Microsoft Word - COMM Evidence Based Library and Information Practice 2008, 3:3 86 Evidence Based Library and Information Practice Commentary Friendly Skepticism about Evidence Based Library and Information Practice Marcus A. Banks Manager of Education and Information Services UC San Francisco Library San Francisco, California, United States of America E-mail: marcus.banks@library.ucsf.edu Received: 01 Sept 2008 Accepted: 02 Sept 2008 © 2008 Banks. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Introduction On July 12, 2008 I blogged up a “screed” about evidence based librarianship, and realised almost immediately that I should have more accurately discussed “evidence based library and information practice,” or EBLIP (Banks). By July 13 two leaders in the EBLIP community, Pam Ryan and Denise Koufogiannakis, had responded eloquently to this post. The next day former JMLA editor T. Scott Plutchak pointed me to the video of a “friendly debate” he participated in with Andrew Booth about EBLIP last year (Booth and Plutchak). Finally, on July 15 Sarah McCord contextualised the conversation by pointing to a recent JAMA paper about varieties of valid evidence (Berwick), to which Ryan responded by noting examples of discussion in the EBLIP community about similar issues (Given; Ryan). I offer this excessively detailed chronology as anecdotal evidence for the power of blogs to rapidly generate thoughtful discussion. While the reaction to many blog posts can dwindle down to flame wars and insults, to me this experience demonstrates the potential of blogs to be a positive professional communication vehicle. I’m willing to stake a good deal on this slight evidence, even though it is undeniably flimsy by the standards of EBLIP. I have similar fuzzy feelings for wikis; for example, I find the “evidence based librarianship” wiki managed by Dean Giustini to be an excellent source of key information about this topic (Giustini). My personal experience with this wiki is my only evidence for such a sweeping endorsement. Booth, one of the chief proponents of EBLIP, recently lamented a general “evaluation bypass” as librarians rush to try out “Web 2.0” tools like blogs and wikis (Booth). Evidence Based Library and Information Practice 2008, 3:3 87 Surely my soft-minded embrace of blogs and wikis would not pass muster with a data-driven man like Booth. And yet, I believe that a sense of infectious experimentation should also be part of our professional toolkit. I suspect that Booth would not disagree with me if I phrase the matter this way, and would protest that he only wants to see “evaluation alongside innovation.” And in turn how could I object to something so equally obvious and compelling? So ultimately this disagreement turns not on ends but on emphases. I would be the first to cheer for a well-designed, randomised, replicable trial with the potential to improve a tangible aspect of information practice. But for me such studies do not have a monopoly on explanatory power. Depending on the context, the so-called evidence pyramid can be a political statement as much as a statement of fact, and for many types of questions, the randomised controlled trial is neither possible nor desirable (Sladen Library; Berwick). EBLIP defenders might object that this is a caricature that frames the matter too sharply. But I believe it is a fair response to the rhetorical thrust of the EBLIP movement. Contributions of EBLIP Before dwelling upon points of disagreement, it is important to recognise the contributions of EBLIP. The current definition of evidence based library and information practice is, “an approach to information practice that promotes the collection, interpretation, and integration of valid, important and applicable user-reported, practitioner- observed and research-derived evidence” (Giustini). It is indeed critically important for librarians to seriously examine why we do what we do, because otherwise the crutch that “we’ve always done it that way” can become seductive. This definition promotes but does not command, and nicely balances the perspectives of users, librarians, and researchers. On a personal note, I sincerely appreciate the willingness of EBLIP’s proponents to engage with people of contrary viewpoints. In addition to her response on my blog, Koufogiannakis immediately invited me to contribute this commentary. The debate between Booth and Plutchak is another example of an open exchange. However much we may disagree, we should all remember that this truly is a dispute between friends. Friendly Disagreements So, where do I disagree with my friends who promote EBLIP? I find three principal points of contention, each of which I will address in turn: o EBLIP is not sufficiently responsive to the reality that local contexts can trump even the most rigorous evidence. o EBLIP cannot answer many types of questions, particularly value-laden questions. o EBLIP can stifle innovation, in effect if not in design. Not Responsive to Local Contexts: In my blog screed, I hypothesise about an academic library that decides to abandon its in-house document scanning service for faculty members. The evidence is unequivocal that this service, however convenient it might be, is no longer cost-effective for the library. In some libraries, this evidence would be persuasive, and in all libraries it should be. But in many libraries, empirically irrelevant Evidence Based Library and Information Practice 2008, 3:3 88 factors would prevail. Library staff may resist the cancellation of this service, seeing it as a core of what they do; or influential faculty members may lobby for it to continue, causing the director to see it as a “loss leader” that generates goodwill. These are among a host of reasons why even persuasive evidence is often insufficient. While the concept of EBLIP makes allowance for practical realities (such as the need to maintain goodwill), it runs the risk of losing coherence if it grants too many exceptions. Less doctrinally committed proponents of the use of evidence in libraries, such as myself, can have our cake and eat it too: we see the value of empirical evidence, but do not put disproportionate weight on this one form of knowing. Much of my critique echoes similar criticisms lodged against evidence based medicine (EBM), the movement in which EBLIP has its roots. Cohen et al. categorized many of the recurrent objections to EBM: an excessive reliance on empiricism; a narrow definition of what constitutes evidence; lack of evidence for the efficacy of EBM itself; limited utility for individual patient care (which is analogous to the limited utility of EBLIP for local library contexts); and a threat to the autonomy of the doctor-patient relationship (Cohen et