The future needs statisticians
The current issue of the journal Science has an editorial on the importance of statistics, and on the increased demand for statisticians in the `Big Data’ future. The writers, Marie Davidian and Tom Louis, call out the need for increased funding in graduate programs — it hasn’t kept up with inflation, let alone with demand.
They also note
The future demands that scientists, policy-makers, and the public be able to interpret increasingly complex information and recognize both the benefi ts and pitfalls of statistical analysis. It is a good sign that the new U.S. Common Core K-12 Mathematics Standards introduce statistics as a key component in precollege education, requiring that students be skilled in describing data, developing statistical models,making inferences, and evaluating the consequences of decisions.
Here, at least, New Zealand is ahead of the game.
Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »
I would suggest it is more than “we need statisticians”. As an engineer with a passion for statistics, I think we need to ensure more professions have adequate and quality statistics teaching/lecturing.
The ‘Big Data’ future will still demand that non-statisticians deal with the data – too many have a little knowledge to be dangerous, but not enough to be useful.
12 years ago
Yes, we need both. Specialist statisticians are needed — people whose mental toolkit includes terms like ‘likelihood’, ‘sufficient statistic’, ‘unbiased estimating function’, and so on.
We also need engineers and programmers who know about things like regression models, shrinkage, cross-validation, overfitting.
And we need informed citizens who understand concepts like randomization, the fact that variance depends on sample size but bias doesn’t, the need to measure the variable you actually want to draw conclusions about, the relationship between risk difference and risk ratio, and so on.
From a public policy point of view these are probably in increasing order of importance but also in increasing order of difficulty. There’s already a good infrastructure set up to make statisticians, and it’s constrained primarily by funding, and graduate students aren’t all that expensive. There’s a lot of US stat and biostat graduate programs whose current structure could handle a 50% increase in students without much difficulty.
12 years ago
It’s not just the engineers/scientists etc that need to understand. I’d rather like the base level of understanding that the average citizen has to increase significantly. To quote Steven Pinker in one of his recent interviews in the guardian:
“I think that a failure of statistical thinking is the major intellectual shortcoming of our universities, journalism and intellectual culture. Cognitive psychology tells us that the unaided human mind is vulnerable to many fallacies and illusions because of its reliance on its memory for vivid anecdotes rather than systematic statistics. Yet pundits continue to hallucinate trends in freak events, like the Norwegian sniper (who shot all those young people on an island) and make wildly innumerate comparisons, such as between Afghanistan and Vietnam, or between today’s human trafficking and the African slave trade”
http://www.guardian.co.uk/science/2011/oct/15/steven-pinker-better-angels-violence-interview?newsfeed=true
12 years ago