Free online access to academic journals increases readership, according to a new study, but it produces no more citations, undermining a key claim of the open access movement.

"There are benefits to providing free access to literature, but academics should not bank on being cited any more through open-access publishing," said communication postdoctoral associate Philip Davis, author of the study published online March 30 by the Federation of American Societies for Experimental Biology (FASEB) Journal.

For the past decade, research into whether free access to scientific literature leads to more citations produced inconsistent results, Davis said, which he attributes to weak methodology.

"It wasn't clear from these studies whether open access was increasing article citations or whether the association was merely spurious," Davis said.

In his study, Davis used randomized, controlled trials to find out what really drives citations. Publishers eager to know allowed Davis to randomly assign open access status to 712 articles appearing in 36 subscription-access journals in the sciences, social sciences and humanities, including articles in the journal Science.

Randomization enabled Davis to weed out other possible explanations for rates of citation. "Through the methodology of our approach, we were able to isolate the effect of access independent of any other possible explanations," Davis said. "We found consistently that free access increases readership from a broader group of readers, but shows no difference in citations to the articles over three years. Our results are consistent across time, disciplines and journals. We're pretty confident that this is a general finding."

Davis said because most academic authors are located at institutions with excellent access to the scholarly literature, it is not surprising that free access has little effect on article citations. He adds, "The real beneficiaries of open access may not be the research community, but communities of practice that consume, but rarely contribute to, the corpus of literature."

"Open access should be promoted because it leads to greater dissemination of scientific knowledge, not because of inaccurate claims that it leads to more citations, and that needs to be acknowledged."