I've put a complete copy of a letter published in Nature this week by Pablo Jensen, on the outreach activities of French scientists, below the fold. The key point is that most don't do anything.

Preliminary data on activities carried out by the 10,403 CNRS scientists, from July 2003 to June 2004, show that three-quarters of them did not get involved in any popularization, or public-outreach, activity during the year — taking this to mean writing a book or presenting a popular-science lecture or poster.

That's actually a generous estimate as far as I'm concerned, since the letter says that social sciences do a better job of popularizing their work (41% are out there making the case for their research) than physicists, chemists, and biologists (17%).

I find this distressing. This is one of our problems, I think; just as science training doesn't involve any instruction in how to teach, we also learn nothing about how to talk to the general public. We know very well how to buzz among our peers in the ivory tower insula, but put us on the street and we're mostly a gang of off-putting fumble-tongued nerds.

Here's the letter:

Your Commentary article "Weapon of mass attraction" (Nature 433, 357−358; 2005) highlighted poor participation by US scientists in public-outreach activities.

It may be interesting to look at the situation in France, where the reform of the national research agency, the CNRS, has led to the creation of a working group on the popularization of science, which I lead.

Preliminary data on activities carried out by the 10,403 CNRS scientists, from July 2003 to June 2004, show that three-quarters of them did not get involved in any popularization, or public-outreach, activity during the year — taking this to mean writing a book or presenting a popular-science lecture or poster.

A closer look at the data reveals that efforts are unequally distributed: the most active 10% of scientists account for 70% of all public-outreach activity and the top 5% account for half.

A fit of the histogram of the number of activities per scientist produced a more precise picture. We discovered that a single Poisson distribution cannot fit all the data. This confirms that scientists do not constitute a homogeneous population.

The best fit we found requires three different Poisson distributions, which can be interpreted as three different sub-populations: the 'silent majority' (76% of scientists, who do hardly anything); the 'open minority' (21%, who carry out some activity once or twice a year); and the 'semi-professional popularizers' (3% of scientists who carry out activities on average six times a year). Researchers in this active minority dedicate a significant fraction of their research time to the public, accounting for a third of all activities in bringing science to the people.

Policies aimed at getting scientists more involved in public outreach should perhaps be tailored to the three sub-populations we found. For example, those who do not yet carry out any public-outreach activities have to be convinced of the importance of doing so. The decline in student numbers seems to have persuaded more physicists to get involved in the World Year of Physics. The 'open minority' might be encouraged by access to simple tools for efficient public outreach, while the 'semi-professional popularizers' may have concerns about institutional rewards.

Contrary to expectations, our data suggest that age is not significant. We find that the average number of public-outreach activities is broadly constant with age, increasing moderately as scientists get older, from 0.45 activities a year for those aged 31−35, to 0.7 a year at 56−60.

Our data also reveal variation among fields: the proportion of scientists carrying out public-outreach activities varies from 17% for general physics, chemistry and biology, to 30% for astrophysics and 41% for social sciences.

Interestingly, while the mean number of activities per scientist varies considerably from field to field (from 1.2 for social sciences to 0.3 for general physics, chemistry and biology), the productivity of active scientists is constant across all fields (close to 2.5 actions a year).

Finally, our data reveal that speaking at conferences on popular science is the most common activity (25%), followed by writing newspaper articles (23%) and giving radio or television presentations (17%).

Again, there are variations across different disciplines. For example, much higher numbers of social scientists appear on radio or television (61%). So any researcher willing to pursue television might be able to learn something from their colleagues in the humanities.

Jensen P (2005) Who's helping to bring science to the people? Nature 434:956.