al.). Proponents of EBM are likely to find these objections to be spurious (Strauss and McAlister). I find them to be persuasive, as long as there is no backsliding towards the older model in which physicians had absolute authority and did not need to consult anyone. The same holds for concerns about EBLIP; they have merit, but I am not arguing that librarians should base our efforts on no evidence whatsoever. Cannot Answer All Questions, Especially Value- Laden Questions: In his debate with Plutchak, Booth’s disdain for answering “philosophical” questions is striking. He prides himself on being a “pragmatist,” whose chief concern is with answering the nitty-gritty questions associated with running a library. I find this to be worthy but narrow. While excessive theorising about the role and purpose of libraries can become tiresome, we are now living through a long period of profound (and to my mind, enriching) professional destabilization brought on by the Internet. It is a critical time for deep and unstructured thinking about our future. Core values of what it means to be a librarian are at stake, and many of the questions we’re grappling with will never have “right” answers. Sometimes the only way to learn is to stumble through the dark until you discover where you are going. Stifling of Innovation: Koufogiannakis has responded to the claim that EBLIP can unwittingly stifle innovation, by positing that there is no inherent tension between the two (Koufogiannakis). She argues that the two actually go hand in hand; an innovative product or service should be grounded in whatever evidence is available, and offered in a way that facilitates ongoing evaluation and refinement. While this is an admirable middle ground, Booth’s concern about an evaluation bypass with regard to the use of Web 2.0 tools in libraries is a more typical EBLIP formulation. He is right that evaluation of these tools’ effectiveness is necessary, but for the time being it is more important to adopt a flexible and experimental mindset. Here’s a slightly absurd parenting analogy to illustrate my point: is it better to let your children play and then do their homework, as long as they get the homework done? Or does the homework always come first, even if it takes so long that it’s dark outside by the time they finish? I say, let the kids play. If they never finish their homework — i.e., if there are still no serious analyses of the use Evidence Based Library and Information Practice 2008, 3:3 89 of Web 2.0 tools in libraries after several more years — then we have a problem. But right now, it’s light outside. Conclusion and Détente In the discussion that ensued after writing my blog screed, I noted that librarianship is merely the “object under discussion” for what is ultimately a disagreement about the nature of knowledge and how to approach the world. My (somewhat tongue-in-cheek) surmise is that most proponents of EBLIP are Sensing (“S”) on the Myers-Briggs Type Indicator, while I rely (sometimes overmuch) on Intuition (“N”) (Wikipedia). In order to make good decisions, S’s want as much data — aka “evidence” — as they can find. N’s are more comfortable making inferences from limited information. I would rather make a mistake based on imperfect information, than wait for superior information only to see an important opportunity pass by. The case for waiting and gathering more information is just as valid. This personality distinction seems hardwired, so I predict that there will be friendly jostling about the validity of EBLIP in our professional community for years to come. What is not in dispute is that librarianship is a deeply service-oriented, collaborative profession. In the book People Come First, Michael Gorman states it well: “Libraries are in the service business. The most important product they have is service” (Montanelli and Stenstrom 2). One of the core satisfactions of librarianship is helping people find that critical nugget of information they would not have found on their own; this is true for all librarians, whether they are explicitly research- oriented or not. My colleagues in the Medical Library Association recently released the well- developed “Research Imperative” (Medical Library Association). Superb research skills are definitely imperative, but for some of us and not for all. The rhetoric of EBLIP strongly suggests that we should all become research experts, but only a subset of librarians will ever have the inclination to develop a sophisticated understanding of research methods. In my view it would be more fruitful to groom this cadre of research leaders. Then the rest of us could learn from these pathfinders as we carry out the essential business of evaluating our work. As long as we don’t stretch it past the breaking point, EBLIP can be a vital philosophical asset for our profession. Works Cited Banks, Marcus. “Thoughts on Evidence Based Librarianship: A Screed.” 12 July 2008. 2 Sept. 2008. . Berwick, Donald M. “The Science of Improvement.” Journal of the American Medical Association 299.10 (2008): 1182-84. Booth, Andrew. “Blogs, Wikis and Podcasts: The ‘Evaluation Bypass’ in Action?” Health Information and Libraries Journal 24.4 (2007): 298-302. Booth, Andrew, and T. Scott Plutchak. “EBLIP: Clear, Simple, and Wrong? A Friendly Debate.” 09 May 2007. 2 Sept. 2008 . Cohen, Aaron M., P Zoe Stavri, and William R. Hersh. “A Categorization and Analysis of the Evidence Based Library and Information Practice 2008, 3:3 90 Criticisms of Evidence Based Medicine.” International Journal of Medical Informatics 73.1 (2004): 35- 43. Giustini, Dean. “Evidence-Based Librarianship.” 11 August 2008. 2 Sept. 2008 . Given, Lisa. “Qualitative Research in Evidence-Based Practice: A Valuable Partnership.” Library Hi- Tech 24.3 (2006): 376-86. Koufogiannakis, Denise. “On Evidence Based Practice and Innovation.” Evidence Based Library and Information Practice 2.4 (2007): 108-11. 2 Sept. 2008 . Medical Library Association. “The Research Imperative: The Research Policy Statement of the Medical Library Association.” 14 Sept. 2007. 2 Sept. 2008 Montanelli, Dale S., and Patricia F. Stenstrom, eds. People Come First: User Centered Academic Library Service. Chicago: Association of College and Research Libraries, 1999. Ryan, Pam. “EBL and Library Assessment: Two Solitudes?” Evidence Based Library and Information Practice. 1.4 (2006): 77- 80. 2 Sept. 2008 . Sladen Library. “Guide to Sladen Library Evidence Based Medicine Resources.” 2 Sept. 2008 . Straus, Sharon E., and Finlay A. McAlister. “Evidence-Based Medicine: A Commentary on Common Criticisms.” Canadian Medical Association Journal 163.7 (2000): 837-41. 2 Sept. 2008 . Wikipedia. “Myers-Briggs Type Indicator.” 31 Aug. 2008. 2 Sept. .