Print version (Adobe Reader required)
Much has been written recently about the “library as place.” This essay approaches the question of library space philosophically, arguing that developing commercial attitudes toward space leads us away from more productive ways of conceiving libraries. A concept called Third Space is introduced, and its relevance to libraries and librarianship is explored. Third Space is defined and applied to various library concepts, especially information literacy. The article contends that thinking about Third Space can help libraries and librarians develop ways of working with increasingly diverse populations in increasingly dynamic contexts.

“Question: What is the first thing that you think of when you think of a library?

Answer: a place of mild climate where I can find adventures”¹

As Charles Osburn notes, “there has been a decided surge of interest in our professional literature about ‘the library as place.’”² This interest reflects various trends and emphases in libraries, especially the transformative social and technological changes that have demanded increasingly innovative thinking about what a library and a librarian should be. Collections, technology, and services can no longer be conceived in traditional twentieth-century terms. Libraries, with their historical ethos of free access for all, struggle to justify their existence in a world of 24/7 access increasingly evaluated by profit-based, commercial metrics. As we think about what library space and librarians should be and become, we need to think broadly and creatively about our options. We have barely begun to develop sophisticated frameworks for thinking about the future of the library as physical space. Libraries are complex institutions, and they need to respond to the demands of the present by adapting in a variety of ways. No doubt we need to justify our existence to our various funding agencies, which will involve economic arguments, but we also need to develop theories about library space that go beyond marketing services and managing buildings. We need to think about intentionally producing unique library spaces. I believe we must be conscious and ambitious about developing guiding theories and that a critical concept called Third Space can help us to do so.

The Conversation in Practice

When we talk about library space, we are usually talking about buildings.

Library buildings give form to the collections of libraries by providing appropriate space specifically designed to house and provide access to the holdings. They also provide other more “mythic” functions by intentionally symbolizing through architecture and design the values that libraries espouse. A number of converging forces have intensified recent questions of library space. Changing technologies have forced reconsideration of how buildings accommodate the new machines that provide service to modern libraries. Along with technical imperatives have come a series of human questions about the impact of new technologies on our ways of teaching, learning, and thinking. An entirely new vocabulary has emerged around learning spaces and how to conceptualize and create them. As Brown and Lippincott note, “New conceptions of the classroom are being driven by the emergence of new methods of teaching and learning, made possible by the rapid evolution and adoption of information technology.”³ We have come to think of learning as a constructive process, which has encouraged us to redesign schools and libraries to foster collaborative learning and active learning, and we are exploring digital environments as spaces we structure and design for learning, as well.⁴

Much of the energy behind these new conceptualizations has been fueled by fundamental questions of library legitimacy. The digital world is replacing libraries, this narrative argues. If we intend to remain relevant (or exist at all) we must adapt quickly to the technological challenges to library legitimacy. This adaptation demands that we compete with various entities that provide desired goods and services in our market. These entities include Google, which has claimed the information market, and also the bookstores and coffee shops that have capitalized on the market for comfortable physical space to interact with books. Space is therefore conceived as both physical and virtual, and libraries face competition in both realms. Consequently, during the past decade, much has been written about how libraries can respond to questions of space. Woven throughout the discussion we find a common anxiety about the changing nature of library space and what will happen as we continue to develop and deploy new technologies that displace or transform traditional libraries, demanding that we justify our stewardship and management of it.

In response to our challenges, we are regularly told that we need to run libraries more like businesses.⁵ ALA Editions’ advertisement for Hernon and Altman’s Assessing Service Quality reflects the concerns outlined above:

Because of technology, the old measures of service quality no longer apply. If libraries are to succeed, they must see themselves in competition with other institutions and sources of information—especially the Web—and make customers feel welcome and valued. [The authors] integrate the use of technology into the customer experience. They offer solid, practical ideas for developing a customer service plan that meets the library’s customer-focused mission, vision, and goals, challenging librarians to think about customer service in new ways.⁶

Another author makes the point that “The Internet, coffee shops, restaurants and even homes are all invading the territory once exclusive to libraries. Bookstores are consciously attempting to recreate the library atmosphere, encouraging customers to linger. … As a result, patrons are abandoning libraries for more favorable environments. Library users are choosing plush recliners and the aroma of coffee over the squeaking of wooden tables and buzzing of fluorescent lights.”⁷ We should note the level of threat implied in these comments. Librarians are “challenged” to think about customer service. Other competitors are “invading” library territory. They are “consciously” imitating libraries. Patrons are “abandoning” us. Anyone following the library literature recognizes such anxious claims, which have been with us for at least the past decade.

In responding to these threats, The Denver Public Library decided to become a “destination library.” To do so, they decided to implement “best ideas and practices in consumer merchandising and marketing and apply these to the library space.” Behaving more like a business meant that “new multiple copies of best sellers and media would be available quickly, displayed more like the local bookstore. Comfortable seating would be available, perhaps with a cafe nearby. The goal would be a popular customer-driven collection in an appealing space that would encourage visits.”⁸ Journals and conferences are infused with this perspective as we focus on marketing services with campaigns like @yourlibrary. Again, this idea of treating libraries like businesses is not new. The managerial segment of the profession has been borrowing techniques from business management for years. However, the idea that we need to market library space as a product that will attract library users seems new. In pointing to this phenomenon, my goal is not to raise the question of whether libraries should behave like businesses. Rather, I want to suggest that when we do, we create a specific kind of space. When we aim to compete with businesses, we infuse the building with advertising and the upbeat signage that “customers” know and recognize. In effect, rather than manage employees or collections or the physical plant, we are managing ambience, trying to create a place that feels familiar and good to the consumers of library services.

A large part of this effort goes into the aesthetics of library space. Demas and Sherer note that “after a generation of intense focus on building the virtual library, librarians have reawakened to the place-making role of the library building.” These authors advocate what they call “esprit de space.” They suggest that libraries should pursue “the timeless design goal of creating transcendent and transportive spaces: transcendent, in the sense of buildings that delimit physicality through imaginative understanding and application of virtues; and transportive, in design that uplifts the patron and enhances the unique experience of sensing past, present, and future simultaneously. It is this transcendent/transportive co-existence, with particular reference to its local, place-specific manifestations that distinguish a library with … esprit de place, or spirit of place.”⁹ Again, it is worth noting that libraries have long been concerned with the aesthetics of their buildings. Library Journal devotes one issue annually to photographs of the most innovative and beautiful new library buildings. Various consultants provide guidance in how to work with architects to develop buildings that both function well and also provide beauty and form that embody library values. Once again, though, we see the emerging emphasis on the feeling of library space and the importance of managing that space to attract and hold library users.

Younger users cause special anxiety, apparently, as a good deal of thinking goes into imagining aesthetically pleasing spaces for them. Kuzyk suggests that libraries need to “put the WOW back in children’s rooms.”¹⁰ Farrelly suggests that we need to compete with the bookstores for the loyalty of teens, noting that “libraries need to be more appealing to teens than Borders, Starbucks, and Barnes and Noble to attract young adults. We also need to do them one better.”¹¹ Gallo suggests that her experience working in a bookstore has provided her with strategies for using displays to attract teens. She suggests that we identify display areas creatively and use color to attract attention.¹² Bolan has been a prolific adviser to libraries about designing teen space. She asks us to consider what would happen “if teens suddenly found the library warm and inviting?”¹³ She has abundant advice about how to make the library an attractive destination for teens. She suggests seeking input through a teen advisory board, noting that “it’s crucial to make room for youngsters’ ideas in everything from creating an advisory board to planning a design team.”¹⁴ She suggests “finding that ambiguous teen style.”¹⁵ Ultimately, Bolan declares, “we’re in the midst of a teen revolution design-wise, that is.”¹⁶ Indeed, as libraries continue to market their services to young people, we see intense focus on the issues played out more generally in the library literature competition with bookstores and coffeehouses, design with the aesthetics of the customer in mind, and the general need to hold on to the library users of the future. The stakes are high.

Toward Critical Conceptions of Space

Two recent publications have approached space philosophically, and they have established a foundation for looking at libraries in the context of critical theories of space. Both these works aim explicitly at providing an alternative to the “library as business” way of looking at space. Interestingly, both these works depend on importing the interdisciplinary research on place studies into librarianship. Place studies can be understood as an effort to bring multiple critical perspectives to bear on the problem of how we use and define the spaces we share and manage. These studies are animated by awareness that when we create and occupy space, we define and develop that space (consciously or unconsciously) to embody cultural codes. Indeed, these studies share a fundamental assumption that place must be understood as the interaction between humans and natural forms. Culture creates space, and once we realize that fact, we can become more conscious and more intentional about what we create.

In “Regaining Place,” Charles B. Osburn argues persuasively that “place is worthy of the most serious consideration, especially at a time when so many fundamental options present themselves for the future of the library.”¹⁷ Osburn goes on to argue that space is “endowed with powerful properties … only by the beholder whose awareness of the experience generates it.”¹⁸ Ultimately the images people have in their minds about space “can have much or little to do with reality, for they are partial and may be either exaggerated or understated.”¹⁹ Osburn provides a useful perspective in that he acknowledges the subjective nature of experiencing space, which moves us past thinking that space is a stable commodity and that we can control how people experience it. Following Osburn’s logic, whatever we do with library space, people who enter libraries will experience that space in their own ways, perhaps as we intend, and perhaps not.

A more critical and more guided discussion of space occurs in the book Library as Place: History, Community, and Culture, edited by Buschman and Leckie. In their introductory essay, the editors summarize a range of theories that provide ways of thinking about space. This survey provides a valuable, concise introduction to the current state of space theories and libraries. Ultimately after presenting a range of critical perspectives, the authors suggest that Jürgen Habermas’s idea of the public sphere should form the central theoretical perspective for how libraries can define themselves and the space they construct. “It is Habermas,” they argue, “who allows us to make normative and democratic claims about libraries as places.”²⁰ This argument aligns with Osburn’s claim that places are social, cultural, and personal constructs that we hold in our minds. In brief, Habermas argues that the rise of the middle class from the eighteenth through the nineteenth centuries involved the development of public spaces where citizens discussed and debated the issues of the day. These debates followed rules of reason and persuasion (in the best Enlightenment tradition), so the most rational argument would prevail. This public sphere formed a critical function providing checks and balances on the powers of government, and it constituted a crucial element of early democracy and a way for the middle class to establish its influence and to define and express its public will.

In summarizing Habermas, Buschman and Leckie acknowledge “crucial problems” with the bourgeois public sphere. They note that in Habermas’s analysis, “the public sphere arose among a highly educated, cohesive class.”²¹ In his analysis of Habermas’s theories of the public sphere, Douglas Kellner succinctly summarizes the most problematic critique of Habermas. He contends that “while [Habermas’s] concept of the public sphere and democracy assume a liberal and populist celebration of diversity, tolerance, debate, and consensus, in actuality, the bourgeois public sphere was dominated by white, property-owning males.”²² Any consensus achieved in such a forum only legitimately reflects the opinions and interests of this narrow class. Summarizing Habermas, Buschman and Leckie suggest that the public sphere began to lose its sway when “democracy became a mass affair during the nineteenth century.”²³ With this change, politics became less reasoned, and competing interests became more effective at undermining the seriousness of conversation in the public sphere. Ultimately, despite its limitations, the authors note that “what we today understand of libraries as public space with democratic undertones is deeply embedded in the historical processes Habermas identifies.”²⁴ Our challenge today, it seems, lies in finding a new way to constitute a truly inclusive public sphere, one broadened beyond the homogeneity of the property-owning bourgeois class.

Habermas traces the decline of the public sphere to increasingly sophisticated capitalist practices that transformed critical citizens into uncritical consumers. These new capitalist practices emerged in the nineteenth century, and as a result, commercial space began to replace the intellectual space of the public sphere. Capitalism has grown increasingly more effective at defining space during the twentieth century. David Harvey argues convincingly that a new, faster form of capitalism began to emerge in 1971. Since then, this new and increasingly sophisticated capitalism has more powerfully defined cultural space. The new capitalism (sometimes called hyper-capitalism, fast-capitalism, or simply late capitalism) compresses space and place by developing increasingly sophisticated ways to collapse time and space to increase the rate of profit. This observation accounts for the fact that toward the end of the twentieth century, we began to experience “an intense phase of time-space compression.” Harvey asserts that “accelerating turnover time in production entails parallel accelerations in exchange and consumption.”²⁵ In other words, the faster we produce, the faster we need to consume to keep inventory from backing up. Profits depend on increasing speed.

