When values meet data

NPR ran a timely story tonight about how people with different value systems will interpret the same information differently [1]. Social scientists are applying these observations to understand reactions to vaccination and climate change data.

The story offered explanations but few ideas. For instance, there was this cautionary tale:

So, should climate scientists hire, say, Newt Gingrich as their spokesman? Kahan says no. “The goal can’t be to create a kind of psychological house of mirrors so that people end up seeing exactly what you want,” he argues. “The goal has to be to create an environment that allows them to be open-minded.” And Kahan says you can’t do that just by publishing more scientific data.

While I am not a fan of creating such an “information ministry” in science just to appeal to different value groups, I was left wondering what, exactly, we can do. Since the mission of science is to pursue more research, the last sentence seems at odds with the instinct of the scientist. Certainly, the present state of public acceptance of the science suggests more science doesn’t mean more people on your side.

I came to my own conclusion, in the end, and it’s pretty much what I have been saying in the last groups of posts on this. Education is the key. The only way to make more people capable of processing data outside their value system is to provide them with the mental toolkit to make decisions informed by values and data, and not just values.

The other conclusion is that you won’t sell one solution as the “best” solution when it comes to problems like climate change. For instance, as reported in the story,

In another experiment, people read a United Nations study about the dangers of global warming. Then the researchers told the participants that the solution to global warming is to regulate industrial pollution. Many in the individualistic group then rejected the climate science. But when more nuclear power was offered as the solution, says Braman, “they said, you know, it turns out global warming is a serious problem.”

Clearly, a rich portfolio of approaches is the key. Economic solutions, including the creation of carbon markets and new energy sectors, are part of it. Regulation of total industrial output of CO2 and methane are another part. Efficiency and conservation are further parts. The above cover personal, corporate, and intranational solutions; international agreements is a further avenue to attack the problem.

In a rich portfolio, everybody wins. Free-market types get new industries and new competition, as well as a new way to generate profit while capping your own emissions. Regulation types get what they want. Those who claim you can’t solve global climate change without global solutions get what they want from international agreements, while those who decry international interference in national sovereignty have their “out” through personal choice (conservation and efficiency) and local control (corporate and governmental).

If we can’t sell the science anymore, maybe we can sell the economic benefits of simply acting more responsibly.

[1] http://www.npr.org/templates/story/story.php?storyId=124008307

Thy will be done

Updated on 2/25/2010: some of the language needed editing for clarity.
Updated on 2/23/2010: added discussion of the current state of the medieval warming period, and whether it was actually a global phenomenon.
George Will’s opinion piece in today’s Washington Post serves as a textbook example of the current argument against acting on climate science. If you have an account on the Post’s website, you can get the piece [1]. I’ll dissect the main points here:

  1. Will begins by linking science to only American progressives
  2. He then quotes a “jerk scientist ” statement, chosen from among things the IPCC Chairman has said of late. He then wags his finger at this behavior.
  3. He then cites the illegally released emails from last November as casting doubt on the scientific process involved in drawing conclusions from climate data
  4. He notes that climate action policies put into effect by a few U.S. western states, or climate coalitions including energy industry players, are beginning to fall apart.
  5. He reports that global average temperature has not increased in 15 years.
  6. He notes the wrong IPCC claim that India’s glaciers will melt by 2035.
  7. He then cites the speculation that if southern hemisphere temperatures in the “medieval warming period” were as high as northern hemisphere temperatures in the same period, then that period would be warmer than now
  8. He repeats that all of this points to an absence of warming
  9. He then calls climate science a religion

It’s interesting to study these points and see whether they form a coherent argument against climate research and its conclusions. Points 1, 2, 4, and 9 are all rhetorical gimmicks As such, they don’t relate to the process or conclusions of climate science. Jerk scientists don’t make for bad science – they just make for good quotes to use to make people think scientists look down on the public. If scientists couldn’t be jerks, they’d had to revoke at least half of the Nobel prizes ever awarded. Just because policy actions or coalitions fail to make an impact in 3 years doesn’t mean that the science itself is flawed. It does mean that, as always, politics and policy are difficult and we have to work harder, not give up.

