Science: practising it, understanding it, communicating it

09/2011

My next blog post will return to the main purpose of this site, I promise! But today I amused myself by processing all the articles I'd gathered under the broad label of "science communication". At some point, this may form part of a small section of the site on scientific literacy. In the meantime, some of you may find my collection of links interesting. Feel free to add any other relevant links in the comments! [Do note that this collection are just articles that washed up on my shores, that I thought interesting enough to dump in a file. I am in no way claiming that they're exhaustive or carefully selected.]

The collection can roughly be divided into four topics:

The modern practice of science

Peer review and the corruption of science

An article attacking that bastion of academic research, peer review, argues that “Pressure on scientists to publish has led to a situation where any paper, however bad, can now be printed in a journal that claims to be peer-reviewed.” It’s undeniable that peer review has flaws — yet the idea of having no independent experts review the research is also worrying.

But perhaps most interesting is the pharmacology professor’s attack on an acupuncture study, and the responding comments from the researchers and the journal in which it was published. This argument emphasizes for me the difficulties involved in performing and evaluating research. Non-researchers tend to believe that science is about “facts”, and any scientist can look at the same facts and come to the same conclusion. If only it was that easy.

I think the main point of all this is that the deluge of research needs better tools than we currently have in place.

Open science, Freedom of Information and the Big Journal monopoly

A discussion of the problems involved in making academic papers freely available. I really loved his example of the journal"Molecular Biology and Evolution" publishing a retraction notice behind a $32 paywall! But as the writer points out, however desirable it is for this stuff to be free, the people involved in finding and shaping this information need to be paid somehow.

What it means to practice science

The author of the book Everyday Practice of Science: Where Intuition and Passion Meet Objectivity and Logic says the book is his answer to the excellent question: Why do you believe what you think you know?

This article is about the real-world messiness of how science is done. About the “ambiguous and convoluted route” to discoveries. About how no one is objective, and “Opportunities for misinterpretation, error, and self-deception abound”. But most importantly, this is about the “credibility process” — the path new findings follow through the scientific community and its practices.

Some non-scientists (usually ones with their own obsessions) are quick to point to any failings in individual scientists or specific claims. But if they understood anything about science, they’d realize that these instances are not “science”. As Frederick Grinnell says: “In the case of science, it is the commons of the mind where we find the answer to Montaigne’s question: Why do you believe what you think you know?”

Scientific papers are given too much weight

Reporting from ScienceOnline London, David Dobbs reports on the discussion of the idea that the scientific paper is now wildly overvalued and overemphasized. It’s far too slow and it consumes far too much money and effort. Instead, it’s argued, scientists should blog and share data using open notebooks.

Moreover, papers are given far too much weight, with “tenure decisions, grant awards, and even university ratings now focus[ing] so heavily on publication in high-impact journals that the paper has largely displaced the real currency of science – the data, methods and ideas that papers are supposed to communicate – with the papers themselves.”

Dobbs throws in another argument — the emphasis on papers means that scientists are inclined to think their job is over when the paper is done. But (as with writing books), the work is only half-done — scientists should really be trying to make their research known and explaining to the world why it matters.

Deluge of scientific data needs to be curated for long-term use

In response to The Center for Informatics Research in Science and Scholarship at Illinois becoming a partner on the Data Conservancy project, this press release talks about the role of data curation in research, and why a data-management plan needs to be in place right from the beginning of any project.

"Saving only the publications that report the results of research simply isn't enough anymore. Researchers also need access to data that can be integrated and re-used in new ways. This is especially important in data-intensive science, where the power of discovery lies in applying computational approaches to large, aggregated data sets." It’s also important for replicating and validating any project.

Digital content is vulnerable, and just putting something online is no guarantee it will stay there. Nor is it just a case of uploading it — data has to be organized usefully to be of value.

Science and politics

A reminder that science should be independent of politics comes when the British government published a set of principles to "clarify the relationship between advice and policy".

A new scientific paradigm?