Historically, space has presented a primary obstacle to this acceleration. Railroads, the telegraph, highways, steam shipping, the Suez Canal, the telephone, and ultimately the Internet, all these technologies have been deployed and perfected to “shrink” space to more rapidly move goods to market to drive commerce.²⁶ Ultimately, Harvey argues, capitalists learned to think of space as broken into distribution nodes connected by communication systems. The resulting fragmentation achieved “the annihilation of space through time.”²⁷ By finding ways to shrink space to speed up commerce, we have arrived at a point where space and time are transformed. Modern communications technologies now allow us to do almost anything almost instantly from almost anywhere. This annihilation and fragmentation of space has had profound consequences for culture. In capitalist culture, we now create disposable spaces and places that can be rapidly “turned over” for profit. Property can be bought, sold, and converted to new uses once it has been fragmented. Place can be played against place for profit.

Harvey argues that what we understand as postmodernism—the fragmentation of place and acceleration of time—results from these advanced capitalist practices. Harvey notes that one strategy for resisting postmodernism has been to “relaunch the Enlightenment project of universal human emancipation in a global space bound together through mechanisms of communication and social intervention.”²⁸ In this response, the autonomous human exercises rationality and free will and can marshal these resources to resist the effects of postmodernism. Habermas’s identification and promotion of a reconstituted public sphere in “global space” represents one such effort to “re-launch the Enlightenment project.” This solution, however, misses the source of the problem of postmodernism. Postmodernism is not a theoretical invention of the academy to be resisted intellectually. The condition of postmodernism results from the very real transformations in culture wrought by increasingly sophisticated capitalism. The Enlightenment project of human emancipation has been increasingly ineffective as a means to resist the capitalist restructuring of culture (hence the transformation of the public sphere chronicled by Habermas). While we might see ourselves as autonomous and rational, the culture we live in undermines our autonomy and subverts our rationality. This new postmodern context must be understood as the defining reality of our age, and within this reality, we must work to define our spaces.

Print version (Adobe Reader required)
This study examines the utilization and preference of popular Internet and communication technologies among students at Sam Houston State University (SHSU), a Carnegie Research Doctoral University in East Texas. The researchers wished to study the local relevance of various technology trends reported in librarianship literature and then to use the survey data to inform decisions regarding library service development. A survey was conducted to investigate student ownership of electronic devices and student usage of technologies such as text messaging, Twitter, RSS, podcasts, social networks, Second Life, and others. Survey results indicated that, while students do not wish to experience an overwhelming library presence on all social networking and Internet media, most do wish to have basic library services easily accessible through a few of the most popular social networking and Internet technologies. The investigators identified some unique trends in usage among their local population and have adjusted certain library services and plans in accordance with their findings. Other libraries are encouraged to study their own users and develop new services based on those users’ needs rather than popular trends or surveys which may be based on radically different user groups.

As social networking and Internet technologies make significant strides in innovation and development, technology mediums for individuals to communicate with one another have increased exponentially. Although librarians have been early adopters of many information technologies, attempting to develop and maintain a presence on all available social networking and Internet communication mediums is a costly and inefficient service model for most libraries. Thus, selecting the most effective communication technologies for delivering library services has become a major challenge for many librarians. Furthermore, rather than blindly adapting the most visible emerging technologies touted by popular media, librarians must tailor library service delivery options to the distinct needs and preferences of their particular user population.

The purpose of this study is to survey student library users’ utilization and preference of popular Internet and communication technologies at Sam Houston State University (SHSU), a Carnegie Research Doctoral University in East Texas. Given that the university campus is located about an hour north of the Houston metropolitan area with a large population of commuter students, providing a virtual library presence beyond the physical campus is vital for improving the user experience. Moreover, since many of the students are also first-generation college attendees or students whose age or life circumstances is considered “nontraditional” among college students, it is also important for the library to examine the needs and technical abilities of these distinct user groups. A survey was developed to study students’ ownership, usage, and perception of popular social networking and Internet technologies, and whether students would like to utilize library services offered through these technologies.

Literature Review

Today’s typical college students have grown up with and been exposed to all manner of technologies in many aspects of their lives. On a daily basis they use computers, online social networks, cell phones, text messages, Twitter, RSS feeds, wikis, blogs, online learning tools, and much more. How students integrate these instruments into their lives has been the focus of much attention and research. Today’s students, being familiar with second-generation technologies popularly known as Web 2.0, can find and use information; produce content in various formats such as posts, blogs, or videos; and push content to recipients by various media such as phones or computers.¹ However, specific skills, attitudes, and practices can vary among groups.

Librarians at Kent State University studying Web 2.0 technology usage in undergraduates found heavy use of Web 2.0 applications, which encourage connection, interaction, and sharing. Results revealed that students made a clear distinction between their social and educational locales online.² The boundaries between learning technologies and social technologies were more subtle among UK university students in varied disciplines who were organized and efficient at finding and handling information, able to multitask, comfortable with combining tools, adept at studying and learning in various environments at various times, and willing to share resources with peers.³

Interaction with peers is a significant factor in students’ lives, and much research has been conducted concerning students’ social behaviors. Past research concerning the use of instant messaging (IM) tools was compared to results of a study published in 2005. Similarities surfaced, including the implication that face-to-face communication was a preferred and useful form of communication, while cell phones and IM were overtaking other forms of interaction such as land-line phones and e-mail.⁴ E-mail also lagged behind in preference for IM in survey results released in 2007 from 545 college students who favored IM technology for personal and social communication.⁵

In the academic realm, libraries have utilized IM technology to answer reference questions with varying results at different institutions. Analysis of one year’s worth of IM session transcripts at Binghamton University Libraries revealed that 5 percent of users were from the campus community and 31 percent of all users were students, while a 2008 pilot project at California State University, Fullerton, increased synchronous virtual reference statistics by 49 percent.⁶ Additionally, student focus group studies at Milner Library (Illinois State University, Normal) led librarians to drop their chat software program and migrate to an IM service after discovering that their chat service was underutilized.⁷

As a related topic, student communication and cell phone use have also been the focus of several studies. In a 2005 survey of 383 college-age young adults in four states, the majority of respondents used their cell phones to socialize, remain available, tell time, leave themselves reminders, and use contact list functions.⁸ An additional popular function of cellular phones is text messaging (or “texting”). A focus group at the University of Huddersfield showed that students were generally open to the idea of receiving text messages about library news and functions, such as a reminder for an overdue book. However, they emphasized the importance of only receiving messages that were useful in nature.⁹

In 2007, Rich Ling and Naomi S. Baron compared texting and IM usage among twenty-two female college students. They looked at 191 text transmissions with 1,473 words and 191 IM transmissions with 1,146 words. They found, among other things, that respondents would text multiple sentence transmissions at least 60 percent of the time, while they would only IM multiple sentence transmissions 34 percent of the time.¹⁰ IM was the most commonly used communication technology in a survey of 268 Canadian university students who were asked how they used IM, mobile phones (talking and texting), and e-mail.¹¹ In 2004, Kevin Lee and Stephen Perry surveyed 409 college students at a small college about their use of IM. They found that students dedicated a substantial amount of time to communicating regularly through IM, often even more than through face-toface contact, and friends were the most important communication partners in their everyday lives.¹²

Many students also build social relationships online by using social networking sites such as MySpace and Facebook. Student usage of Facebook was the subject of a study in which ninety-two undergraduates kept a diary-like log each day for a week. Results published in 2009 showed that students spent approximately 30 minutes on Facebook throughout the day, mostly reading and observing content.¹³ In 2008, a survey of sixty-eight undergraduates found that 37 percent had one social networking account, 53 percent had two accounts, and 9 percent had three accounts. Students visited their accounts on average three times per day.¹⁴ Highlights of the 2010 EDUCAUSE Center for Applied Research (ECAR) Study of Undergraduate Students and Information Technology, which surveyed over 36,950 college freshmen and seniors, indicated 96.6 percent of survey respondents used Facebook and 22.5 percent used MySpace.¹⁵

Students are not the only ones with online social accounts these days. At the time this paper was written, limited literature was available on student usage of the popular Twitter technology, although the 2010 ECAR study did report that 43.3 percent of students followed or used microblogs such as Twitter, but only 4.3 percent used Twitter in college courses. Similarly, information on student use of RSS feeds was scant.¹⁶ However, what literature is available suggests that some colleges and universities use these technologies to communicate with students. Some institutions use Twitter to dispatch news and information to students, while others use RSS feeds to direct information into course management systems.

Additionally, some libraries are creating Face-book profiles. In 2007, librarians Sarah Miller and Lauren Jensen offered tips for increasing student response to library services and information via Facebook, such as understanding that Facebook is all about connecting with the students. The librarians suggest, among other things, keeping the profile active, updated, and current; using the newsfeed feature; replying promptly to student questions; sharing favorite books, interests, and quotes; and promoting new databases and library services.¹⁷

In a related vein, academic libraries have recently started to explore the possibilities that e-readers hold for their patrons. E-readers are a popular topic among media outlets. However, only a few studies have been published in scholarly journals about e-reader usage in academic libraries. One notable study was conducted at the Texas A&M University Libraries. They purchased forty Kindle e-book readers in 2008, at a cost of $399 each, and conducted a year-long study of usage. They concluded that the e-reader was an effective device for popular reading but had limited value for academic reading because of poor graphics, high cost, and limited content.

Students rarely selected academic titles for the library to purchase on the Kindle.¹⁸ E-readers were also not common among college students in the 2010 ECAR Study, with only 3.1 percent reporting that they owned a dedicated e-reader.¹⁹

Podcasts, on the other hand, are somewhat more promising academic tools in some cases. Researchers at the University of West England found that students did not want to relinquish being taught by traditional methods but believed that improvement in learning resulted when pod-casts were used as an additional academic tool.²⁰ Additionally, the use of podcast lectures reportedly improved study and grades among undergraduate nursing students at the Washington State University College of Nursing.²¹

Many academic libraries produce podcasts, including the SHSU library. Reviewing the literature about podcasts shows that most institutions report positive results from their podcasts. They are an effective way to instruct, promote library services, and involve the student body with the library. Curtin University Library in Australia developed its own podcast series at the end of 2005. They used the series to promote library services and provide basic instruction in using library resources. In an article discussing the program, they wrote, “We have been amazed at the popularity of our podcasting series; download statistics have increased week by week. In October 2007 we had our thirty-three thousandth download since the release of our podcasting series in February 2006.”²² Moraine Valley Community College also had a successful podcast; Michael Stephens wrote in a 2007 issue of Library Technology Reports that “the Library Event PodCasts are intended to be a flexible, portable record of the events held within the Moraine Valley Library.”²³

Method

The population for this survey was the undergraduate and graduate student body of SHSU early in the Spring 2010 semester. The survey included both on-campus and distance-learning students, since the library strives to serve the diverse needs of both groups. Approximately 6,240 students, or 37 percent of the total student body, were selected using stratified random sampling to participate.

The survey questionnaire consisted of fifty-four questions, although not all questions required responses. Some questions were designed to appear only when participants selected a particular response in a prior question, so participants may not have encountered all fifty-four questions. Furthermore, not all questions required participants to enter a response; therefore, unless otherwise specified, the percentages quoted in this paper indicate a percentage of an individual question’s respondents, rather than a percentage of the total survey respondents.

The first section of the survey asked about participants’ access to the Internet, type of Internet access, and computer and netbook ownership to assess levels of technological adaptation. The next section investigated participants’ familiarity with and usage of a number of popular technologies, including e-readers, Twitter, RSS, podcasts, social networking sites, and mobile phones. Participants familiar with a popular technology were also asked to indicate whether they would be interested in library services offered through that specific technology platform. Demographic questions, including year of birth, gender, and student classification, were placed in the last section of the survey to avoid undue influence on participants’ confidence and mindset. Access the complete survey instrument (PDF).

An initial invitation e-mail was sent to all subjects’ official SHSU e-mail accounts on April 2, 2010. The invitation stated the purpose of the study and provided a Web link to the survey hosted through commercial online survey tool Survey Monkey. As an incentive for participation, the e-mail also mentioned a chance to enter a drawing for gift certificates upon survey completion. The survey was open for the duration of April, and a reminder e-mail was sent to all subjects one week prior to the survey closing date of May 7, 2010. Overall, 702 of the subjects responded and completed the survey for a response rate of 11.25%. A random number generator was used to select three winners out of all participants who chose to provide an e-mail address for the gift drawing. Winners were contacted by e-mail during the week following survey closing.