So let’s focus on the points that would suggest a problem with the science. Some of these have recently been addressed. The British MET, embarrassed by the emails mentioned in point 3,  re-examined data. They did so  in lieu of additional criticism from third parties about climate data analysis. The results of the re-analysis had no impact on the conclusion that the earth is slowly warming and that this is connected to human activity. Will might have missed that, since it’s recent news (science is always on the move) [2]. The land-based temperature record [3] was re-evaluated by the MET and is shown below:

Land-based temperature record (British MET Office)
Land-based temperature record (British MET Office)

The above record would seem to indicate that for about 10-15 years, within uncertainties on the year-by-year numbers, there have been only slower or no increases in temperature. So this would seem to confirm point 5.

We can also look at NASA studies from multiple sources of temperature data [5] -” . . . weather data from more than a thousand meteorological stations around the world; satellite observations of sea surface temperature; and Antarctic research station measurements . . . ” – combining land, sea, and polar data. Since the earth’s surface is about 70% water, it seems a little disingenuous to neglect data from other important sources. What do we see?

NASA study of land, sea, and polar climate data
NASA study of land, sea, and polar climate data

We see that not only has the last decade shown SUSTAINED warming in both hemispheres of the planet, but 2009 marked RECORD HIGH temperatures since 1880.

A quote from the NASA article sums up the danger of mistaking short-term behavior for long-term trends:

“There’s always an interest in the annual temperature numbers and on a given year’s ranking, but usually that misses the point,” said James Hansen, the director of GISS. “There’s substantial year-to-year variability of global temperature caused by the tropical El Niño-La Niña cycle. But when we average temperature over five or ten years to minimize that variability, we find that global warming is continuing unabated.”

Is Will drawing his conclusions by only choosing to look at data from the land-temperature record and discounting other records? He shouldn’t be afraid of using more data. Is he drawing his conclusion by mistaking a 10-15 year effect for a 100-year effect? What about other periods in recent history?

Consider the land-temperature data from about 1940-1970. That seems almost flat, if not declining, and it was sustained for far longer than the present land-data trend. Amplifying this period, the same trend is also very clear in the combined-data NASA study. Yet, after the 1970s the pace of warming – in all data – accelerated. Generally speaking, from the 1880s to now, the trend is far more up than not. When the temperature climbs, it does so quickly. We again see the danger of mistaking short-term effects for long-term effects

Ten years – even 30 years – is not that long on the scale of climate. Yet the temperature goes up quickly, and does so even on the background of decadal periods of natural warming and cooling from causes like solar activity and the El Nino/La Nina cycle. So what happened during that earlier 30-year period?

Let us not forget also that 1940-1970 was the time during which we shoved industrial aerosols into the atmosphere; these are now understood to have a cooling effect [5]. It was only with the beginning of clean-air movements in the 1970s that aerosol levels began to decline. This decline removing an anthropogenic forcing factor from the climate that drove cooling. In effect, another negative human activity helped to mask our larger warming impact on the climate. Once the air got cleaned up we revealed the warming again – with a vengeance. Cleaning up the air in our hemisphere took the collar off the pit bull. It’s worth noting that in the Southern Hemisphere, temperatures steadily climbed during that same period.

What of the medieval warming period  mentioned in point 7? This period is illustrated below [6]:

Temperature in the Northern Hemisphere (2000 years)
Temperature in the Northern Hemisphere (2000 years)

The curves are derived from 10 different data sources which can be used as proxies for  temperature at times in the past. Thinking of the color-coding like the visible spectrum of light, more recent data sources are shifted toward the red while older sources of proxy behavior are shifted toward the blue. The variation in temperature records from different sources are then related to the uncertainty on the actual temperature during a given period.

The medieval warm period is fairly apparent in the above plot, but the only strong evidence that such a period occurred comes from studies of climate in the northern hemisphere. Data on climate in the southern hemisphere is less available, and where available does not reflect this warming. In fact, taken all together the existing data on both the northern and southern hemispheres suggests that the globe was cooler overall than in the mid-20th century. The data from outside the north Atlantic is more scattered and points to a significant region-by-region variation in temperature during that period. Will is relying on incomplete data to draw his conclusion that the current period is not “the warmest on record.” His reliance on this warming period seems shaky, at best.

This leaves only Will’s point 6, but making a mistake in a report does not mean all of climate science is bunk. In fact, the different studies by the MET and NASA point to a convergence of conclusions not only from multiple independent researchers but largely independent data samples.

