that you can't
prove?" The Guardian calls Edge a forum for the world's most brilliant minds, and the 174 essays submitted so far for this year's query offer a dazzling cognitive buffet.
make social tradeoffs, Christakis says, and examining variance will
help us probe such questions as whether we want a richer, less equal
society, whether we want educational programs to increase equality of
test scores, or average performance, and even whether cancer patients
might prefer a drug that extends lives for some but kills others. Other
thinkers have proposed that we jettison current notions of infinity,
information overload, big data, cause and effect, free will, and truth.
author and psychology professor at Princeton, would scrap the idea that
opposites can't both be right. He says sadness and happiness, stupidity
and wisdom and goodness and evil can all co-exist, and context matters.
He cites a study in which seminary students, immersed in Biblical and
ethical learning, were asked to deliver lecture on the parable of the
Good Samaritan. Half were told they were comfortably ahead of schedule,
and half were told they were late. On their way to the talk, all
encountered a presumably injured man slumped in a doorway groaning. Most
of those who thought they had time stopped to help. Among those who
thought they were late, only 10 percent tried to help, and the rest
stepped over the victim and rushed along.
Nicolas G. Carr, author of The Shallows and The Big Switch,
thinks we should ditch "anti-anecdotalism" and recognize that life is
made up of those little stories. Science that scorns them risks veering
away from the actual experience of life, he asserts, adding "The
empirical, if it's to provide anything like a full picture, needs to
make room for both the statistical and the anecdotal." To W. Daniel Hillis, a computer scientist with the technology company Applied Minds, the concept of cause and effect is just an artifact of our brains' proclivity for storytelling. He'd let that go. Gary Klein, a psychologist with MacroCognition,
thinks the idea of evidence-based medicine impedes progress because it
discourages exploration of treatments not tested in randomized
controlled trials. He notes patients suffer from far more conditions
than can be controlled for. Dean Ornish, founder and president of the nonprofit Preventive Medicine Research Institute,
thinks we could sideline large randomized controlled trials. Size
doesn't always matter, he says, and a randomized controlled trial "may
introduce it own biases." He calls for more creative experimental
designs. Click here to view all the essays.