Demographics

Undergraduate students represented just over 50 percent of the survey respondents, with 13.6 percent classified as freshmen, 14.4 percent as sophomores, 9.8 percent as juniors, and 12.8 percent as seniors. Master’s students were the largest group represented in the survey results at 41.4 percent, while doctoral students comprised only 6.1 percent of respondents. Seven respondents (1 percent) were nondegree seeking or continuing education students, and six respondents (0.9 percent) selected “Other” as their classification with no explanation provided (see figure 1).

Of the 666 respondents who chose to reveal their gender, 71.6 percent were female and 28.4 percent were male. Of the 674 respondents who chose to reveal their age, 375 respondents were 25 years of age or younger (55.6 percent); 267 respondents were between the ages of 26 and 50 (39.6 percent); and 32 respondents were over 50 years old (4.8 percent).

In addition to student classification, gender, and age, survey respondents were asked to report in what college they were seeking a major or degree. At the time of this survey, Sam Houston State University was divided into five colleges. The largest group of respondents, 31.5 percent, selected the College of Education. The College of Arts & Sciences had 20.8 percent respondents, and the College of Humanities and Social Sciences had 17.1 percent. Another 16.2 percent of respondents were from the College of Criminal Justice, and 12.9 percent were from the College of Business Administration. Eleven respondents (1.6 percent) reported an undecided major.

Finally, students were asked to report the manner in which they attend classes (strictly face-toface at the SHSU main campus or the SHSU annex campus, The University Center in The Woodlands, Spring, Texas; online; or blended face-to-face classes with online components). Students were allowed to check all that applied, so the percentages will total more than 100 percent. The largest response, 55.3 percent, was for face-to-face (F2F) classes on the main campus blended with an online component. Main campus F2F classes without an online component received the second highest response rate, 42.7 percent. Another 13.8 percent of respondents take F2F classes at the annex campus blended with an online component, while 8.6 percent take F2F classes at the annex campus without an online component. Strictly online classes were taken by 30.6 percent of the respondents.

Results

At the time of the survey, about 97 percent of student respondents had Internet access at home, with almost 76 percent of those having high-speed DSL/cable Internet. However, almost 2 percent of students had only dial-up Internet access, and 3.4 percent of students did not have any access to the Internet at home. Just under 56 percent of students used a desktop computer at home, and 90 percent of those desktops were PCs. Almost 92 percent of students used a laptop computer (though the survey did not distinguish between those who personally owned a laptop and those who might borrow one from a family member, roommate, etc.), and 86 percent of those laptops were PCs. Less than 8 percent of students owned a netbook. A little over 97 percent of student laptops had wireless Internet access, but just over 2 percent did not. Student computers, both desktops and laptops, averaged between 1 and 4 years old, with laptops slightly more likely to be 1–2 years old and desktops slightly more likely to be 2–4 years old.

Students were asked to indicate which in a list of popular or emerging technologies they recognized by name. Social networks like Facebook or MySpace had high name recognition (over 98 percent), while newer location-based social networking services like Foursquare, Gowalla, and Loopt had low name recognition (less than 12 percent). Figure 2 shows the full list of responses.

Students were then asked a series of questions about several technologies of particular interest to the library. The following sections investigate the specific responses pertaining to each technology.

Print version (Adobe Reader required)
Effects of social marketing strategies on student research behaviors were investigated. Three objectives were identified as target behaviors for change: (1) decrease procrastination due to the illusion of immediacy (2) increase students’ willingness to seek expert assistance when it is warranted, and (3) increase the selection of information sources based on criteria other than the information need itself, which includes the habituated and automatic use of Internet sources based on the assumption that they are more convenient, reliable, and easy to use. Findings suggest a positive impact as a result of marketing strategies attempting to achieve these objectives. Students who received messages based on a social marketing framework that emphasized these objectives appeared more willing to engage in discussions about the research process and were more likely to seek assistance from a librarian. A number of students reported successful encounters with librarians in meeting their research needs. Students who only received skills instruction reported attempting to use research tools like databases, but gave up in frustration. Due to relatively little research on how social marketing strategies can be used to change student research behaviors, more research is warranted to explore this connection. More investigation is also needed regarding how to help librarians learn how to package and deliver messages using a social marketing framework.

Cognitive ability is but one determinant of human behavior. If people’s actions were determined by knowledge alone, changing behavior would be a relatively easy task. Good behaviors, such as maintaining a healthy weight through diet and exercise, would be ensured through the simple delivery of appropriate information. Likewise, good information literacy instruction would ensure effective information seeking behaviors. Research suggests, however, that just as obesity persists in the United States despite the efforts of public health educators, college students continue to engage in less than ideal information behaviors despite the substantial efforts of instructional librarians across the nation’s colleges and universities.¹

Since 1951, when Wiebe asked, “Can brotherhood be sold like soap?” professionals in nonprofit arenas have seriously considered how they might apply commercial marketing principles to effect behavioral change in their clients.² What has been since dubbed “social marketing” has met with success in addressing and improving numerous health and safety behaviors.³ While there is a substantial body of literature on marketing information literacy services and programs, little research has been done heretofore on applying these concepts to the actual information behaviors of students. This study will compare the impact of two distinct approaches to library instruction aimed at improving the research behaviors of freshmen college students: one which is primarily cognitive in nature and aims to increase information seeking skills and the other which applies social marketing techniques to increase effective information seeking behaviors. As the first of its kind, this research is exploratory and expected to generate questions and hypotheses for further research in this area.

Literature Review

In order to change information seeking behavior of college students, one must first understand their current behaviors and motivations. Fortunately, several recent research studies provide an excellent foundation for this work. A review of this literature will be followed by a brief review of current instructional practice in academic libraries. Finally, a discussion of social marketing and a review of its literature will highlight how social marketing techniques are applied to noncommercial aims and how such an approach differs from cognitively oriented library instruction.

The Information Seeking Behaviors of College Students

Recent studies provide a surprisingly consistent picture of how college students identify, select, evaluate, and use information. Clearly, from findings across all studies, the majority of students do not practice the multi-source, diversified, high-information use strategies library and information professionals prescribe.⁴ Rather, they apply a predictable and highly consistent strategy for finding information. Whether they seek information for their course work or to fulfill every-day life needs, students turned to a small set of common information resources, demonstrating “little inclination to vary the frequency or order of their use regardless of their information goals and despite the plethora of other online and in-person resources that were available to them.”⁵ De Rosa et al. also conclude that the “majority of college students are not making high use of the array of electronic resources libraries make available.”⁶ Although Dervin and Huber found greater impact of context on source selection, their study describes the same limited information repertoire, particularly for undergraduates.⁷

Information seeking research has generally found that most people prefer informal, personal forms of information to formal information and that holds true for college students.⁸ However, Dervin and Huber found that a preference for electronic information is fast approaching equal status among undergraduates.⁹ Specifically, students prefer Internet search engines to other sources of information. While De Rosa et al indicate 89 percent of respondents reported beginning their search for information with an Internet search engine, Head and Eisenberg found students turn to Google (99 percent), Wikipedia (92 percent) and friends (85 percent) for everyday-life information and to course readings (97 percent) and Google (95 percent) for course-related research.¹⁰ Other studies also confirm that Internet search engines are the go-to source for college students when they begin their research.¹¹ Furthermore, students rely primarily on the Internet to cross reference sources and verify the information they have gathered.¹²

Current research also provides critical insight into what motivates college students’ seemingly intractable unacademic behavior.¹³ Students value and select information for very different reasons than do librarians. What Head and Eisenberg call a “disconnect” is key to understanding student behavior and librarians’ frustrations.¹⁴ While librarians select information based on thoroughness, credibility, and authority, students value other attributes more: “brevity, consensus, and currency,” according to Head and Eisenberg, and “familiarity, convenience and currency,” according to Dervin and Huber.¹⁵ Students have apparently absorbed the message that library resources are more trustworthy. When asked what types of sources are more reliable and accurate, 77 percent and 76 percent, respectively, of all student respondents identified library resources. But they rate Internet search engines more reliable (63 percent), cost-effective (71 percent), easy to use (87 percent), convenient (84 percent), and fast (90 percent).¹⁶ When students rate how well information sources “fit” with their needs and lifestyles, 64 percent reported that Internet search engines are a perfect fit compared to 24 percent who felt that way about the library.¹⁷ If students rate the Internet more highly on the criteria they value most, then it is perfectly rational for them to select it as their primary source of information.

The good news is that students do report knowing about scholarly research databases. De Rosa et al, Dervin and Huber, and Eisenberg and Head all found that a significant number of students have used a scholarly research database at least once in the past.¹⁸ Yet students selected databases for reasons that are less than ideal, for example because they have a single search box or because they are already known.¹⁹

There are several consequences to students’ nearly exclusive reliance on electronic sources of information. First, students do report experiencing frustrations due to “information overload and the sense of being inundated by all the resources available to them.”²⁰ This response to information overload, and other responses, has the possibility to be maladaptive.

One of them, escaping, is, by definition, dysfunctional.²¹ Head and Eisenberg describe students as being driven by familiarity and habit, sticking with known resources. In doing so, college students are able to escape information environments that are unfamiliar to them; environments that present the possibility of failure.²²

A second consequence is what Head and Eisenberg call the illusion of immediacy: that is, the illusion that appropriate information is always immediately available.²³ This misconception leads students to misjudge how much time they truly needed to complete course-related research assignments. So while fear of failure is still a significant reason for procrastination on research projects, many students (40 percent) now cite “juggling their needs to meet competing course demands from other classes” as a reason for delaying work.²⁴

Finally, most students do not seek help with these issues in libraries or from librarians. Eight out of ten reported rarely, if ever, asking a librarian for help with course-related research. Most used very few of the resources or services available to them.²⁵ Only 12 percent had ever used online reference services or attended on-site library training sessions.²⁶ A full 39 percent of students indicate they use the library less now due to Internet use, and most (62 percent) indicated they expect their library use to remain flat or decline in the future.²⁷

Research findings give some indication of why students tend to ignore these potential sources of help. Primarily, 70 percent of all college students still associate libraries first and foremost with books, while only 7 percent with information and 5 percent with research and 2 percent with reference services; though when asked to describe the main purpose of libraries, 49 percent used the word information in their answer.²⁸ Secondly, their experiences apparently reinforce their avoidance behaviors. When they did seek help, 52 percent of college students described the assistance they received from a search engine as equally helpful as assistance from a librarian. Only 32 percent thought it was better.²⁹ When asked if they had ever started with a search engine and ended up at and using the library’s website, a full 76 percent of those students that answered yes reported that the library’s site did not completely fill their information needs (66 percent partially and 10 percent not at all).³⁰

Head and Eisenberg, in summary of their own findings, express well the general consensus of the literature:

Students conceptualized research, especially tasks associated with seeking information, as a competency to be learned by rote, rather than as an opportunity to learn, develop, or expand upon an information-gathering strategy which leverages the wide range of resources available to them in a digital age.³¹

Information Literacy and Instruction

Generally speaking, the literature of information literacy emphasizes cognitive skill development. The most widely accepted articulation of information literacy, the learning outcomes described in professional documents from the Association of College and Research Libraries, heavily emphasize skill acquisition.³² Notable exceptions, such as work by Kuhlthau and Bruce, attend to the affective dimensions of information behavior and provide a foundation for this work.³³ Kuhlthau’s zones of intervention are particularly useful in helping understand how library and information professionals might better communicate with students during the various stages of their research.³⁴ Evidence that library instruction, as it is currently practiced, is having a substantial, macro-level impact on use behaviors is not overwhelming. Existing studies on instruction are rarely research based and, when they are, tend to rely on case-study methodologies.³⁵ That is not to say that traditional, skills-based instruction should be abandoned, but rather that much research is yet to be done. The findings on college students’ current information seeking behaviors discussed above suggest that library and information science (LIS) professionals need to continue researching new strategies to address these issues.

Social Marketing

Social marketing is essentially the application of commercial marketing techniques to the resolution of social and health problems. Wiebe is generally credited for the development of social marketing after he evaluated four different social change campaigns, and concluded that the more similarities they had with commercial marketing, the more successful they were. Social marketing is unique in that it is solution focused and seeks to change behavior in specific and measurable ways. Essentially, it provides a mechanism for systematically understanding barriers to individual behavior change and designing interventions accordingly.

