Without question, the report, which looked at 4.2 million papers published over three decades, “indicates a remarkable and nearly universal rise since 1975 in the frequency of collaborations between authors located at different universities.” But those collaborations aren’t necessarily happening between researchers located on opposite sides of the planet. Over the 30-year interval in question, the mean distance between collaborators in science and engineering fields grew from 750 miles to 800 miles. They conclude: “It is not the length of a scientist’s reach that has changed but rather the incidence of reaching across university boundaries.”

But the more interesting findings involve social distance rather than physical distance.

The data suggests that collaborations between researchers at different top-tier universities have a measurably larger impact than papers published by collaborators within a single top-tier school. For science and engineering fields, a cross-school study is about 6 percent more likely to be high impact than work published by colleagues on the same campus. In the social sciences, this probability increases to almost 12 percent.

The authors divide schools into four tiers and note that based on the number of schools in each tier, researchers at the top-flight institutions tend to work together more frequently than expected. Likewise for pairings that involve two schools in the fourth, or bottom, tier. While a same-tier pairing improves impact ratings for the first through third tiers, for a bottom-tier school, it can actually damage the probability of producing a high impact paper. “Partnership choice increasingly appears to be based on who the collaborators are rather than where they are,” the authors write, “with an emphasis on in-group status matching.”

The analysis leads them to conclude that while geography is measurably less important to doing high-impact research, social stratification actually helps increase the gravitational pull of already dominant academic centers.

According to the Chronicle, researchers conducting the study searched a database of millions of court documents provided by Merck for liability cases involving its drug Vioxx, which was pulled from the market because of safety concerns. They found that papers reporting clinical trial results seemed to be written by Merck employees, only to have the names of academic authors added later.

The report includes an editorial by the editor-in-chief and executive deputy editor of JAMA, condemning such ghostwriting as “unprofessional and demeaning to the medical profession and to scientific research.” The editorial includes new guidelines to prevent such behavior: in particular, it calls for authors to disclose any financial conflicts of interest to their supervisors.

The study adds to the growing list (subscription) of recent similar breaches of scientific integrity. A clear and simple conflict of interest standard adopted by all scientific journals could go a long way to preventing future conflicts like this.

A provision in the Labor-HHS-Education bill (H.R.3043/S.1710) before the Senate today will change the NIH public access policy and require that researchers publish all peer-reviewed articles produced from research conducted with NIH funds in open access journals. The House version of the bill requires that the NIH provide access to the articles within 12 months of their appearance in refereed journals. Under the current system, the NIH simply “requests and strongly encourages” making final peer-reviewed manuscripts publicly available.

Unfortunately, Senator James Inhofe (R-OK), introduced two ammendments Friday that would derail open access by placing decisions about making peer-reviewed research public in the hands of for-profit publishers, rather than in the hands of scientists. This would limit the ability of the public and other researchers to benefit from taxpayer-funded research.

Inhofe’s proposed changes to the legislation would limit access to scientific work conducted in the service of the common good in order to protect the out-dated business model of some scientific publishers. Scientific progress requires access to shared knowledge, and the new NIH rules would be a strong mandate for knowledge that empowers scientists and the public alike.

For more on the issue:

Heather Morrison explains that the NIH already underwrites publications costs; while full and immediate open access would be in the public interest, the 12-month open access policy is a reasonable compromise (Via A Blog Around the Clock).

The Scientist: the NIH is the world’s largest biomed funder, and open access is a “natural next step.”

Scientific American ran a backgrounder on the issue in January.

U.S. labs that handle deadly germs have reported “100 accidents and missing shipments since 2003,” reports the AP. No one was hurt, but the number of incidents are going up with number of labs approved to handle the pathogens. The House Energy and Commerce Committee will hold a hearing today.

The latest on framing science: Matthew Nisbet & Dietram Scheufele in The Scientist on The Future of Public Engagement (now out from behind the subscription veil). Nisbet has it on Framing Science.

How the vagaries of scientific publishing don’t necessarily lead to better science getting published.

The EPA decided not to approve highly toxic methyl iodide as a pesticide. The agency balked after receiving a letter from 54 scientists, including 6 Nobel laureates, who “were astonished EPA was considering approving such a toxic chemical for agricultural use.” (Via Wired Science.)

PBS and Wired premier a new science magazine TV series last night, which (not to be confused with the publication’s blog) is called Wired Science.

Smithsonian magazine just issued a Fall 2007 special issue titled “37 Under 36: America’s Young Innovators in the Arts and Sciences,” comprised of 37 crisply written profiles of smart and creative young men and women making a difference. Oftentimes, the editorial content of magazine packages such as these come across as glossy as the accompanying photos, especially in special issues. In this case, the written words and the photos are equally compelling.