Is there really a crisis in access to the scientific literature? The answer is largely a matter of perspective.

If you ask librarians, the answer would be a resounding, “Yes!” Faced with stagnating and declining acquisition budgets and a growing production of scientific literature, librarians have been forced to make hard decisions on what to keep and what to cancel. As someone who conducted three journal cancellation projections, I have first-hand experience that these decisions are never easy to make, and each year, they only get harder. There is only so far you can go before the ability of the library to provide core titles to its researchers becomes compromised.

A matter of perspective. Image via Jean-Christophe Benoist.

If you ask researchers, however, you may hear a resounding, “No!” Big Deals have provided readers with access to more titles than ever before, and many subscription-access journals now provide hybrid open access publishing models. The number of open access journals and articles has been steadily increasing, as are the number of articles made freely-accessible through voluntary or mandated policies to self-archive. Add to this a growing list of programs that provide free access to developing countries, free access to backfiles, free access to research articles, and articles selected by editors to have important public health implications, and the access landscape starts looking very different.

From the perspective of readers, the world of access looks a lot different than the perspective of librarians.

In an article published in the July issue of the Journal of the Medical Library Association, The Impact of Free Access to the Scientific Literature: A Review of Recent Research,” Bill Walters and I review the existing research on the access problem and the impact of free access on the behavior of scientists as readers and authors. We also examine the extent to which the biomedical literature is used by the general public.

We report that there is little evidence to support the idea that there is a crisis in access to the scholarly literature, although this statement does come with a big caveat: most of the studies to date have focused on the behavior and opinion of scientific authors, most of whom are located at institutions with excellent access to the scientific literature.

In building the answer to this larger picture puzzle, we also found many missing pieces. For inquisitive graduate students looking for interesting research topics, the field is wide open.

First, little is known about whether free access is making a difference in non-research contexts–in teaching, clinical, extension, government and industry settings.

Second, there is a lack of understanding on the use of the primary scholarly literature by the lay public. We know that the public searches the Internet for health and medical topics, but their use of the Internet appears to be limited to public-directed summaries, such as those available at MedlinePlus and social media sites that help facilitate patient interest groups.

Last, most of what we know about access to the literature is provided through traditional channels–publishers and library intermediaries. If peer-to-peer sharing of scientific articles resembles the sharing of other forms of media–such as music and video–publisher website usage may capture only a small channel in a diverse multi-channel environment.

I’m no Pollyanna, and do not, for one instant, believe that access conditions are perfect. Access to the scientific literature could always be better. On the other hand, the data simply do not support a widely vocalized view that there is a crisis in access (or in scholarly communication in general).

The problem with getting an accurate view on the access situation is that the messages we hear are dominated by advocates and astroturfers — individuals who want to change the system and those who wish to profit from such a changed system. As David Crotty argued recently, it’s difficult to distinguish prescient views from outliers when the status quo remains largely silent.

Those who work for the complaints department know that silence usually means satisfied.

Enhanced by Zemanta
Phil Davis

Phil Davis

Phil Davis is a publishing consultant specializing in the statistical analysis of citation, readership, publication and survey data. He has a Ph.D. in science communication from Cornell University (2010), extensive experience as a science librarian (1995-2006) and was trained as a life scientist.


23 Thoughts on "Challenging the Access Crisis"

See the RIN report Overcoming barriers: access to research information for a through review of the extent to which readers in academia, healthcare, industry etc can access all sorts of information resources, not just journals. From memory, there were fewest problems for academics accessing journals, more significant problems for other types of publication and other sectors

I’d be one of that amorphous “public.” When I finished grad school my access to scholarly literature was essentially cut off since I can’t afford to pay for every article that looks like it might be interesting (and that’s when I able to find them) I’m an author (I write fiction) and I do a lot of research. A lot. But that research is now a lot less targeted and productive than I used to be able to do.

A recent letter in Nature Biotechnology provides some interesting perspective from a scientist at a small biotech company, with a few possible solutions for small businesses (that might be beneficial for publishers, too!).