The literature on social marketing is extensive. It has been applied primarily in public health and safety contexts to promote behaviors such as safe sex practices, family planning, smoking cessation, vaccinating children, drug and alcohol avoidance, and healthy diet and exercise. This literature is vast, and comprehensive reviews exist.³⁶ Overall these reviews conclude that social marketing interventions are effective:

The main point is this: research illustrated that social marketing interventions of different types can be effective and that they can work with different target groups: young people, adults, minority ethnic and disadvantaged groups. Furthermore the results demonstrated that social marketing interventions can be effective in a range of different settings: schools, the workplace, church based, community and family-based settings, clinical practices, supermarkets and media-based settings amongst others. The reviews also produced evidence that both narrow and broad focus social marketing interventions can be effective. Overall the effectiveness reviews reported in this article identify social marketing as a very promising health behavior intervention approach.³⁷

Social marketing methods have also been applied effectively in other areas, including education, so it seems likely to have a positive impact when applied to information behavior.³⁸

The literature on social marketing within the LIS field is limited. Librarians have applied social marketing techniques to increase librarian-teacher collaboration, promote the use of archives, promote youth services and collections, and to reduce users’ mistreatment of books.³⁹ The application of social marketing techniques is equally limited in the area of library instruction and user education. Much of it is confined to attracting students to user education programming.⁴⁰ Others fall into the “how-to” category but do not actually test methods for engaging in effective social media.⁴¹ This study will pose important questions and add valuable knowledge about the application of social marketing techniques to information behavior to this still developing body of literature.

Procedures

This quasi-experimental study compares two distinct interventions with college students to improve their information seeking behaviors.

Students were administered a preliminary survey to assess their current behaviors, two types of interventions were designed and delivered, and data were collected from student research journals and an instructional artifact, a works cited page, to compare the impact of each intervention. Students were assigned a number to maintain confidentiality. Multiple data sources ensured triangulation of data.

The process of constructing a social marketing campaign will be used as a framework to discuss the additional procedural details. That process has four major components: customer and competitor assessment, market segmentation, identification of target behaviors, and the design and delivery of an intervention intended to alter those behaviors.

Social Marketing Component 1: Customer and Competitor Assessment

The first step in a social marketing campaign is to thoroughly assess both customers and competitors. In noncommercial applications, this often refers to understanding what behaviors clients engage in that compete with more desirable behaviors. Existing literature, reviewed above, provides a thorough understanding of how college students, in general, seek information and why they are attracted to the Internet to fulfill so many of their needs. For this study, customer participants were drawn from three sections of the first of a two-sequence freshman composition course, all taught by the same instructor. Each of the three sections was randomly assigned to receive a traditional instructional treatment (hereafter called the Skills group), a social marketing treatment (hereafter called the Behavior group, or no treatment (hereafter called the Control group). A preliminary survey (see appendix A) was administered to assess the characteristics and entry behaviors of the particular student participants involved in this study prior to the treatment provision. It consisted of three demographic items and six open-ended questions intended to assess students’ past research behaviors. Open-ended questions were used to avoid influencing students’ descriptions of their previous activities. Surveys were coded by assigning numbers to specific answers that occurred frequently. For example, survey question five asks, “If you needed help with your research, would you ask for help? Where would you go to get it?” Answers varied, but most students used a combination of instructor, friend or family, a librarian, or the Internet. Possible answers were coded quantitatively and calculated into percentages. This method was used for each of the survey questions. Data were calculated for each of the separate classes to determine if there were any significant differences in student research processes prior to intervention. No significant differences were identified.

Print version (Adobe Reader required)
Are the resources provided in answers to reference questions reusable for answering future reference questions? This study seeks to answer this question as a means to address the scalability problem of human-mediated reference work. Using the Internet Public Library’s archive of over eighty thousand records of answered reference questions, this study identifies (1) what resources are provided in responses to digital reference questions, (2) the extent to which these resources are reusable in future responses, and (3) the useful lifespan of a resource that has been provided. The distribution of resources provided in these answer records was found to display a classic power law distribution. The half-life of these resources was found to be approximately eleven years, far longer than the half-lives of resources in other corpora that have been studied. The relevance of these resources was found to be remarkably high, even after more than a decade.

Ever since the advent of digital reference services, there has existed a belief that, as Coffman puts it, “if we could somehow access the work another librarian had done before, there would be no need to start over answering every question from scratch.”¹ This is a seductive notion, as it offers at least a partial solution to a perennial problem of digital reference work and perhaps all reference work: scalability.² Human labor is time consuming and expensive, and significant savings of both could be realized if even some of the products of that labor could be reused.

On the other hand, some researchers suggest that the context that gives rise to an information need is unique for every individual.³ A consequence of this position is that an answer that is useful to an individual in a particular context and the set of information resources provided to support that answer will not be useful to others in other contexts. If this is indeed the case, then even similarly phrased questions cannot be treated as actually being similar. It may therefore be misguided for reference services to attempt to reuse answers across ostensibly similar questions.

This is a critical issue, both practically and theoretically. As a practical matter for reference services, if answers are indeed reusable across questions, then Coffman is correct, and there is no need for reference librarians to answer every question from scratch. On the other hand, if answers are not reusable across questions, then this dramatically limits the scalability of reference services. As a theoretical matter, if answers are reusable across questions, then even though the individual contexts out of which information needs arise may be unique, commonalities exist in how those information needs may be fulfilled. Despite this being an issue central to reference work, however, there has been no research to date on the reusability of answers provided by reference services.

The resources used in answers to reference questions are a different matter. Resources are of course reused all the time. The reference section of a library is full of materials that are used again and again. It’s well known that dictionaries, encyclopedias, and telephone directories are among the most commonly used reference sources in any library—and, at the risk of being tautological, a source does not become commonly used unless it is reused. Indeed, a library’s entire collection is reused, though of course some materials more than others.⁴ The argument could be made that reuse of materials is one of the raisons d’etre of libraries. Ranganathan said it best: Books are for use.

This paper reports on a study to investigate the reusability of the information resources in answers provided by digital reference services. Specifically, the information resources investigated in this study are the URLs provided in answers from a digital reference service. The research questions guiding this project were

1. To what extent are the URLs provided in responses reusable for future responses?
2. What is the useful lifespan of URLs provided in responses to digital reference questions?

In order to answer these questions, it is necessary to have a corpus of answered questions to work with. This is, of course, why no research has been conducted on the reusability of reference answers: these corpora either do not exist or are unavailable to researchers. Questions and answers are entirely transitory at reference desks; there is a long tradition of capturing data about the interaction at reference desks, but the interaction itself is rarely, if ever, captured.⁵ The entire interaction is captured by digital reference services—e-mail exchanges, instant messaging transcripts, etc.—but due to significant user privacy concerns, those artifacts are rarely made available outside of the service itself.⁶

Background on the Internet Public Library

There are, however, two large corpora of answered questions available outside of the services that answered the questions: the QuestionPoint Global Knowledge Base, and the Internet Public Library’s Archive of Reference Questions (ARQ). QuestionPoint is the most widely-used digital reference management system (www.oclc.org/questionpoint), and the Global Knowledge Base is the repository of answered questions submitted by libraries around the world that use QuestionPoint. As of February 2009, the Global Knowledge Base contained 20,061 searchable records.⁷ The Global Knowledge Base is searchable by library users, though it is not clear how many libraries have implemented this feature.⁸ The Global Knowledge Base is not, however, readily accessible for research, while the IPL’s ARQ is, at least to the current author.

As of 2006, the IPL is partly supported by the memberships of information and library science programs in the United States and around the world. For their membership fees, these programs receive access to the ARQ. The ARQ contains a record of every question answered by the IPL, going back to the IPL’s inception in 1995. This study found that as of June 2010, the ARQ contains 81,385 records, thus making it several times larger than the QuestionPoint Global Knowledge Base, even assuming growth of the Global Knowledge Base since February 2009. The author’s institution, the School of Information and Library Science at the University of North Carolina at Chapel Hill is also an IPL member, thus making the ARQ readily accessible to the author.

The Internet Public Library (ipl.org) describes itself as “a public service organization and a learning/teaching environment.”⁹ The IPL provides two primary services: collections of vetted and annotated resources, and an Ask a Librarian service. The IPL maintains resource collections on a wide range of topics, several special collections (e.g., on topics such as the U.S. Presidents and the fifty states), and collections for young children and teenagers. These are not collections of content developed by the IPL, but rather collections of links to materials developed by others, on the free web, that have been vetted by the IPL and judged to be authoritative and trustworthy. For example, the IPL’s Earth Sciences category contains resources from NASA’s National Space Science Data Center, the U.S. Geological Survey, and the British Geological Survey, among many others.

The Ask an IPL Librarian service maintains a question submission webform (ipl.org/div/askus) that allows users to submit a question on any topic. Submitted questions enter a queue, and answerers may claim the unanswered question of their choice. IPL policy states that once an answerer claims a question, she should submit a response to the user within twenty-four hours.¹⁰ IPL policy also states that answers should contain two to four sources.¹¹ Once an answerer provides a response to a question, the response is sent as an e-mail to the user, and the question-and-answer records are stored in the IPL’s content management system, QRC.¹² Every six months or so, new question-answer records are deidentified and made available in the ARQ. The IPL has a small full-time staff, and much of the work of vetting resources and question answering is conducted by volunteers and students in information and library science programs, in particular in courses on collection development and reference.

In January 2010, the IPL merged with the Librarians’ Internet Index (LII), and was renamed ipl2. The LII maintained extensive collections of vetted resources, similar to those maintained by the IPL. The LII did not, however, maintain a question answering service. While the data collected for this study spans the time period of the transition from the IPL to the ipl2, the Ask an IPL Librarian service was largely unaffected by this transition.

Literature Review

In the quote at the beginning of this paper, Coffman suggests that it would be useful to reuse the answers provided by librarians in response to reference questions.¹³ Indeed, he seems to suggest that the only hurdle to this reuse is the technical capability to mine these answer corpora.

This is not, however, a widely accepted position in the reference community. Indeed, judging by the number of digital reference services that do in fact reuse answers provided by librarians, this position has been rejected wholesale by the reference community. This rejection of resource mining may be a principled stance by the reference community, a belief that information needs are subjective and unique to individuals.¹⁴ If this is the case, then this principled stance is held despite the repeat nature of many questions submitted to reference services.¹⁵ The rejection of resource mining may, however, simply be a technical limitation: even the most sophisticated question answering systems are not yet capable of answering many questions submitted to a reference service, which may be on any subject, ambiguous, poorly phrased, and the thousand shocks that natural language is heir to.¹⁶

There is, in fact, only one digital reference service of which the author is aware that reuses answers in response to new questions: the Mad Scientist Network (MADSci, www.madsci.org). Bry describes the process employed by the MADSci Ask-A-Scientist service: when the user submits a question, a CGI script searches the MADSci archive of previously answered questions.¹⁷ Bry states that “approximately 63 percent of questions are matched with archived files”—however, “only 25 percent of users deem their questions answered by this process (15 percent of all submitted questions).”¹⁸ While the MadSci Ask-A-Scientist service is still operational, Bry’s article is a decade old now, and it is not clear how or if the algorithms used by the service have changed over that time. In the interim, other forms of question-answering systems and services have emerged.