Microsoft have produced a book, Data-Intensive Scientific Discovery, discussing what Jim Gray, a database software pioneer, called the “fourth paradigm” that is transforming the practice of science. That is, the use of computers to handle the flood of data. He argued that this flood demands “a world in which all of the science literature is online, all of the science data is online, and they interoperate with each other.”

The book contain essays on how computers and networks are transforming scholarly communication; on the new scientific instruments that can capture mindboggling amounts of data; on the need to share data (for a variety of reasons); on how ideas taken from computer science might redefine modern science.

Statistical literacy

A common statistical error

Ben Goldacre talks about a recent finding that the same statistical error – ignoring the "difference in differences" – is so widespread it appeared in half the papers in which it could have been made (157), out of 513 papers published in five prestigious neuroscience journals over two years. Extending the survey, they found 25 out of 120 cellular and molecular articles in Nature Neuroscience similarly misunderstood what it takes to measure a significant difference in response between two groups. Moreover, not a single paper analyzed differences in effect sizes correctly.

Statistical noise

Ben Goldacre uses recent unemployment figures, and their media presentation, to discuss samples and confidence intervals, and how to understand whether changes are real or just “statistical noise, the gentle static fuzz of random variation and sampling error, making figures drift up and down, following no pattern at all, like the changing roll of a dice.”

Any set of figures needs adjusting before it can be usefully reported

Ben Goldacre discusses, using a recent news report claiming unplanned children were slower to develop, the importance of accounting for social and demographic factors. I have to say I find it completely inexplicable that these reports (and the initial press release, so you can’t just blame the media) even mentioned a finding that disappeared when these factors were accounted for. Still, it provided Ben with an excuse to nicely explain why and how you have to account for various relevant factors. [Note: I usually mention this in my news reports when I think it’s of particular interest, but otherwise I tend to assume that everyone realizes that results have taken relevant factors into account — in the same way I assume that anyone reading my reports knows what a control group is. It’s the absence of these factors I would usually find worth mentioning — if indeed I bothered to report on any study that failed in these basic requirements.]

Communicating science

Why media literacy is part of scientific literacy

Why do so many people fall for the climate change skeptics’ arguments? A lack of media literacy, this researcher claims. “Research shows that laypeople and the media tend to view all scientific viewpoints as equally valid and, therefore, give too much credence to the minority viewpoint of skeptical scientists.”

So science educators should also tackle media literacy education, to encourage people to ask questions about the people or organizations behind reports, as well as about the content and its credibility and completeness. I’m a big fan of this — except I wouldn’t put the onus totally on science educators. This is core stuff that should be taught at school. When I was at secondary school, our English curriculum included instruction on this, at least in regard to advertising. I was surprised and disappointed when my sons went through secondary school, to find that it was apparently no longer part of the curriculum. And yet the need for this sort of critical (and dare I say cynical) thinking has never been greater.

Another approach, it was suggested, is through citizen science — science projects that involve non-scientists, such as Galaxy Zoo and Field Expedition: Mongolia. I’m a huge fan of those too. (See this Guardian article for more about Galaxy Zoo)

A third approach is by broadening the media used in science communication — as described in the next article.

Is theatre exposing the gutlessness of TV science?

The article attacks the conventions of TV science shows — “the impression that in the world of TV, science is 'other', 'something else', a thing that has its place, that has to be tackled a certain way and shown to the public in small, carefully controlled doses in case it leaks out and contaminates a debate show or CBeebies.” And describes a number of less-conventional science communications, including:

And then there’s the rise of science comedy.

Here’s another way of involving the public in science:

Using Twitter to create data

A nice example of creating data through crowdsourcing, using Twitter. A snow map of Britain at a time of heavy snowfalls, generated by followers tweeting part of their postcode, with a rating of snowfall.

Time for change in science journalism?

A discussion of the problem in science journalism where journalists typically treat the publication of a paper in a journal as a newsworthy, validating event, and so everybody reports on it at the same time (in response to the press release) and no one waits to see how the stories develop. It’s suggested that if everyone waited six months, they might find that the paper was not so important after all, or that it would be better understood in conjunction with other related work. With less emphasis on what’s new, more stories might be written “about trends in research, or the accretion of ideas within fields, or more deeply analytical pieces. Articles could do a better, more thoughtful job of providing context.”

