Complexity and the Information Landscape
This morning the newswires are buzzing with a story that alleges Britain’s Climatic Research Unit fudged some of its climate change data and suggesting that a ‘bunker mentality’ took hold in the unit, which led to this kind of skewing of the data and science. One scientist told Doug Saunders from the Globe and Mail that “It wouldn’t be an exaggeration to say that this has set the climate-change debate back 20 years.” Indeed, with the Copenhagen Climate Summit about to start, there is real concern that these allegations – whether proven true or not — will impair the delegates’ ability to reach a deal.
On a different, yet related note, yesterday I went and got my H1N1 shot and was told by the official guiding people through the clinic that about 37 percent of the population of Toronto have had the vaccination. I went to the downtown clinic and waited about 2 minutes to see someone, which is in stark contrast to what we saw a few weeks ago.Why? The threat of H1N1 seems much less in the here and now than it did a few weeks ago when, in the span of one weekend, when U.S. President Obama declared swine flu a national emergency, and two young people in Ottawa died from H1N1. Towards the end of October, H1N1 seemed a lot more scary and that made the issue a lot simpler: get protected or die (or so it seemed)
So what do these two stories have in common? Both illustrate the problem of complexity in the information landscape. H.L. Mencken is quoted as saying: “For every complex problem there is an answer that is clear, simple, and wrong“.
The problem that public health and scientific research faces is that it is in the business of complexity, yet the business of the media is too often in simplicity. This caused that. That person is bad, this person is a hero and so on. The archetypes and stereotypes come in spades and that is the problem. On the issue of climate change, most scientists worth their salt looking at the data are concerned about what is happening to our climate, not because they know for sure, but because they don’t. In a complex system like the environment, the overlaying causes, consequences and potential confounders of data make it impossible to say for sure that something causes something else in a specific dose. What can be done is that we can observe large scale patterns of behaviour and anticipate changes based on models developed using past, current and possible future (estimated) data and scenario planning.
In public discourse however, this makes for a less compelling story. Many like to think that buying a hybrid car, recycling, and carrying a reusable shopping bag will help solve the problem of climate change, when the truth is an entire system of small changes needs to take place if we really want to make a difference. This speaks to a fundamental lack of understanding of complexity.
With the H1N1 example, complexity is less about the cause and effect relationship of the disease and host and more about the vaccine developed to help prevent it. There are an entire littany of websites, pundits and voices who have turned something that is complicated like a vaccine, with potential complex outcomes in rare events such as allergic reactions, into overly complex issues around patient safety, conspiracy theories and the like. I commented on some of these issues in a previous post. At issue here is a fundamental lack of understanding of statistics and probability.
The problem is that the two are related. For those of us in public health, this is an issue that can lead to sleepless nights. How to both make complex information accessible and interpretable to those without the interest, time or ability to sift through it and make reasoned, informed decisions AND how to enhance people’s understanding of probability? Just yesterday in my course on health behaviour change a student in epidemiology remarked that even something as fundamental as an odds ratio to her field gets debated and misunderstood among her peers. John Sterman at MIT has studied his students — ones that learn about system dynamics — and found that many of them have difficulty grasping the fundamentals of the ‘bathtub problem’ and accumulation, which I discussed in a previous post.
I would argue that this is one of our most fundamental challenges as educators, scientists and members of society.
Think you know about stats and complexity? You might be surprised (and entertained) by how randomness creeps into our lives by listening to the recent podcast on recent episode on stochasticity, or randomness, from WNYC’s Radio Lab.