Perhaps the most significant alternative to digital reference services to emerge in the past few years is social Q&A sites. Examples abound: Yahoo! Answers, Ask Meta-Filter, WikiAnswers (wiki. answers.com), the Wikipedia Reference Desk, the now-defunct Google Answers, Aardvark (acquired by Google in February 2010), to name only a few. Due to the different mechanisms and policies according to which these sites operate, it is difficult to define precisely what social Q&A is. The most basic characteristics that all of these sites possess in common are (1) a mechanism for users to submit questions in natural language, (2) a mechanism for users to respond to submitted questions, and (3) a community built around participation in this question answering.¹⁹ There is a great deal of variability in the types of questions asked and the quality of the answers provided on social Q&A sites; so much so in fact that some librarians maintain that social Q&A and library reference services are not even offering the same service.²⁰ Whether or not these two types of services are in competition will not be addressed here. The important point for present purposes is that not only the resources provided in answers, but the entire content of previous answers, are reused in answers to new questions on social Q&A sites: of Yahoo! Answers responses that refer to online sources, fully 59 percent of those online sources are previous Yahoo! Answers responses.²¹

A less recent but equally significant alternative form of question answering evolved from the information retrieval community. The Text Retrieval Conference (TREC) was instrumental in promoting the development of question-answering (QA) systems, by hosting a Question Answering Track for nine years, 1999–2007. The goal of the Question Answering Track was to develop systems “to retrieve small snippets of text that contain the actual answer to a question rather than the document lists traditionally returned by text retrieval systems.”²² These systems were also built to answer open-domain questions: in other words, questions could be on any subject, and the corpus of documents from which answers were retrieved was a newspaper collection, which could support answering on any subject. A persistent criticism of these systems, however, was that they were able to answer only “factoid” questions—such as, “Who is the conductor of the Boston Pops?”²³ In other words, these were systems designed to answer ready reference questions. One of the long-standing desiderata for question answering systems was expert-level answering of expert-level questions.²⁴ That was never achieved by any QA system developed for TREC, perhaps due to the difficulty of interpreting questions and formulating answers at an expert level in an open domain. At the other end of the spectrum, however, expert-level QA systems have been developed with considerable success in restricted domains, where it is feasible to develop domain-specific rules for identifying answers in texts.²⁵

Out of this work on developing restricted-domain QA systems has come some research on enabling QA systems to reuse information provided in previous answers.²⁶ One of the categories of reuse articulated in that study is the reuse of one document to answer more than one question. This type of reuse was found to be ubiquitous in the corpus of questions used by Light et al. This finding should come as a surprise to no one; the existence and popularity of frequently-asked question (FAQ) lists attests to the usefulness of a single document in the answer to multiple questions. Indeed, a FAQ list operates as a single document that answers multiple questions in two ways: first, multiple answers are provided in a single FAQ document, and second, a single answer may be a response to multiple information needs that can be reformulated as a particular question in a FAQ list.

The popularity and usefulness of FAQs provides evidence that one person can find the answer to someone else’s question useful. Almost since the advent of the web, researchers have worked on developing QA systems that make use of FAQ lists as corpora. One of the first of these was FAQ-Finder, which, even in the early days of its development, showed considerable success in retrieving QA pairs that satisfactorily answered new questions put to the system.²⁷ More recent work has demonstrated similar results, both in restricted and, like FAQ-Finder, in open domains.²⁸

While FAQ lists are often created in response to actual frequently-asked questions, as opposed to lists of questions being created to proactively respond to anticipated questions, they are also often cleaned-up versions of those questions. Most social Q&A sites, on the other hand, make the entire content of previous questions and answers available online. Thus, unlike in a FAQ list, where multiple real questions may be reformulated into a single FAQ, in social Q&A sites, every question, no matter how similar, is retained and made available as written. Building on methods derived from FAQ-based QA systems, considerable success has been reported in answering new questions by making use of the corpus of answered questions on social Q&A sites.²⁹ This work provides good reason to believe that the answers provided by digital reference questions may likewise be reusable.

Method

This project used data from the IPL’s ARQ. The ARQ contains all answered questions submitted to the Ask an IPL Librarian reference service from the inception of the IPL in September 1995. These question-answer records contain fielded data as an artifact of the IPL’s Ask an IPL Librarian question submission webform. Some of these fields are closed-ended, such as the drop-down list of subjects by which the user can categorize her question. Other fields are open ended, such as the field in which the user specifies her question. The records in the ARQ contain all of the data from the webform, the response provided by the IPL answerer, and system-level administrative data supplied by the QRC (e.g., timestamps for when the question was submitted, when an answerer claimed the question, and when the answer was sent).³⁰ The IPL has developed an algorithm to deidentify these records to a claimed level of 90–95 percent accuracy; this algorithm is run on all records prior to their being made available in the archive.³¹

The author wrote a simple web crawler and in June 2010 downloaded every record in the ARQ. The earliest records in the archive were answered in August 1995, the latest in June 2009. The most recent data was a year old at the time of this data collection because the IPL deidentifies question-answer records every six months, and then there is a lag before they are made available in the archive; the latest batch was made available shortly before this data collection.³²

The author also developed a parser to tokenize question-answer records. In other words, records were analyzed automatically, and certain blocks of text were identified within each record. The following blocks of text were identified in this way: the text blocks that are the question and the answer, URLs provided within answers, subject categories, and timestamps. The IPL has a policy dictating that for any URL over sixty-five characters in length, an additional shortened link be provided using TinyURL; all TinyURLs in answers were therefore eliminated as duplicates of other URLs.³³

URLs are relatively easy to identify using regular expressions. A regular expression is a search string that matches a specific pattern of text: as a simple example, “lib” matches the patterns “library,” “librarian,” “ad-lib,” “calibrate,” and of course many others. There are several regular expressions that one can find on the web to match URLs. This ease of identifying URLs is due to their inherent structure and syntax: all contain a top-level domain (e.g., .edu), most contain a protocol prefix (e.g., http://), many contain the character /in the middle, most end with a /or with a three- or four-letter suffix (e.g., .html, .pdf, .asp). Even many mistyped URLs may be identified using a regular expression that employs these syntactical rules. Other types of resources, such as books and journal articles, are also provided in IPL responses but far less frequently than URLs. This is the result of another IPL policy dictating that answerers should prioritize providing “freely-available sources” online to ensure “that all our patrons will have access to the information,” though a related policy states that print and subscription sources may be included in answers under appropriate circumstances.³⁴ Future work will be required to develop regular expressions to identify types of resources other than URLs in answers.

Print version (Adobe Reader required)
As virtual reference services (VRS) have become vital alternatives to traditional face-to-face (FtF) and phone reference, guidelines for best practices emerged to ensure that users are being well served across different modes of service delivery. Yet much remains to be understood about whether and how the essential process of query clarification occurs in the live chat reference environment. This study is among the first large-scale analyses of chat transcripts from libraries of different types, sizes, and world-wide geographic regions. It reports results from a qualitative analysis of a random sample of 850 transcripts from QuestionPoint, a 24/7/365 cooperative reference service administered by OCLC Online Computer Library Center with one thousand plus member libraries across the world. A Query Clarification Coding Scheme was developed as part of the analysis by using the constant comparative method. Close reading and evaluation of the transcripts revealed that librarians ask clarifying questions heavily before searching, while users generally offer clarification of their information need more often during the searching process. Both groups clarify to seek or offer information about the query topic and background; search history; and extent, depth, and type of resource needed. They also clarify to verify understanding and correct misunderstanding. Results indicate that accuracy was enhanced for librarians who used clarifying questions in answering ready reference (factual) questions. Although highly recommended query clarification techniques, especially using the follow-up question before logging off, are generally prescribed to improve accuracy, only 50 percent of librarians used follow-up questions and 33 percent of all questions asked to users were open questions. These findings suggest that emphasis on increasing these two techniques during VRS librarian training will improve the effectiveness of chat reference. Additionally, users with queries related to school, academic, or work-related information seeking were found to comprise 26 percent of the analyzed transactions, suggesting important implications for further research efforts as well as VRS librarian training and practice.¹

Though research-based guidelines for traditional, face-to-face (FtF) reference are firmly established, investigation into query clarification practice in virtual reference services (VRS), including live chat reference, is still in the early stages. As VRS have continued to grow as popular alternatives to FtF and phone reference, guidelines for best practices have emerged to ensure that users are being well served across different modes of service delivery.²

Because of the proliferation of VRS, it is increasingly important to have a deeper understanding of current practice as well as research-based recommendations for service excellence, as several scholars have pointed out.³ Lankes articulates important assumptions and central issues for a research framework that need to be addressed to advance the field’s understanding of VRS as emerging phenomena.⁴ He posits that human expertise is useful to incorporate into information systems, which Curry calls “the human touch.”⁵ In addition, Lankes believes that the digital nature of reference systems provides a significant differentiating context from FtF reference.⁶

Several authors have made a strong case for the need for query clarification and conducting reference interviews in VRS encounters, but little is known about current practice beyond sporadic data collected at the local level.⁷ What is the frequency and nature of query clarification in VRS? What questioning techniques are used? At what point in the chat session are clarifications requested by librarians or offered by users? What do they seek to clarify? This article addresses the above questions by reporting the results of a large-scale international research project that seeks to further the understanding of query clarification practices of both VRS librarians and users. Results are obtained through an in-depth analysis of a random sample of 850 live chat transcripts randomly selected from an international VRS provider, OCLC Online Computer Center’s QuestionPoint.⁸

Query Clarification in Face-to-Face (FtF) Reference

Research in query clarification techniques of traditional, FtF reference has been plentiful since Taylor’s seminal article, which proposes a theoretical model that conceptualizes reference as a process of question negotiation.⁹ The essence of this negotiation is one person’s attempt “to describe for another person not something he knows, but rather something he does not know.”¹⁰ Taylor’s major contribution has been to raise awareness of the complexity of the interaction and to stimulate research.¹¹ Since Taylor, the framing of reference as a process of query clarification has produced a body of research concerning the reference interaction, defined as “a conversation between a member of the library reference staff and library user for the purpose of clarifying the user’s needs and aiding the user in meeting those needs.”¹² Radford prefers the term “reference encounter” rather than “reference interview” to acknowledge that complex relational, interpersonal dimensions are present in addition to the information exchange.¹³ While Wilson notably challenges the unquestioning assumption that there is a need for clarification in every reference transaction, Ross, Nilsen, and Radford and Ross suggest that there is almost no situation in which a reference interview is not needed to clarify the user’s query.¹⁴

Research on reference interviews in the FtF environment has found that asking questions makes a librarian more likely to produce a correct answer and the user more likely to be satisfied, but it also indicates that librarians seek clarification less than 60 percent of the time. Gers and Seward find that librarians who do not negotiate the query almost never provide a correct answer, while those who do negotiate provide a correct answer 62 percent of the time; nevertheless, clarification occurs in only 49 percent of reference encounters.¹⁵ Arguing that willingness to return is a better measure of the reference interaction than accuracy of the answer, Durrance finds that students are more likely to return to librarians that they rate highly for interviewing skill, ability to determine need, and search strategy.¹⁶ Building upon Durrance’s work, researchers at the University of Western Ontario conducted the library visit study using MLS students as participant-observers in FtF reference encounters. They found that librarians asked clarifying questions in only 51 percent of encounters in phase 1 (1991–1993), and 48 percent in phase 2 (1998–2000).¹⁷

Best practices guidelines for reference service recommend a variety of query clarification techniques such as the strategic use of open questions to elicit information about the user’s situation, closed questions to verify the question, and follow-up questions to ensure that the user is satisfied.¹⁸ Dervin and Dewdney propose that librarians engage in “neutral questioning,” which involves asking open questions that encourage users to express the query in their own terms.¹⁹ Durrance finds that users are most likely to be willing to return to librarians who started the reference encounter with an open question; librarians are, however, most likely to begin with a statement (37 percent) or closed question (31 percent) rather than an open question (27 percent).²⁰ Studies of reference encounters suggest that the follow-up question is a particularly effective technique of query clarification because of its potential to correct misunderstandings and repair unsatisfactory interactions, which often leads to a re-negotiation or more complete negotiation of the query.²¹ Gers and Seward found that librarians who follow up the query negotiation by asking “Does this answer your question?” provide a correct answer 76 percent of the time, compared to 52 percent when this question is not asked.²² Despite its apparent effectiveness, they found that the follow-up was only asked in 12 percent of reference interactions. Interpreting the follow-up question more broadly, Dewdney and Ross found that variations of the follow-up question occur in only 37 percent of the reference interactions.²³ Further studies confirm similar occurrence rates of 37 percent and 35 percent.²⁴ In situations in which the reference librarian is truly unable to adequately help the user, the most beneficial follow-up behavior of referring the user to other resources has been recommended.²⁵

In addition to increasing the accuracy of librarians’ answers, clarification techniques also have an effect on the relational, interpersonal dimension of the reference interaction. Many elements of interpersonal communication have been shown to impact the librarian’s effectiveness during the reference encounter: showing interest in and comfort with the user’s question; approachability, friendliness, and confidence; and a nonjudgmental attitude.²⁶ While Katz frames the reference interaction from the librarian’s perspective in which the goal is to find a satisfactory answer, Radford applies interpersonal communication theory in an analysis of FtF reference in academic libraries to show that the content-oriented (information) dimensions of a reference encounter cannot be separated from its relational (interpersonal) dimensions.²⁷ Radford found that 74 percent of users reported that relational, interpersonal factors were more important than content-related factors in assessing FtF reference interactions, with three important categories of relational factors emerging from the analysis: attitude, relationship quality, and approachability.²⁸

