How libraries and publishers can work together

New trends and challenges in research information make it crucial to ‘work smarter’

The Author

Alicia Wise, PhDAs Director of Universal Access for Elsevier, Dr. Alicia Wise (@wisealic) is responsible for delivering Elsevier’s vision for universal access to high-quality scientific publications. She leads strategy and policy in areas such as open access, philanthropic access programs, content accessibility, and access technologies. Based in Oxford, she has a PhD in anthropology from the University of North Carolina at Chapel Hill.[divider]

Recently, I was invited by Professor Qiang Zhu, Director of the Peking University Library, to speak at his library’s 110th anniversary celebration.

The topic – how libraries and publishers can work together – is extremely close to my heart, and the audience was a very interesting mix of Chinese and international (largely North American) library colleagues.

One of the nicest things for me, besides the opportunity to visit Beijing again, was that the topic was broader than open access. Don’t get me wrong: open access is hugely important, Elsevier is doing an extraordinary amount with open-access publishing (see this recent article in Elsevier Connect), and I love talking about this at conferences. Still, it was nice to put that work into the broader context of how libraries and publishers collaborate.

The key thing that unites academic librarians and publishers is that we all serve science.

Here’s a summary of my presentation:

Research is pivotal to economic growth and addressing societal challenges. The world of research is large and growing, with global research and development (R&D) spending $1.4 trillion in 2012 according to Battelle R&D Magazine’s Annual Global Funding Forecast. R&D spending as percent of gross domestic product (GDP) has been relatively stable at around 2.5 percent in developed countries, and it increased from around 1 percent to 2 percent between 2000 and 2010, according to the OECD. Even in these difficult economic times, governments around the world are protecting their investment in R&D to help drive economic recovery.

Growth in research & development investmentGrowth in research & development investment is strongly correlated with growth in research outputs. (Sources: internal Elsevier analysis based on Scopus data and “Main Science and Technology Indicators,” OECD Science, Technology and R&D Statistics database)

While this investment in R&D for economic growth is of course a good thing, it drives an increase in the number of researchers and research output, which can cause challenges for both librarians and publishers. Librarians struggle to afford this explosion in content, and publishers have to make significant annual efficiency gains simply to try and keep pace.

Four key trends

Given this context, I believe four key trends will continue to increase the value of research information, and it is therefore crucial that librarians and publishers work smarter rather than relying on our finite capacity to work ever harder. [pullquote align="right"]It is crucial that librarians and publishers work smarter rather than relying on our finite capacity to work ever harder.[/pullquote]These trends are that research is increasingly interdisciplinary, increasingly collaborative across national boundaries, increasingly done in emergent countries such as China, and increasingly is data intensive.

A unique vantage point

Librarians and publishers have a unique vantage point on research.

Each year scientific publishers handle over 3 million submitted articles, coordinate the work of more than 300,000 peer reviewers, and publish over 1.5 million articles. Research libraries make these articles available to 30 million readers worldwide, who download more than 2 billion articles each year and make 30 million article citations.

What unites librarians and publishers

While there are things that divide librarians and publishers there are more things that unite us:

1. Our contributions to research start with the highest quality information, which increases the efficiency of researchers using this information. For example, Elsevier handles about 1 million submitted articles each year. This year, submissions to our journals have grown by more than 10 percent over the previous year. Through investments to drive quality, such as our new Article of the Future format, researchers can, on average, determine the relevance of an article 24 seconds faster, spend 60 percent more time interacting with relevant articles, and are five times more likely to use underlying data and tools.

2. We share the responsibility of providing researchers with access to information, now and in the future. 10 million researchers from around the world use Elsevier’s platforms each month. 85 percent of researchers have access to this platform, including researchers in 105 low- and middle-income countries through Research4Life. But access is important for others, too, which is why we have a wide array of programs to enable access by members of the public and by businesses big and small. As interest grows in shifting the costs of the scholarly publishing system from user to researcher/funder, we are well placed to scale up open-access publishing. We have already launched 28 fully open-access titles, we have more than 1500 hybrid open access titles, we have more than 80 titles with open archives, and all our author agreements allow voluntary posting of accepted manuscripts.