Perhaps a better question to ask about access is not the either/or one but the qualified one, viz., access within how much time? Because people have become so used to virtually instantaneous access through the Internet, they may be inclined to say that they have no access if they cannot reach the desired source with, say, 24 hours or if they cannot do so online.

Thanks for writing this, Phil. I think your big caveat “most of the studies to date have focused on the behavior and opinion of scientific authors, most of whom are located at institutions with excellent access to the scientific literature”) is akin to “members of country clubs do not complain about access to golf courses”. In the scholarly field, there seems to be a black/white belief that there is a concentrated group of scholarly readers who are trained to read the sophisticated literature of their field and have satisfactory access, and then there is the broad morass of unaffiliated “laypeople” who are neither qualified nor interested to read this material. This view simplifies a large sub-category of unaffiliated users, namely trained professionals who graduated from those very same institutions and now work in corporate America, particularly those in small to mid-sized businesses .

According to the 2008 U.S. Census Bureau Statistic of U.S. Businesses (, there are over 6MM businesses with fewer than 500 employees. Health Care and Prof’l, Scientific & Tech Services represent over 11M employees, but as you can see, the sectors have interest in more than just STM but also Soc Sciences and Humanities. Overall, Small/Mid Businesses (SMB) represent nearly the same number of employees as Large Businesses (59M vs. 61M respectively) indicating a large pool of potential users. In short, there is a broad spectrum of unaffiliated users who fill in the middle between Institutional researchers and laypeople, and they quite likely comprise a substantial amount of the hundreds of millions of so-called “turn-aways” that scholarly publisher sites experience on an annual basis.

Thanks for this contribution. In the review paper, we also explore the studies that deal with use of the scientific literature by laypersons. Unfortunately, all of these studies have problems in defining the primary scientific literature and distinguishing it from the popular literature. This is particularly true for topics like health and medicine where there is a great deal of interest outside of academia.

It is also true, however, that some universities now offer access to journal and other library databases to alumni who join up to become Alumni Association members. This is true for Penn State, for instance, and gives me access to all of the materials i could access while working there.

In looking at, it looks to me like Penn State has negotiated alumni access to a very small percent of the e-resources they license for faculty and students: (there are hundreds listed) or e-journals of which there are approximately 70,000 listed in their A-Z list. Libraries simply cannot afford to license a lot on behalf of their alumni and many providers don’t have reasonable options for this. This is why public institutions strive to include “walk-ins” in all their licenses.

Perhaps if I were a scientists, i would find the access insufficient. But as someone who works mainly with the humanities and social sciences, the Muse, JSTOR, and ProQuest databases satisfy almost all of my needs.

Ah, but if libraries only had to support humanists, I suspect we wouldn’t even be having this discussion. Either that, or libraries would be supporting you very, very well. 😉

If only there were some way to ban the term “crisis” from policy discourse. Someone should do a fun catalog of all the supposed crises being cited today, no doubt thousands.

The main point is that there is a complex technological revolution going on in scientific communication, with lots of interesting opportunities. An opportunity to make things better is not a crisis. As you note, what we need is a much better understanding of what is actually going on, to see where new access policies can make a difference.

At this point we need experimentation, not regimentation.

Phil, The term ‘crisis’ in this context, whether you attach the term serials, library or access, is not resonating with enough of those it is intended to influence: researchers and authors of papers. However, you note the improvement in access, in particular the growth of open access. You do not suggest this is a bad thing. Where should we draw the line? Some of us have been arguing for 100% open access to research publications, meaning access to at least an author-produced version of published works. Many research funding councils would appear to agree. I believe it follows from your analysis this is where we should be heading, but from my point of view it is all too slow.

You have long questioned the case for open access. I hope to the extent that if you are for open access, or even not against it, that you are motivated to seek the best outcomes on access, so that we should avoid the wrong solutions for the wrong reasons. In this case I agree, the use of ‘crisis’ is unhelpful, but beyond that here you seem to lose the thread about where we should be heading and what would be the best approach to continuing the improvement in access provided by new opportunities in the digital world. “I’m no Pollyanna, and do not, for one instant, believe that access conditions are perfect. Access to the scientific literature could always be better.” By then emphasising lack of ‘crisis’ again, you are avoiding the most important part, the access trend that you have observed and how to take it forward.