According to Gross, most of the findings that have contributed to the creation of best practices guidelines for reference assume that the user is the originator of the query in question, that the query is self-generated.²⁹ Her research indicates that many users are seeking information related to imposed queries. Imposed queries are defined as those in which the person pursuing the question has been put on the course by someone else, usually a teacher, parent, work colleague or supervisor, family member, or friend.³⁰ Gross suggests that recommended techniques designed for self-generated queries are not translatable to imposed queries in which the query originator is absent from the reference process. Recommendations in light of the distinction between imposed and self-generated queries include developing a professional stance of contextual empathy, recognizing that right answers are probabilistic, and recognizing the absence of true feedback.³¹ Ultimately, Gross suggests that the first and most important task of query clarification is the identification of question type.³² In an important study that formally incorporated question type into its data collection, 25 percent of adult public library users indicated that they were seeking information for someone else.³³

Query Clarification in Virtual Reference Services (VRS)

VRS, in asynchronous (i.e., e-mail or SMS text messaging) and synchronous (i.e., live chat or instant messaging [IM]) forms, have burgeoned in number and frequency of use since their emergence in the mid-1990s. Librarians and researchers have since been grappling with understanding what effect the virtual environment has on the nature of reference services. Is there a need for query clarification in VRS? If almost every question benefits from a reference interview in the FtF environment, then it is likely that there is even more of a need in the virtual environment in which librarians operate without visual or auditory cues.³⁴ Ronan suggests that without the sensory cues of the FtF encounter, VRS librarians should rely much more heavily on classic questioning techniques.³⁵ Luo’s survey finds that librarians identify “reference interview skills” as one of three universal competencies for reference work, essential both to in-person and remote environments.³⁶

But does reference interviewing occur in the virtual reference environment? Nilsen and Ross report that “the reference interview has almost disappeared,” with only 20 percent of VRS encounters (eighty-five visits) including a reference interview, though the percentage increases to 40 percent if only chat reference and not e-mail reference is considered.³⁷ Analysis of 114 chat transcripts of librarians helping students with homework by Walter and Mediavilla find that only 32 percent of librarians probed the students for more information before referring them to the Live Homework Help service.³⁸ Marsteller and Mizzy report that “the reference interview seems to be alive and well in the brave new world of synchronous digital reference,” occurring in 64 percent of 865 analyzed transcripts.³⁹ Because of the localized nature of most VRS studies and differing definitions of “reference interview,” it is difficult to generalize these findings.

One of the earliest concerns raised about VRS was that it is not conducive to complex research questions that require extensive reference interviewing, and that users would not tolerate librarians’ efforts at query clarification. In a survey conducted by Janes, 80 percent of reference librarians thought ready reference questions would be well served by VRS, but only 33 percent thought detailed research questions would be well served;⁴⁰ 46 percent thought research questions would actually be poorly served in VRS environments. Librarians may assume that users are impatient and unwilling to engage in query clarification, which causes feelings of pressure-induced anxiety, leading to fewer clarification attempts.⁴¹ While librarians assume that chat users are in “such a rush that they don’t have patience for the traditional reference interview,” only 5 out of 270 chat reference transcripts examined in the Marstellar and Mizzy study contained a user displaying a negative response to clarifying questions on the part of the librarian.⁴² Kern suggests that, contrary to librarians’ perceptions, chat reference users “don’t use online chat because they are in a hurry, but because it is more convenient.”⁴³ Connaway and Radford have confirmed convenience as a factor in users’ decision to use VRS.⁴⁴

One of the ways in which the profession encourages best practices like query clarification in reference encounters is to create guidelines that can be used for implementation, training, and evaluation purposes. Both the RUSA Guidelines for Behavioral Performance of Reference and Information Services Professionals and the Digital Reference Guidelines promote query clarification techniques such as using open and neutral questions to elicit the query, closed or clarifying questions to refine the query, and follow-up questions to confirm that the user’s query has been adequately answered.⁴⁵ Kwon and Gregory test the effect of the RUSA Guidelines on user satisfaction with chat reference by analyzing 422 transcripts and corresponding user surveys from a public library system participating in a national chat consortium.⁴⁶ They find that asking users the follow-up inquiry of whether their question was answered completely to be a strong predictor of satisfaction. Kwon’s further analyses of the transcripts and user surveys reveal chat’s inherent advantage in dealing with queries requiring a level of subject expertise and its corresponding disadvantage in handling questions specific to local library’s hours, policies, resources, and services.⁴⁷ She identifies two types of referrals: redirectional local referrals, which are detrimental to collaborative VRS, and expert research referrals, which are conducive to collaborative VRS.⁴⁸ Kwon also finds that question type has an effect on outcome measures of answer accuracy and user satisfaction: local-specific questions, defined as circulation-related or inquiries about local library services, were answered less completely and users expressed lower levels of satisfaction compared to nonlocal questions.⁴⁹

One thousand four-hundred thirty-five transcripts from Texas A&M University Libraries’ chat service were reviewed for compliance to RUSA’s guidelines by van Duinkerken, Stephens, and MacDonald, who found poor compliance with the guidelines related to “listening/inquiring behaviors” typically associated with reference interviews.⁵⁰ Only 10 percent of librarians restated the question; 50 percent asked open-ended questions; 57 percent refined the original question with close-ended confirming questions; and only 31 percent asked follow-up questions to see if the need had been fully answered. The authors surmise that poor compliance is related to time, as 805 of the 1,435 users indicated that they were in a hurry to get the information they needed. In an unobtrusive evaluation of a sample of fifty academic and public libraries offering chat reference, Ronan, Reakes, and Ochoa found that 41 percent of VRS librarians adhered to the RUSA guidelines regarding use of open-ended questions and 70 percent for close-ended questions.⁵¹ Despite the existence of clear professional guidelines for chat VRS, levels of adherence to these guidelines regarding reference interviewing are generally low.⁵²

Luo derives an original list of chat reference competencies based on a survey completed by 597 chat librarians.⁵³ The resulting competencies are classified as general reference competencies, competencies highlighted in chat, competencies specific to chat, and competencies not as important in chat. The essential competencies include reference interview techniques in addition to customer service mentality, keeping users informed by constantly notifying them what the librarian is doing; the ability to work under pressure; and a knowledge of effective online communication skills. Radford’s use of interpersonal communication theory sheds light on the workings of these online communication skills, revealing that both librarians and users engage in relational facilitation through rapport building, shows of deference, compensation for lack of nonverbal cues, and greeting and closing rituals.⁵⁴ Clarifying questions can be used to both build rapport (e.g., “Is this what you are looking for?”) and to give deference to the user (e.g., “Do you agree that this is a good search term?”).⁵⁵ Of particular note in this approach is that transcripts are analyzed as communicative encounters, which considers librarians and users as equal participants in maintaining the interaction. Similarly, Westbrook uses politeness theory as a framework for a discourse analysis of 402 transcripts from one academic year at a large, public university, identifying syntactic and content-based markers of formality levels in chat reference encounters.⁵⁶

The present study builds upon the communicative approach to frame query clarification in chat reference as a complex process between librarians and users, with both content and relational dimensions.⁵⁷ It is among the first large-scale analyses of transcripts from an international consortium representing a wide variety of public, academic, special, and consortial VRS that seeks to understand if and how query clarification occurs in the live chat environment. It seeks to investigate the following research questions:

• What is the state of current practice of query clarification in chat reference?
• What kinds of query clarification techniques are being utilized in chat reference?
• At what stages of the chat reference interaction is query clarification occurring?
• What question types are being used in chat reference query clarification?
• What are the patterns of librarian versus user clarification?
• What is the frequency of self-initiated versus imposed queries in chat reference?
• What is the impact of query clarification on accuracy for ready reference questions?

Print version (Adobe Reader required)
Most librarians and staff who perform library instruction in academic settings place a high value on assessment. They understand that determining what our students bring with them to the instruction lab and what they learn during the hour we teach them helps improve teaching. Some study the latest in assessment techniques and educational theory and form high ideals for themselves and for their libraries’ instruction programs.

However, when faced with realities of our programs, successful though they may be, improving assessment that is performed across instructional programs can be a daunting task. Academic libraries’ efforts to institute or improve library instruction assessment can lead to anxiety as librarians are pulled between using old favorite assessment tools and making major overhauls.

The authors of this article have developed an approach to assessment wherein librarians study the components of learning and use these to enhance familiar tools they already use for library assessment during instruction. This allows librarians to greatly improve the quality of their tools while feeling a sense of ownership without losing particular features of importance.

This study discusses three popular assessment tools (pretest/posttest sets, posttests, and activities) that were enhanced using this technique and the evidence of learning that each gathers. The authors use data collected with each tool to discuss the strengths and weaknesses of each tool and suggests strategies for tool selection.

Innovative and enthusiastic instruction librarians work hard to improve teaching in their departments. They reach out to new campus faculty, create engaging online tools, and continually try new techniques in the classroom. They assess their students’ learning, and encourage their colleagues to follow suit. The question of how best to assess often arises at this point. With many library instructors, many majors, unconventional library research assignments, and often only seventy-five minutes per semester with students, challenges of choosing thorough, informed assessments quickly outweigh ideals. The popular “one-minute paper” starts to look enticing again.

The authors of this paper bring a range of perspectives on higher education as well as a wide variety of assessment techniques and theory. All are actively engaged with the education of college and university students. Based on their research at the Auraria Library in Denver, Colorado, they recommend continuous reexamination of a department’s assessment tools, based on the detailed breakdown of student learning (into factors of knowledge, skills, and attitudes) that is examined later in this paper. Further, their findings indicate that even a simple awareness of what one’s tools actually test encourages improvement of these tools. Briefly studying the components of learning can lead many instructors to make positive changes in content or evaluation methods, whether simple or large scale.

The process designed for this study began by collecting assessment tools being used in a library instruction program and identifying the factors of learning covered by each. Tools may be revised to cover additional factors of learning. They are then used in the classroom and revised again, in response to “real life” factors, such as skills necessary for a particular course or adjusting the overall level of difficulty. The techniques also work on a smaller scale, for better incorporating learning theories into an individual’s assessment efforts. This article includes sample rubrics and instruments for testing information literacy across subject boundaries.

The major strength of this analytical process is that it allows a particular library instruction program, or an individual instructor, to revise assessment tools that are already in use. Rather than discarding current instruments and starting from scratch, instructors study the components of learning and use these to improve the tools they are already comfortable with. This may entail updating tools’ content or changing how students’ answers are evaluated. Either way, both departments and individuals will find this method easier than creating completely new tools.

Literature Review

It is important to note that this review focuses on one-shot library instruction sessions. Librarians embedded in college classes, for example, have proposed more in-depth methods of evaluation appropriate to their situations. In general, these do not apply smoothly to one-shot scenarios.

Current Trends in Assessment of Learning

One major theme in recent assessment-related library literature is the culture of assessment on college and university campuses. Colleges and universities develop this culture and often campus-wide standards to go with them. Libraries can then adopt or adapt the standards and fit in with the culture. Some libraries also choose to create their own cultures of assessment before their institutions. This freedom to innovate is one of academic libraries’ most fortunate and productive strengths.

