Crowdsourcing Reference Help: Using Technology to Help Users Help Each Other

Ilana Stonebraker,Tao Zhang

2015-01-01

Abstract:Introduction: What is Crowdsourcing in Libraries? Crowdsourcing has recently been defined as an “online distributed problem-solving and production model that leverages the collective intelligence of online communities to serve specific organizational goals.”1 Crowdsourcing encourages users to participate in knowledge creation with other members of their academic community. In both the library and public contexts, crowdsourcing has emerged as an effective model for incorporation of new ideas and perspectives. We have developed a crowdsourced reference system, CrowdAsk, in order to provide additional useful information for students and assist in fulfilling our mission as librarians to educate and encourage an information literate and informed populace. By helping users help each other, we solidify our relationship with our expert users as well as better serving non-expert users of the community. Crowdsourcing was first coined in 2006, and is derivative of the term outsourcing.2 Within libraries, crowdsourcing has become an effective means of leveraging online collective intelligence to transcribe and catalog numerous items. One of the examples is the New York Public Library’s (NYPL) 2011 “What’s On the Menu?” project. With assistance from an IMLS Sparks! Ignition Grant, the NYPL utilized its online user base to transcribe 9,000 restaurant menus.3 The crowdsourcing concept spanned multiple similar archival transcription projects including the Australian Newspapers Transcription Project and Family Search Indexing, among others. We believe that a similar model of utilizing collective intelligence can be implemented in libraries’ reference workflows. While our websites grow, the traditional model of librarians passively waiting for users to seek help while most users find help outside libraries, has not changed. Users, especially students, tend to seek reference help from faculty advisors and their peers. This kind of knowledge sharing does not have a well-structured platform within the library environment and expert knowledge is not well utilized, especially for experts outside traditional library roles. There have been some efforts to create searchable help content, but adding new questions and answers, as well as validating and updating answers that may be out of date, inevitably takes up librarians’ valuable time. This paper examines the various ways crowdsourcing help can provide academic users with quality answers and engage users with libraries in ways previously not administratively or technologically possible. Through crowdsourcing, libraries can bring new perspectives to problems. CrowdAsk is a crowdsourced questions and answers system for library help developed at Purdue University Libraries with support from an IMLS Sparks! Ignition Grant and is openly available at http://crowdask.lib.purdue.edu.