3. In the digital age, the roles of publishers and librarians change. We support you in making that transition, and we’re changing too. Together librarians and publishers work to ensure the digital preservation of the scholarly record. By providing analytic tools such as SciVal, libraries help institutions compete for research talent and resources and make smarter investment decisions.

4. Young researchers are the future of science. We help secure that future by supporting and developing the global research community. Librarians already know about the myriad ways the library community supports students, but may not be as familiar with some of the ways publishers contribute. For example, more than 18,000 students have attended the Publishing Connect author workshops, which are presented by Elsevier staff and editors in partnership with the local library.  Additionally, some 250,000 researchers worldwide have connected to similar Publishing Connect online webcasts. We have programs that support the development of library leaders of the future, too.

I always enjoy giving presentations because I always learn something in the process. And this time, it was a pretty inspirational something: together, librarians and publishers provide quality information, which drives quality research, which improves the quality of life.

comments powered by Disqus

5 Archived Comments

Alicia Wise December 7, 2012 at 5:19 pm

Hi David,

Thank you very much for the comment. This is my first contribution to Elsevier Connect so I’m glad to see it’s being read and that you took the time to respond.

I’ve posted a few more slides which relate to your points on slideshare ( Investment in R&D is rather resilient in developed countries, and this investment continues to grow in developing countries as a proportion of GDP. In that context I think my interpretation of these data holds: investment in R&D continues to grow as do the number of researchers and research articles. Yes, there are likely to be more complex connections – for example lag times between funding and subsequent impact on researchers and outputs such as articles – but there are nevertheless connections (see, for example:

All the best,



Alison Bert December 6, 2012 at 10:15 pm

Thank you for your comment, Dr. Roberts. Alicia will get back to you soon.

David Roberts January 1, 2003 at 12:00 am

Hi Alicia, do you have time series data corresponding to the figure? That R&D expenditure and number of articles published are correlated is not at all surprising. And it should be followed up by a chart with R&D expenditure as a percentage of GDP, rather than absolute dollar figures (and a time series for this). What I would be interested to see is how (if at all) the change in the number of articles published relates to shifting patterns of absolute/%GDP R&D funding. I ask this because the reason given by Elsevier (and other publishers) for increasing journal prices so dramatically each year is that scholarly output is increasing. Your argument is that

"... this investment in R&D for economic growth ... drives an increase in the number of researchers and research output"

whereas your graph doesn't show this at all, only that richer governments can afford to indirectly employ more researchers. I would imagine that in the current economic climate a number of developed countries are in fact cutting down on their spending on basic research (to some extent this is happening even here in Australia, where our economy is in far better shape than some other countries). If scholarly output flattens out, would publishers be happy to follow that trend with their prices? (Well, apart from the fact price increases are already built into existing contracts.) You also say "This year, submissions to our journals have grown by more than 10 percent over the previous year." whereas I don't think government R&D spending has increased by a similar figure.

I would argue that it is not (only) government spending that is driving the number of publications, but a change in publishing culture among researchers, an intensification of the publish-or-perish environment. The blame for this cannot be placed at the feet of government spending policy.

Also, I disagree with the subtly disingenuous logic of your paragraph

"While this investment in R&D for economic growth is of course a good thing, it drives an increase in the number of researchers and research output, which can cause challenges for both librarians and publishers. Librarians struggle to afford this explosion in content, and publishers have to make significant annual efficiency gains simply to try and keep pace."

which I take to be: governments spend more -> researchers output more -> librarians struggle with this -> publishers help. Rather, the real flow of effect is: (multiple complex factors) -> researchers output more -> publishers charge more -> librarians struggle with this.

I don't blame you for the faulty analysis of the figures, as I'm sure that was done by internal analysts (and I know *exactly* how that works, having been in a similar position myself before), but I do think it sad that a company that prides itself of furthering (among other things) science, would try to justify an argument by a graph that tells nothing of the sort.

