For a larger version, visit Jorge Cham at PhD comics. It’s worth it. Especially if you are killing time not finishing your thesis (Sara…)

This comic came to me by way of Laura (thanks) and Language Log, where the true nerds among you can go for an illuminating discussion of the difference between ρ and p. Really.

In the last couple of weeks the site has picked up studies that purport to show that passion fruit peel cures asthma, that cocoa helps diabetics ward off heart disease (that one was funded by confectionery manufacturer Mars, funnily enough), and that diabetic men with droopy dicks are headed for heart attacks. Each study is evaluated and put into context, its findings are clarified and its limitations pointed out. The site could become a study of its own: which of Britain’s major news sources are most likely to turn bad science into great headlines?

If you can’t get there, this very good report from the BBC might give you a clue. It focuses on the Bush administration’s uneasy relationship with science. Some White House flunky calls potential scientific advisers to ask them how they voted in the last election. Disturbing, but we’ve seen political vetting of scientists in other places (China, for example, Cuba, the Soviet Union). What’s more disturbing to me is that an American scientist would feel compelled to answer such a question.

“Sam Gosling, a psychologist and an associate professor at the University of Texas at Austin, who has about 300 students on his MySpace page, said there are students today who think professors are not doing their jobs unless they convey information in zany, interactive ways.”

Hmm. A good educator conveys information, certainly (rather than simply spouting it). If interactive teaching techniques help implant information in the mind of a student, by all means use them. But I’d argue that you should ensure that the information you’re conveying is of some use to the students. More use than what your cats had for breakfast or what you like to listen to on your i-pod when you’re out kayaking.

Many professors are apparently spending their time making friends on Facebook in the hopes of making themselves more likeable. More likeable may mean higher ratings on the (sometimes quite vicious) student rating system RateMyProfessors.com. And who knows, in this list-obsessed world of ours, higher ratings may turn into a higher salary, or a better job. I note with interest that the schools with the highest rated professors are not the Ivy League by any means. Number one is the Mormon Brigham Young University in Utah. Hot on its heels is Southeastern Louisiana University. Perhaps the faculty at these lesser known schools have to try harder. Or perhaps the students at Ivy League schools are too busy learning useful things to spend time rating their professors on line. Many professors obviously take student’s acerbic comments rather seriously, seriously enough to respond on video. But I wonder how important “likeability” is in a teacher. A good teacher has a passion for their subject and a drive to communicate that passion. You can be really quite unlikeable and still do that very well.

“The rise of electronic publishing [1], preprint archives, blogs, and wikis is raising concerns among publishers, editors, and scientists about the present day relevance of academic journals and traditional peer review [2]. These concerns are especially fuelled by the ability of search engines to automatically identify and sort information [1]. It appears that academic journals can only remain relevant if acceptance of research for publication within a journal allows readers to infer immediate, reliable information on the value of that research.”

But the discussion (which is usually the Plain English section of a paper) veers deep into the Land of Nerd. For example:

Our finding that the distribution of number of citations is log-normal is in agreement with recent generative models of the citation network [21], [22] that predict a log-normal distribution for subsets of papers related by content similarity. Note that this result is not in disagreement with prior claims about the power-law behavior of the citation distribution [23], as the convolution of many log-normal distributions with different means can yield a distribution that can be hard to distinguish from a power law….

Our findings thus suggest the possibility of ranking journals according to q̅(J). To this end, we turn to a heuristic used in information retrieval called the Probability Ranking Principle [24]. This principle dictates that the optimal ranking of a set of journals will be the one that maximizes the probability that given a pair of papers (a,b) from journals A and B, respectively, q(a)>q(b) if A is above B in that ranking. This probability is also known as the multi-class “area under curve” (AUC) statistic [25]–[27]

The role of open access publishing in a changing scientific landscape is an important one, and good bloggers can increase its importance. But if we want to push that role with a wider public, we’re going to have to summarise our findings in some way that everyone can understand.