Many libraries have found that taking part in university-wide assessment programs and publicizing their assessment work to faculty members outside the library validates their work to these faculty and encourages more productive collaboration between libraries and academic departments. Susan E. Searing of the University of Illinois at Urbana-Champaign writes that “even small assessment efforts can make a meaningful difference in the acceptance of information literacy as a critical component of the curriculum. By sharing assessment results, librarians generate good will between the library and the academic program.”¹

Claudia Ruediger and Donald Jung highlight the fact that now “information literacy skills are common learning indicators found in accreditation and assessment documents.”² Regional and professional accreditation groups have incorporated information literacy into their standards as well.³

Even if a university has not begun an institution-wide assessment program, librarians can “demonstrate to a reluctant faculty and administration the utility of a new initiative.”⁴ Showing faculty that students learn to apply their information literacy skills and assessing how well students are learning with the resources the library has demonstrates dedication to students’ learning.⁵ Nancy O’Hanlon of the Ohio State University has applied an outcomes assessment model in situations where universities have not yet added information literacy to curriculum requirements. She advocates libraries’ creation of their own information literacy programs. Since direct measurement of what students learn through these programs is difficult, collecting faculty perceptions of students’ learning and development after these programs can help gain faculty support.⁶

Diller and Phelps have researched the use of portfolios, rubrics, and other more holistic methods of evaluation to help students understand expectations and measure their own growth. They provide an example of a matrix created to help students understand goals related to information literacy and other areas of learning throughout a general education curriculum.⁷ They also provide a highly detailed “Communication and Information Literacy Rubric,” which could be used on a single assignment or throughout a course.⁸ Knight provides another rubric meant to measure information literacy in assignments in a freshman composition class.⁹ Her rubric assesses student work using criteria from ACRL’s “Information Literacy Competency Standards for Higher Education.”¹⁰

The State of Library Instruction and Assessment

As assessment receives more and more attention throughout all units of colleges and universities, the body of literature on assessment of library instruction is growing rapidly. To make the review manageable, and also to reflect changing trends in library instruction and its assessment, this review is primarily limited to works published in 2000 and later. One exception is the inclusion of Evan Ira Farber’s work. From 1962 through 1994, he developed practices to guide librarians’ collaboration with faculty in other departments.¹¹ Although the exact methods librarians use to provide such philosophies over time should change, the philosophies themselves provide continuing guidance.

Libraries in the United States have used the ACRL standards to shape their information literacy programs in multiple ways. Numerous instruction departments have used them to design rubrics to assess information literacy in student assignments. Some are designed to be used by librarians and a few by academic faculty.¹² One of the most important aspects of the standards has been in the design of learning outcomes. Avril Cunningham notes that “instruction librarians now routinely make it a practice to write three to five learning outcomes for a class based on ACRL’s ‘Information Literacy Competency Standards.’”¹³

A small but significant body of literature in library science journals has covered a shift in the types of questions librarians ask students during assessment. In the past, assessment forms frequently asked students to rate librarians’ teaching style, level of familiarity with the material covered in class, and so on. Library researchers label these “affective surveys.”¹⁴ This change allows librarians to gather significant information on student learning, rather than simply finding out whether students found the hour vaguely satisfying.

The majority of recent (2000–2008) literature on assessment of library instruction sessions focuses on formats of assessment used and questions asked to more directly measure learning and comprehension. This recent literature on assessment of library instruction has also suggested a twist on effective surveying. Houlson notes that students’ responses to questions about the most useful thing learned and which resources they would recommend to others¹⁵ can prove quite revealing. While these responses do not directly reveal librarians’ proficiencies, they suggest comfort levels and may help project usage of specific research tools.

In many library instruction programs, instructors assess learning as a group but do not collect information on students as individuals. Houlson advocates getting students to work in groups, then share their results on whiteboards around the classroom.¹⁶ While individual results are not collected, students learn from each other, and librarians gauge the class’s overall understanding. Ondrusek and colleagues suggest “informal performance exercises”¹⁷ or ungraded online activities.

The debate over online versus paper formats has increased over the past several years. The current consensus seems to be that online tools have a variety of advantages: easy collection of data and holding student interest, but that most libraries still use paper tools. Currently, discussions of online tools focus more on content and course design than on technical details.¹⁸

Formats of Assessment Tools

A review of the literature on assessment of library instruction produces an enormous list of potential formats of the assessments themselves and of the questions they contain. Some formats are recommended many times, while others show experimentation on the librarians’ part. Below is a partial list of assessment formats described in literature since 2000. Types that are novel or particularly well described are cited here.

• One-minute paper: students are asked to write about a given topic for one minute. Topics often focus on students’ perceived learning or satisfaction.

• Pretest: a skills test administered before library instruction, generally within the first few minutes of class.¹⁹

• Posttest only: a skills test given after students have received library instruction, generally during the last few minutes of class. Some library instructors also choose to test students several days or weeks after library instruction.²⁰

• Pre and posttest sets: a skills tests given to the same group of students before and after library instruction. Tests may be administered so that the librarian can judge individuals’ before-and-after performance.²¹

• Free response questions on skills learned: students are asked to write a brief essay either displaying or discussing skills learned during library instruction.

• Viewing student research papers: librarians arrange to read papers written by students who have received their library instruction. Librarians may then evaluate the quality of research apparent in the papers.²²

• Viewing student portfolios: a similar technique, using student portfolios rather than papers.²³

• Pretest, posttest, and post-posttest sets: students complete quizzes before library instruction, immediately after receiving library instruction, and several days, weeks, or months after receiving library instruction. Quizzes may or may not be matched to measure individual students’ learning.²⁴

Most librarians do not suggest using significantly different assessment tools and questions for subject-specific classes and basic courses such as freshman composition.

Dimensions of Learning: Knowledge, Skills, and Attitudes

This study’s model of student learning, based on Bloom’s Taxonomy of Learning Domains, discusses three dimensions of learning: knowledge, skills, and attitudes.²⁵ Current literatures often discuss these dimensions in the context of learning in particular disciplines. The following materials discuss them in more general terms.

Peter Ackerman discusses the idea of the “knowledge repertoire,” a collection of information that students can be expected to possess and apply at particular stages in their education, such as the end of high school.²⁶ Armin Weinberger, Karsten Stegmann, and Frank Fischer discuss “knowledge convergence,” which refers to the knowledge that students bring together and share when they work in groups or other interactive settings.²⁷ This concept may be revealing when posttests are used or when students are allowed to consult each other during class activities.

The report titled How Should Colleges Assess and Improve Student Learning?: Employers’ Views on the Accountability Challenge provides insight on students’ use of classroom skills in the real world; it is one of the select resources that discusses the concept of skills in general. Not surprisingly, employers strongly advocate the development and assessment of skills at colleges and universities through projects, internships, and other practical applications.²⁸

Numerous resources discuss students’ attitudes toward learning. Fewer cover assessment of those attitudes. However, their methodologies can provide useful guidance. Educating the Net Generation and Serving the Millennial Generation discuss the generation of students that make up the bulk of today’s undergraduates in the United States.²⁹ Remedios and Lieberman’s article discusses the question of whether easy courses necessarily receive more positive evaluations, a frequent concern among educators. While much of the article specifically discusses learning throughout a semester-long course, comparisons of students’ knowledge before and after lessons or units can help library instructors understand how results of single evaluations or assignments affect students’ attitudes toward the class.³⁰

What Is Missing?

Choinski and Emanuel note that relatively little literature is currently available on outcomes-based assessment for one-shot library instruction classes.³¹ There is also a lack of studies that compare effectiveness or usefulness of multiple assessment tools in an academic library setting. Our research aims to fill in both of these gaps in knowledge.

Knowledge, Skills, and Attitudes

Early in the project, the researchers chose to focus on knowledge, skills, and attitudes as the three overarching components of learning. The following definitions, from discussions based on Bloom’s work, apply to this study:

• Knowledge: “relates to the acquisition and application of knowledge and understanding. It deals mainly with learning of an intellectual nature, covering the range from simple recall through to analysis and evaluation of information.”³²

• Skills: the ability to apply prior knowledge and use it to carry out tasks.

• Attitudes: “deals with learning that has a substantial emotional basis and covers the range from having an awareness of feelings through to amending behavior so that it becomes consistent with new values and beliefs.”³³

Breaking our research into Bloom’s categories allowed us to clearly define our areas of analysis. It also helped us align our work with much other research and many past studies conducted in more traditional classroom settings.

Print version (Adobe Reader required)
Although much has appeared in the literature regarding the initiation of virtual reference services, to date a case study discussing online reference service’s integration into an academic library’s current suite of reference services has not been written. At the University of Saskatchewan, the integration process forced the library to take a broader look at reference services as a whole and address several questions: What is reference? How should it be delivered? Who should provide it? On what reference service values are decisions being based? To facilitate the integration, it was necessary to identify the needs of our users and re-examine the core values of our reference service accordingly. This analysis resulted in system-wide changes to all of the library’s reference services. The paper concludes with a set of planning recommendations that will be useful for college and university libraries currently considering, or presently offering, a virtual reference or instant messaging service and aspiring to incorporate the service permanently into their suite of reference services.

Although much has appeared in the literature regarding the initiation of virtual reference services and the “need to integrate ‘digital reference’ into reference,”¹ to date a case study discussing the integration of virtual reference into a library’s current suite of reference services has not been written. At the University of Saskatchewan Library (U of S Library), the process of integrating virtual reference has proven to be both challenging and successful. The library was forced to take a broader look at reference services as a whole and address several questions: What is reference? How should it be delivered? Who should provide it? On what reference service values are we basing our decisions? To facilitate the integration it was necessary to identify the needs of our users and re-examine the core values of our reference service accordingly. This case study will benefit all libraries currently considering, or presently offering, virtual reference and aspiring to incorporate this service permanently into their suite of reference services. The authors also provide a discussion on some lessons learned as well as recommendations for libraries moving forward in this direction.

Virtual Reference Service at the University of Saskatchewan Library

Many academic libraries have attempted to serve the needs of their remote users by offering appropriate reference services based on the technology available, such as telephone, e-mail, instant messaging (IM), and short messaging service (SMS).² A brief history of virtual reference at the University of Saskatchewan Library provides a context for outlining our process of eventually integrating our virtual reference service.

The University of Saskatchewan (U of S) is located in Saskatoon, Saskatchewan in Canada. Geographically, Saskatchewan is a large province, approximately half the size of Alaska, with a small population of just over a million people concentrated in the southern portion of the province. The U of S is a doctoral degree-granting university with approximately 18,500 students (full and part-time).³ The library consists of seven on-campus branch libraries and employs approximately 150 staff members, 40 of whom are librarians.

For the purposes of this paper, virtual reference, also known as digital reference, is defined as the synchronous exchange of information between library reference staff and patrons, using online chat software. “Ask a Librarian Live” was the name given to the virtual reference service at the U of S Library.⁴

In July 2002, a small virtual reference pilot project was initiated at the U of S Health Sciences Library in response to a growing number of health sciences students and professionals studying and working outside Saskatoon. HumanClick, a simple version of chat software, was used. The off-campus library services librarian expressed an interest in the project, and from January 2003 until July 2004 both the Health Sciences Library and Off Campus Library Services used this software, now called Live Person. In 2003, the authors of this paper, along with the U of S library reference coordinator, were asked to establish a library-wide service serving all seven branch libraries. A virtual reference service interest group was formed, consisting of librarians and library assistants who provided traditional reference service and who were interested in providing reference virtually. It is important to note that interest group members volunteered their time to both evaluating potential software and staffing the service throughout its trial period. They recommended trying more sophisticated chat software to ensure that the service trial would not be handicapped by software limitations. LSSI was chosen because of its intuitive interface, attractive pricing, good customer and technical support, proxy server compatibility, and the absence of a patron download. After two years with LSSI, a switch was made to Docutek as it was the only chat reference software at the time that did not require a user download, which might have presented an obstacle to using the service. Unfortunately, Docutek proved unstable, and at the end of the pilot’s third year a decision was made to move to an instant messaging software, Meebo. Table 1 provides a summary of software used for virtual reference at the U of S Library from 2004 to the present.

When virtual reference service was first introduced at the U of S Library, the new service was not integrated into the existing traditional reference model. For three years, there were two separate staffing models, schedules, and statistical tracking systems. The authors clearly recognized the challenges of sustaining two separate reference models, virtual and traditional. Issues surrounding staffing, duplication of services and workload, and assessment were identified as roadblocks for efficient and effective reference service. The need to integrate virtual reference into existing reference service is a challenge faced by many libraries and is vividly described by Janes:

It was as though two trains were running on parallel tracks, headed for the same destination, taking effectively the same route but, yet, somehow not seen as the same … However, if there are two trains going to the same place, eventually somebody is going to ask why. Justifying a separate add-on service, potentially seen as duplicative or even wasteful, in tight budgetary times, is not easy.⁵

Staffing

Our most significant challenge was the different staffing models within the two services. Within the traditional model, a designated reference co-coordinator was assigned the task of overseeing reference services across the library system. Librarians and library assistants provided traditional reference service as part of their assigned duties. During the three-year trial period of virtual reference, volunteers were depended upon to staff the service. This presented human resource issues because the two virtual reference administrators, while recognized as leaders on this project, did not have formal administrative responsibilities. Occasionally, this presented problems when issues of service quality arose or when the virtual reference administrators were asked to evaluate the quality of service provision by volunteer staff. The virtual reference software provided transcripts to assess reference service that was not possible in traditional reference. While the virtual reference administrators utilized many qualitative and quantitative statistics from the software for program assessment, not all of the staff were comfortable with having their transcripts used to formally evaluate their performance, especially since they were volunteering their time. This type of evaluation was not standard procedure for staff providing traditional reference, and it was unlikely that staff would continue to volunteer if this idea was implemented.