Will’s argument is a good snapshot of the current attempt to mistake climate science for unsuccessful policies and politics, to mistake scientists for the scientific consensus, and to mistake short-term effects for long-term trends. It also is a good illustration of how quickly  we forget that we helped cool the earth when we pumped aerosols into the atmosphere. (That cooling was also helped by having volcanoes erupt periodically – but that was out of our control). Cleaning up our aerosol mess (which helped destroy the ozone layer) only revealed the bigger mess we had already made.

I encourage every scientist to read Will’s piece. As a blueprint, it will help you to understand what mistakes are made when talking about scientific methods and results. Climate science is science, and like all science:

  • It is a web of evidence linked by hypothesis-testing. That web is growing, not shrinking, as researchers continue to question each other’s methods and conclusions.
  • Science is not defined by one result or one researcher. Remember that there is a big picture, and don’t mistake a few examples of human error for a complete failure of the scientific method.
  • Climate conclusions are not owned by one source or one agency – scientists come from all over the world, collect independent lines of data, and constantly criticize each other’s conclusions. Yet, the warming appears to remain as an unmistakable feature of the natural world, tracking beautifully with our continued industrialization of energy and food production.

[1] http://www.washingtonpost.com/wp-dyn/content/article/2010/02/19/AR2010021903046.html

[2] http://www.foxnews.com/scitech/2010/02/16/key-climate-change-data-laden-errors/

[3] http://www.metoffice.gov.uk/climatechange/science/monitoring/data-graphic.GIF

[4] http://www.nasa.gov/topics/earth/features/temp-analysis-2009.html

[5] http://en.wikipedia.org/wiki/Global_cooling#Physical_mechanisms

[6] http://en.wikipedia.org/wiki/Medieval_Warm_Period

Don’t be a science jerk

Of late, I’ve written some things in my blog that, upon reflection, make me a science jerk. For instance, in my recent discussion of media coverage of the climate science mistakes uncovered in the last four months, I pretty much made it sound like anybody who doubts climate science is an idiot and should suck it up . . . and then walk it off.

As a scientist, it’s difficult to watch a great number of research peers come under attack by a public that largely ignores science until it either (a) produces a new toy or (b) conflicts with societal notions, common wisdom, or values. Climate science is one of those things where nature appears to be trying to tell us something, and most people in the U.S. don’t seem to know how to listen. Ultimately, the failure is ours. If the public fails to comprehend the scientific method and therefore disregards its conclusions, then it’s only because we failed to live up to our end of the societal bargain wherein good public education produces an informed electorate. If companies have better information machines that cast doubt on the science, it’s our fault for not having similarly organized grass-roots machines that communicate our science.

Being upset is not a license to bash. Even if I feel angry about how the media is portraying the climate e-mails, or climate data analysis, or some of the IPCC report conclusions, I shouldn’t bash my fellow citizens. At the time, I didn’t think of it as bashing. Everything looks different in the cold light that makes hindsight possible. Science as a whole only further fails the society that it criticizes for not understanding science. It’s more a sign that we have failed in our role as educators, rather than a sign that people can’t understand science.

I still stand by my fundamental concerns, and reiterate them here more as points than anything else:

  • Science is not a chain, it is a web. At first, as pieces enter the web and connections are made, the strength of the web is tenuous and, really, the web may be just as likely to fail as to succeed. The web can collapse not because its conclusions run afoul of common wisdom or societal values, but because its claims simply fail to hold under withering scientific scrutiny. As a result, an area of science is not slave to the weakest conclusions in the web, but instead by the number of connections joining one conclusion to many experiments, which themselves obtain reinforcing results. A chain is defined by its weakest link; science is not such thing, and by its nature tends to weed the bad or useless ideas and identify the useful or correct ones.
  • Mistaking short-term fluctuations for long-term trends is at the heart of the climate science discussion. Anyone who claims that snow trucked into Vancouver is evidence of global warming is just as dangerous to public understanding as people who claim snow in Dallas is evidence against global warming. Neither of these is evidence of anything, but rather are observations in a large web of observations, some of which may be connected by hypotheses and other which are irrelevant. It is incumbent upon all scientists to keep a focus on the long-term, using the short-term merely to add richness to the data set that is needed to establish trend and outcomes.

