The authors thesis is that the only mandatory communication of results is in peer reviewed journal articles. Scientists aren't required to do other communicating and often leave communication to the public to the media. They ask if is this is adequate given the very low percentage of scientific articles that ever make it into the press, particularly in areas outside of health and medicine, and also given the fact that for everyone out of formal education, the media is their primary source of science education.
Recent studies do show that scientists often don't mind talking to reporters and do so more frequently that one might think [1-2]. They do get kind of frustrated that their work is misrepresented - even if that misrepresentation is failing to include qualifying statements. Newspapers in general covered a lot more science over time (as studied in the time period 1951-1971, I know). Fancy journals that issue press releases for papers find that those papers are more likely to be reported in the news media. The authors cite another study that some 84% of the newspaper stories originated from press releases.
This study was just about how much makes it to the media and is that percentage staying steady as the number of papers increases. When they actually did the work, they only looked at parts of 2 years, 1990 and 2001, and two media outlets, Time and NBC News. They didn't use the WaPo or NYT because better educated people read them (???). Plus, they found that only 25-50% of news pieces actually mention the article's author and venue, so they probably missed a ton.
So this is quite disappointing, really. The study narrowed the coverage of the search so much, that I don't think it's really representative of anything. Of course only a few articles get discussed in the media, but if you want numbers, this paper won't help. These articles also need to start discussing things like Nova and National Geographic and Discovery Channel. We watch that stuff all the time and so do a lot of people we know (of course I'm pretty well educated, I guess).
They mention journal press releases, but for big science there are also lab press releases and media officers. There are also scientists talking directly to the public on blogs.
One thing you can probably take away, if you work outside of biomed and/or are not publishing in Science or Nature and have a really cool result, don't wait for the press to come a knockin' - get it out there another way.
Here's the citation:
Suleski, J., & Ibaraki, M. (2010). Scientists are talking, but mostly to each other: a quantitative analysis of research represented in mass media Public Understanding of Science, 19 (1), 115-125 DOI: 10.1177/0963662508096776
 Peters, H. P., Brossard, D., de Cheveigne, S., Dunwoody, S., Kallfass, M., Miller, S., & Tsuchida, S. (2008). Science-Media Interface: It's Time to Reconsider. Science Communication, 30(2), 266-276. doi:10.1177/1075547008324809
 Dunwoody, S., Brossard, D., & Dudo, A. (2009). Socialization or rewards? Predicting U.S. scientist-media interactions. Journalism and Mass Communication Quarterly, 86(2), 299-314. Retrieved from http://aejmc.org/topics/wp-content/uploads/2009/09/3-Dunwoody-et-al.pdf