There is another group of dis-enfranchised user: small government and not for profit agencies of various sorts with scientists and researchers dealing with technical issues but having extremely limited access to the literature. The budgets for these agencies are small, they typical have no (or a minute) library, and scientists report up to bureaucrats who no understanding of the role of the literature in a scientist’s execution of his job. And then there are retirees from various sectors, still inquisitive and active, pursuing their own interests and wanting to satisfy their intellectual curiosity, but finding them now cutoff from resources they have been accustomed to use.

I point out in our review the dearth of studies surrounding access to the groups you identified above. Mark Ware (citation 26 in the review) did study small-to-mid-sized companies in the UK. Quoting from our paper:

More than 70% of respondents claimed that they had “very easy” or “fairly easy” access to the journal literature, and 60% reported that their level of access had gotten better over the past 5 years. Among enterprises with more than 250 employees, 82% reported having easy access to the journal literature. Ware’s study used a sample of convenience, however, and only 4% of potential respondents completed the survey.

Just to clarify: the 70% who claimed “very easy” or “fairly easy” access was a sub-group of small/mid-size business respondents who had previously rated access to journal articles as important.

For all SME respondents: 2% rated access as “Excellent”, 26% as “Good”, 56% as “Varies”, 14% as “Poor” and 3% as “Very Poor”.

Bill, thanks for the further clarification. Mark Ware was right to focus on those who rely on the journal literature as it makes little sense to ask those who have little need/desire to read the academic and professional literature to rate their access to it.

I think it’s unclear. In his report, “Important” required a ranking of 6 or higher out of 7 which is a pretty high bar. And his study also indicated that the average SME respondent reads 112 articles per year which would suggest that it’s pretty important even for those who may not have answered it as such.

Certainly there is a large section of the general public which wants to access the scholarly and technical literature. Although astrophysics and physics are at least as difficult for the lay person to understand as other scholarly areas, the ADS ( alone last year answered over 30 million queries from the general public, 60% with abstracts, 40% with full text. I imagine PubMed has much more than ten times more traffic than we do.

A log analysis (see Kurtz & Bollen ARIST 2010) shows that these are not research queries, but a sampling of the actual queries shows that a large fraction of them are seeking information contained in the downloaded document. For the vast majority of these general public queries access after a one year embargo period would be fine, as there is no bias toward newer articles.

Access has a different meaning for researchers in the areas of text mining and large scale semantic data bases. Here only PMC with the NLM-DTD provide a basis for researchers outside of the world of text providers to attempt to create next generation research tools, and even there the data is a year out of date. Organizations which provide text (e.g. publishers) are also working with there own corpuses, but this is highly restrictive. The ADS, for example, has currently 0.7 FTE working on these issues.

The fact that changes in access (open to closed or closed to open) do not presage changes in citation rates (e.g. or any of several of Phil’s papers) shows directly that researchers do not have any major difficulty accessing the journals they publish in. This is, however, a very static analysis. Opening up access beyond their narrow focus groups is necessary for us to take the best advantage of the massive changes as society moves from paper based communication to electronic.

Access can easily get both better and worse at the same time, as this is a complex system. For example access to one’s favorite, narrowly focused journals can get better while the number of unavailable journals in all disciplines increases dramatically.

I have a special problem with researchers who say they can follow their field more easily, namely that there is much more than one’s field to follow. There is also one’s methods, math or technologies, which may span many fields. For example, a recent advance in Monte Carlo method, developed in forest management, has implications for nuclear medicine and even for most of science.

It is not a question just of the ability to follow one’s field, it is a question of the efficient diffusion of new knowledge. Today a proposal is expected to demonstrate knowledge of one’s field. Tomorrow we should expect a broader reach. This is what access is really all about, the acceleration of diffusion.

Comments are closed.