The communication gap between scientists and the public

A workshop involving scientists, engineers, public policy experts, lawyers, ethicists, and journalists explored the communications gap between scientists and the public. A review of the workshop findings and recommendations is available at http://www.amacad.org/publications/scientistsUnderstand.aspx.

Helping scientists become better at public communication

An analysis of science communication as it related to the establishment of "marine reserves" off the California coast outlines a comprehensive communication strategy that might be applicable to other areas of science and natural resource management. The approach included such innovations (sorry, a bit sarky, but none of this is rocket science; amazing that these things aren’t routine in this domain) as:

  • identifying the needs and background of the audience they wanted to communicate with;
  • pinpointing a few "main messages", such as: what’s the problem, why does to matter to you; what needs to be done; how you’d benefit from that;
  • using several media, including printed materials, websites, and presentations,
  • tracking the success of the strategies.

Why scientists should blog

On which subject, here’s Chad Orzel reminding us that it was “not until wide and open dissemination of scientific results became the norm that we saw the tremendous explosion of scientific knowledge that has shaped the modern world.” But of course, as he points out, modern scientific papers are not, as a rule, intelligible to the general public. This in a time where science has become more important than ever. Which is why scientists need to blog.

Example of poor science communication, and example of good

Great infographic on climate change. But the real interest here is from David McCandless’s discussion of how he researched this information: “I deliberately chose not speak directly to any climate experts or leading scientists in the field. I used only publicly available web sources [in order to] simulate what it's like for people trying to learn about climate change online. … I was generally shocked and appalled by how difficult it was to source counter arguments. … This must be one of the reasons why scientists and leaders are struggling to convince sections of the populace about the threat of climate change. Because they're doing such a terrible, terrible job explaining it...”

The science/art divide

On the fiftieth anniversary of C.P. Snow’s famous essay on the two cultures, here’s a discussion of whether the science/art divide still exists. I really liked Collini’s comment about the need for specialists to “remain ‘bilingual’ - able to communicate with a wider society that does not share their idiom” — because that takes the problem, quite properly, to the overarching problem of any translation between any specialist topic and anyone outside that specialty.

I also want to call attention to Grayling’s comments, reminding us that scientific illiteracy among both our leaders and the general public is of particular significance (as opposed to, say, ignorance about art) because science is so vital to modern society.

"Streams of Content, Limited Attention: The Flow of Information through Social Media"

An interesting discussion of the idea that we are now living in content streams, streams of information. “The idea is that you're living inside the stream: adding to it, consuming it, redirecting it.” About the move from broadcast media to networked media, and how that “fundamentally alters the structure by which information flows”. About the fact that this doesn’t mean that all information is created equal, nor that people will ‘naturally’ flow to the best, most useful, most truthful, most important, information.

She predicts several changes, including:

  • information spaces will get more niche
  • “the key is not going to be to create distinct destinations organized around topics, but to find ways in which content can be surfaced in context, regardless of where it resides”
  • new tools will allow people “to more easily contextualize relevant content regardless of where they are and what they are doing and tools that allow people to slice and dice content so as to not reach information overload”

Science education

College students lack scientific literacy, study finds

Assessment of fundamental science knowledge of more than 500 students at 13 U.S. colleges in courses ranging from introductory biology to advanced ecology revealed that most didn’t truly understand the carbon cycle, in particular the principle of conservation of matter.

Actually, this account of the problem reminds me of recent educational research into the importance of tackling students’ preconceptions. It seems clear that there is a big gap between formal understanding (the knowledge in the textbooks) and informal understanding (what “everybody knows”). We all ‘know’ plants get most of their nutrients from the soil, because that’s what we get told when we’re kids. Whatever you learn in class is not to going to touch that belief unless instructors go head to head with it and prove it wrong (even then it’s an uphill battle!).

Science education for children

A commentator in the Guardian notes that “Of course scientists can always improve the way we present our work to the public, but well-taught, well-designed science curricula that have the freedom to be difficult and exciting will go a long way to harnessing and developing the fascination that children have with science.”

Comments

Add new comment