David Roberts December 10, 2012 at 1:39 am

Thanks, Alicia, for having a closer look at this.

These would have been better (though not the best) figures to include in your article. However, as with all data, where are the footnotes? How is R&D spend defined? For instance, if I look at the document which details the coverage of the dataset from which the R&D spending and number of researchers data is derived, total national R&D spend seems to include business and enterprise expenditure, and number of researchers includes those in industry. If the argument is about taxpayer money, then I would, if I were performing this analysis myself, restrict myself to government-funded research, and researchers who are employed (directly or indirectly) by the government. In this I would exclude the military or intelligence sectors because their work is (by and large) not intended for publication. I've not seen any arguments by open access advocates that cover private-sector research.

Looking at your slide 3 (R&D spending as %GDP in developed and developing countries) I see that you've placed developed markets (=EU27+US+Japan) gross expenditure on R&D (GERD) as %GDP at roughly 2.3%, and quote this figure in your article: "R&D spending as percent of gross domestic product (GDP) has been relatively stable at around 2.5 percent in developed countries"), whereas from figure 2 (page 14) in the (100+ page!) article you link to, the _public sector_ figure is closer to 0.75% of GDP spend on R&D. The goverment's share of GERD in the US, EU and Japan decreased between the years 1995 and 2005 (table 3, page 60), and the business enterprise sector's share has risen, though these figures unfortunately don't include higher education. So I'm afraid I'm slightly less impressed by the choice of graphs in the slides as they don't clarify the public/private split in R&D funding.

If the majority of Elsevier's customers are libraries in higher education institutions, and this certainly where I see the argument coming from, then it would be good to see the figures broken down to the level of higher education spending and researchers (this data is available in the data set, I've checked).

I would not go so far as saying at this point that your thesis is invalid, merely that your supporting data at this point is, firstly, not in the context of the general argument that is occurring around taxpayer-funded research, and secondly, doesn't go into enough detail to independently verify what you say is holding (definitions of the quantities measured are important for this). One can do statistical analyses on the sorts of questions you are trying to answer, and a simple scatterplot as in the original post does not answer them, I'm afraid to say.

Section 4.2.1 of, a very short section at the end, _does_ detail some more in-depth statistical modelling of the relationship between public R&D spending, number of researchers and number of papers output. This is what you should have pointed to, as the people interested in these questions are generally highly numerate, rather than the paper as a whole. However I'm slightly nonplussed by some of the choices of variables. I can't find where they list the source of their data on numbers of publications (the list of sources is on pp 80-85), and I should point out that these are only available for the years 2000-2004 (see page 88). (There is also a shocking systematic error on p88 for half of the last column, but this could happen to anyone). From that table one can see that number of publications (however that is measured) grew for the US by 7% in the years 2000-04, shrank in the UK by about 2%, and was pretty much flat in France and Germany. Italy has 16% increase, Spain a 25% increase and Poland a 32% increase. Countries which are newer member states in the EU seemed to have the biggest increases (though they has a lot smaller baseline in general). Unfortunately there isn't any aggregated data for the EU, for instance, nor for other regions such as China, where I imagine most of the growth is happening of late.

With all that said, this is how I would present the data which the original figure attempted to indicate, drawing on the article you linked.

"The authors of the report find that the number of publications in the EU27+US+Japan between 2000 and 2004 was proportional, all other things being equal, to the square root of (the sum of government-funded R&D and higher-ed-sector R&D, as a percentage of GDP)."

A positive correlation, yes, but not a linear one, and one that still deserves further investigation, if one wanted to continue flogging what one might say is a horse that has rung down the curtain and joined the choir invisible.

One could write and publish a paper in a peer-reviewed journal on this, I'm sure...

Alicia Wise December 11, 2012 at 11:15 am

Thanks, David

Agree that this area is worth exploring in more detail, and is worthy of a journal article as you suggest!

With kind wishes,




Related Stories