Magazine article Information Outlook

The Librarianship Conference Report: Convincing Evidence; Seeking out the Best Available Evidence

Magazine article Information Outlook

The Librarianship Conference Report: Convincing Evidence; Seeking out the Best Available Evidence

Article excerpt

In the January 2003 issue of Information Outlook, SLA's Research Committee, along with Joanne Gard Marshall, discussed how SLA's research statement is based on the concept of evidence-based librarianship (EBL). (1) EBL was developed by medical librarians who sought to apply the principles of evidence-based medicine to our profession. However, the basic principles of EBL are applicable to all areas of librarianship, regardless of library type or subject specialization. The main thrust of EBL is that librarians should base their professional decisions and actions on the best available evidence.

[ILLUSTRATION OMITTED]

I must confess that before reading the Information Outlook article, my impression was that EBL was just something that health librarians did. It really wasn't applicable to me in my position as an academic business librarian. The article, however, piqued my curiosity. Coincidentally, my employer, the University of Alberta Libraries, was hosting the second biennial International Evidence Based Librarianship Conference on June 4-6, 2003. I attended the conference with little knowledge of EBL.

What Is Evidence?

The presenters and health librarians in attendance were committed to making EBL principles fundamental to our profession and not just another management fad, such as management by objectives, reengineering, and total quality management (TQM). But, if practicing librarians are to make informed decisions based on the "evidence," what is the evidence and how do we lay our hands on it? In an introductory session titled "An ABC of EBL: What Is It and Where Has It Come From?" Andrew Booth and Jonathan Eldredge defined evidence as findings reported in the literature that utilize research methodology. They referred to EBL as a hierarchy of evidence determined by the research methodological rigor employed. The methodologies used in higher levels of evidence minimize bias. Human and systemic biases are more likely to occur in the lower levels of evidence.

Random controlled trials (RCTs) are considered to be the gold standard for research rigor. RCTs involve subjects or library users/clients who are randomly assigned to one of two or more groups. Each group is subjected to different interventions or no intervention at all (e.g., instruction would be an intervention). All subjects in the different groups are studied to measure the effects, if any, of the intervention. (2) Next on the hierarchy are controlled comparison studies, followed by cohort design studies, descriptive surveys, decision analyses, case studies, and, finally, qualitative research (e.g., focus groups, ethnographic studies). (3) As someone coming from a social science background, I found the evidence-based hierarchy very "discipline centric." All of the research methodologies employed by social scientists are ranked at the bottom of the hierarchy.

Supporters of EBL emphasize that it is not a substitute for a librarian's personal experience and that the evidence must be moderated by local circumstances. They are also quick to note that not all decisions warrant use of evidence, and that deadlines may preclude or limit the ability to integrate research into practice.

Bibliomining as Evidence

Scott Nicholson, School of Information Studies, Syracuse University, suggested a different approach to EBL. He has coined the term bibliomining to refer to library data mining. Most libraries already collect and/or generate substantial datasets through common functional processes and customer service applications, such as collection assessment, circulation, and end-user searching. Nicholson talked about applying statistical tools to these data in order to build a different form of evidence base. Bibliomining looks for patterns from system data in order to identify and/or understand different user communities. Aggregated data (which most libraries tend to use) do not provide patterns of user information. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.