Most importantly, don’t be a science jerk. Don’t tell people who disagree with you that they are stupid. It’s a great way to make more enemies defined not by their understanding of science but by their opposition to individual scientists. Rather, focus on education. Focus on the quality of your own work, and on the work of others. Keep science beautiful, work to define your own brand in the enterprise, don’t let anyone tell you what you mean or who you are, and never tell anybody they are incapable of understanding your work. The only way we can ever make progress on anything in our society is to respect its members while at the same time arming those members with the very tools of skepticism that enable science to make progress in the natural world.

We will necessarily create more skeptics, armed with the freedom to inquire and the responsibility to respect the process, and while this will leave some people who disagree with the science it will create many more people who can act as independent and responsible citizens.


Athletics at SMU can raised quite heated responses from members of the SMU academic community. This response is rooted in many things, but primarily a perception that student athletes are expected, or themselves expect, to achieve less in the academic realm than their non-athlete peers. Setting aside the reality for a moment, which is always more complicated, it’s useful to focus on the perceptions. Thinking about the meaning of the perception may lead to useful conclusions.

What are the perceptions of athletics from academics? As with any school with a visible athletics program, athletes are assumed to have less preparation before coming to college and are assumed to have been admitted not because they could prove a valuable scholar but because they can run, jump, or throw (or all of these). It’s also assumed  that a healthy athletics program draws money and attention into the university, and has the potential to benefit the entire university. As with anything involving a large, up-front investment of money and human capital, there is a tipping point that must be reached before the benefits outweigh the costs. I think it’s safe to say that it is also assumed that it is the duty of the academics to make sure any under-prepared athletes get the education they deserve.

Truth usually (but not always) lies somewhere in the tangled mess of common wisdom. We can begin simply by asking “What is a scholar?” Merriam-Webster keeps it simple: “2 a : a person who has done advanced study in a special field b : a learned person” It’s not really specific about what field, or what kind of study. I think we all assume that this means academics, but why must the scholar be divorced from athletics? Exercise of any kind, coupled with activities of the mind, improves both.

Speaking from personal experience, a little exercise every day keeps my own stress levels in check and gets my creativity moving. Without some exercise, I grow irritable, incapable of executing tasks and keeping things organized, and in extreme cases feel the effects of panic or indecision. In my own life, dividing the academic from the athlete (however amateur) is a fruitless choice.

Thinking more broadly  about the university, athletics clearly has a place. It serves to attract to the school individuals who might have never considered SMU, and in that way serves to enrich the diversity of both background and mentality present in the university population. Where research or humanities fail to draw large singular  crowds and attention to the university, athletics does so with ease. Athletics forms loyalty, which can last more than a lifetime. Media are more likely to latch onto athletics than to academics – it’s easier for the media to understand and communicate (and, to be fair, it really is easier to communicate athletics simply because of its visceral, human-scale drama).

Taking a lesson from CERN and “Angels and Demons”: all press is good press. Sure, Dan Brown decimated the science to tell his story, but CERN seized on this opportunity to educate the public about the real subatomic particle research at CERN. SMU has a similar opportunity. Faced with the goal of expanding the research profile of the university, having a visible athletics program draws attention (and money) to the university from segments of the population otherwise un-inclined to care about a university. That money should not only serve the athletics program . . .  but that’s for higher-paid people than I to decide and manage.

It is also, however, incumbent upon the athletics program to act like part of the scholarly mission of the university, and if it intends to recruit students with weaker academics it must pay for tutoring and other extra academic support. I want students to have the same opportunities, and if all an athlete needs it a little extra math tutoring to take a crack at cool science classes then by God let’s see money for tutors. Academics also have a responsibility in this. They must strive to do their best to support students who need a little extra work and encouragement. Students don’t come fully formed to university – otherwise, why are they paying us to teach? All students have a talent, a love, a passion – if sports is that passion, why not draw them into science with a “Sports Science” class? It can be fun, relevant, and teach principles of modern science in biology, chemistry, and physics.

I don’t see a dual role for SMU, or any other such university, in the coming decades. Wanting to become a better scholarly institution should never be at odds with having great athletes. We shouldn’t label people as “academic scholars” or “scholar athletes,” as if those people are divorced from one another. Rather, let us labor to produce scholars, young and talented people with a love of academics and athletics, and let us give them the value system to enjoy a lifetime full of both.