The virtual reference service operated on a volunteer model, which proved to be problematic over the long term. Volunteers moved to new positions or became overwhelmed with additional responsibilities. Staffing the service became an issue since virtual reference shifts were in addition to reference desk shifts and were not recognized as reference hours contributed. As the administrators grappled with trying to find additional volunteers, it became clear that the library had never clearly defined the minimum educational requirements for staffing reference service. This presented a major obstacle when trying to recruit additional volunteers for the service. Ideally, anyone who was staffing our traditional reference desk had the ability to staff the virtual reference service, but there were staff members who were reluctant to provide virtual reference service for a number of reasons. The virtual reference administrators could not mandate participation because they lacked the authority to do so and because the service remained a pilot project. Although there were a handful of dedicated volunteers, there were other library staff members who believed that the service was an add-on service that should only be offered when the library was fully staffed. However, with medical, parental, sabbatical, and administrative leaves, it was rare that the library was fully staffed. This sentiment has also been reported in the literature.⁶The differing staffing models between the two services was not feasible for the long term.

Duplication

Another challenge with providing two reference services concurrently was the duplication of effort. There were separate meetings for all reference staff and for virtual reference staff; there were two different schedules for both reference models, as well as two separate means of gathering and organizing reference statistics. Our virtual reference service was often available at the same time as our traditional reference services. Our reference desks are normally staffed from 9 a.m. to 9 p.m. Mondays through Thursdays; 9 a.m. to 5 p.m. Fridays; and limited hours on the weekends. The virtual reference service was offered from 10 a.m. to 4 p.m. from Monday through Friday. Two different reference staff members had to be assigned to monitor our two services since our virtual reference service was often available at the same time as our traditional reference services. While this may not have been deemed negative, it is ideal when one staff member can monitor the virtual reference service while working at the traditional reference desk.

Pilot Project or Permanent Service?

The pilot-project nature of the service was yet another challenge for the virtual reference administrators. Unlike the traditional reference model, the virtual reference service started as a library-wide trial project and maintained this status for three years. The temporary status proved challenging for planning, marketing, funding, and staffing the service. Although the administrators were aware that marketing the service would likely increase its use, they were reluctant to do so, in case user expectations were created for a service that might cease to exist by the next academic year. This was yet another reason for the recommendation to integrate the virtual service into the current suite of reference services.

Response to the Service

From the service’s inception, both qualitative and quantitative statistics were collected for assessment. Some statistics were essential and have been collected from the beginning, and some questions were introduced later as we tried to compose a clearer picture of our users. It is important to note that the service moved from a commercial, virtual reference product at the beginning of the 2007–2008 academic year to an instant messaging service that lacked the ability to gather usage statistics and administer user surveys. At this time, the user survey was separated from the online chat form and appeared as a separate link on the “Ask a Librarian” page. Patrons were not prompted to answer the survey at the end of their session, but some chose to do so. Survey responses decreased significantly once that separation was made.

Of prime interest was the number of questions we received through virtual reference, and how the number compared to traditional reference questions (comprised of in-person, telephone and e-mail questions) (table 2). Questions asked through virtual reference do not represent a large percentage of the total number of questions asked, but they do represent a growing percentage of the whole. Significant to note is the downward trend of traditional reference questions and the growing number of virtual reference questions. Over the last five years the number of virtual reference questions has increased 291 percent. Undergraduates remain our largest patron group, although not as markedly as at the beginning of the project, while graduate students’ questions now make up almost one quarter of our total, up from 16 percent in 2004–2005 (table 3).

For the first three years, we tracked the type of virtual reference questions being asked; research questions represent the majority (table 4). Research questions were defined as those about how to find information on a specific topic, how to find journal articles or books, how to search a database, or how to cite a resource. Definitions of other types of questions can be found in appendix A.

Recommendation for Integration

The aforementioned challenges prompted the virtual reference administrators to include the following as the primary recommendation in the final reports for both Year II (2005–2006) and Year III (2006–2007) of the virtual reference pilot project: “that an integrated reference model be developed and that virtual reference be subsumed under this service model.”⁷

The authors stated:

Ask a Librarian Live has become a vital component of reference service at the University of Saskatchewan Library, and should be mainstreamed into reference service offered at the University of Saskatchewan Library … In this context, it is essential that virtual reference become part of an integrated approach to reference service … An integrated reference service would treat all types of reference service at the Library equally. This would require providing reference service via one or more modes, and in one or more locations, depending on the needs and anticipated needs of our users, and within available resources.⁸

The process of integrating the virtual reference service at the U of S necessitated that the needs of library users be reassessed and that what is meant by ‘reference service’ at the library be re-examined. To undertake such an assessment there needed to be a return to the foundations of reference service.

Origins of Reference Service

Samuel Swett Green is often considered the “American pioneer of reference service.”⁹ In 1876, at the first conference of the American Library Association, Green presented a talk titled “The Desirableness of Establishing Personal Relations Between Librarians and Readers in Popular Libraries.” The ideas in this presentation were seen as quite revolutionary and controversial for the time. That same year, Green published an article on the same topic in Library Journal.¹⁰ In these two works, Green stressed that the role of librarians was not only to be the keepers of books and organizers of information, but that librarians must expand their role to include reference services. This was the first time reference services were clearly identified as an important component of librarianship. Tyckoson summarized Green’s four core functions of reference librarians:

• Instructing the reader in the ways of the library
• Assisting the reader with his queries
• Aiding the reader in the selection of good works
• Promoting the library within the community.¹¹

Tyckoson argues that although there have been many changes and improvements to reference services in libraries since Green’s time, these founding four philosophical functions remain unchanged.¹²

Values of Reference Service

As Green identified the original core functions of reference librarians in the late eighteen hundreds, core values for reference service have also been identified.¹³ Tyckoson identified these values as accuracy, thoroughness, timeliness, authority, instruction, access, individualization, and knowledge.¹⁴ In Our Enduring Values: Librarianship in the 21st Century, Gorman discusses broader values for the librarian profession, such as stewardship, intellectual freedom, rationalism, privacy, and equity of access to recorded knowledge and information.¹⁵ While these values are foundational and important, each library must consider the needs of its patrons and place different emphasis on particular values to best serve them.

In the 1997 article “The Shape of Things to Come: Values-Based Reference Service for the Largely Digital Library,” Ferguson and Bunge make a compelling argument to retain traditional library values “while exploring new values such as integrating technologies, maintaining holistic computing environment, delivering core services through the network, making technology work for all, and collaborating across administrative lines.”¹⁶ He advocates providing services to users where they are and when they need it as well as maintaining personal assistance. To move forward, libraries need to rethink their core values and perhaps identify new values to meet users’ changing needs and expectations.

It was the value of “equity of access” discussed by Gorman ¹⁷ that stood out for the authors as they struggled with integrating reference services. At the U of S Library, patrons have increasingly accessed library resources online. Because of the vast size of Saskatchewan, many students choose to study by distance or online. These students may never have the opportunity to visit campus or use the library in person but, nevertheless, deserve equal access to library services and resources. Gorman points out that technology such as virtual reference can be used to provide equitable access to reference service, explaining that “none of these methods is as effective as human-to-human reference, but they are far better than no reference service for the rural, the home bound, or other such seekers of knowledge and information.”¹⁸

Print version (Adobe Reader required)
Over the past decade and a half, reference and information services have increasingly moved away from library reference desks and away from libraries’ print collections into the electronic world. This article describes a study that addressed two research questions related to the changing reference and information services environment:

1. What are the current trends in the provision of reference and information services in terms of user behaviors, librarian behaviors, and information sources being used?
2. What is the basic model of the current reference process?

Data were gathered via focus group interviews with reference and information service educators and via discussions at a town hall-style meeting of faculty members and doctoral students interested in virtual reference education. The study results indicate a shift toward an increasingly interactive, collaborative reference model, in which both the reference librarian and the reference user play the roles of information seeker, information receiver, and information creator. The article concludes with a model of this process and with a discussion of implications for the provision of reference and information services.

In response to the electronic information explosion, members of the public are increasingly using the Internet and the World Wide Web to search for information to support their personal, business, and other information needs. At the same time, a growing number of libraries are offering reference and information services (RIS) via a variety of virtual technologies, and there has been an increasing shift from face-to-face reference interactions that relied on paper-based information resources to virtual interactions using electronic information resources. It is possible that these developments have changed the ways in which people use RIS and possibly even the ways in which reference providers deliver reference services to their users.

With their inside view on research in this area and with their close connections to practice, RIS educators are in a unique position to assess changes occurring in the field of reference. This study sought to learn from RIS educators their views on the current state of reference services, with a focus on the effects of the online information explosion on the virtual reference process.

Research Questions

Because of this movement from face-to-face to virtual venues and from paper to electronic resources, today’s reference transactions take place within a dynamic and rapidly changing information environment. The fundamental question now becomes “what changes, if any, has the exploding digital information environment caused in the reference process?” Without an understanding of these changes, reference providers cannot tailor their services to meet the needs of today’s information users.

Two research questions drove this study:

1. What are the current trends in the provision of reference and information services, in terms of user behaviors, librarian behaviors, and information sources being used?
2. What is the basic model of the current reference process?

Literature Review

There is a large research literature concerning the design and delivery of reference and information services. This review includes recent empirical studies and theoretical analyses related to current trends in the delivery of reference services and major models of the reference process.

There has been no firm consensus on the use of terminology defining virtual and digital in the context of reference service.¹ Lipow deemed the phrases “virtual reference” and “digital reference” interchangeable.² Lankes and Shostack noted that digital, virtual, and e-reference are the same type of service, in which librarians serve as intermediaries to assist users in finding information in a digital environment.³ This article will use the term virtual reference to cover all of these terms.

Trends in the Delivery of Reference and Information Services

Tenopir and Ennis sent questionnaires to the directors of large academic libraries in the United States and Canada every three years from 1991 to 2001. They identified a number of changes in reference services, including an increase in the range of available electronic resources, a de-emphasis on mediated services, an increase in the range of available online services, an increase in the web as the standard platform for resources and services, a decrease in the number and an increase in the complexity of questions, and an increase in users’ computer skills.⁴ To be sure, virtual reference services are now standard offerings in most public, academic, and large special libraries. Lankes has suggested that “digital reference is a phenomenon that is firmly rooted and ever-expanding in practice.”⁵ He proposed a digital reference research agenda that combined the examination of policies, systems, evaluation, and behaviors to examine a central question: “How can human expertise be incorporated effectively and efficiently into information systems to answer information seekers’ questions?”⁶

It’s not just reference services that are increasingly digital in nature, but the sources used in virtual reference work as well. Shachaf and Shaw analyzed 1,851 e-mail and chat reference transcripts from one academic and one public library and found that 96 percent of the sources used in the responses were electronic.⁷ Similarly, a study by Bradford, Costello, and Lenholt showed that only 1.8 percent of an academic library’s 9,587 titles in the print collection were used to answer reference questions, and that academic reference used online sources significantly more than print sources.⁸

Many studies have examined specific aspects of virtual reference services, focusing on issues such as reference interviewing techniques, verbal and nonverbal communication strategies, and teaching practices.⁹ Additionally, studies have reported on collaboration during the research process and on instruction to users.¹⁰

Other studies have evaluated reference transcripts to assess librarian performance, question-answering success, user satisfaction, and answer completeness and accuracy.¹¹ Unobtrusive analyses of reference data have looked to categorize the types of questions posed to virtual reference services.¹² Still other related work has studied the expansion of RIS into new technologies, such as instant messaging (IM), text messaging or SMS, and Second Life.¹³

Major Models of Reference and Information Services

A number of researchers have proposed models of the reference process. The traditional face-to-face reference service model anchors the interaction between the user and the librarian in some service point, typically a reference desk. In this environment, a library user may approach a reference librarian with an inquiry of any type or complexity.¹⁴ While this model symbolizes values including “ease of access, equity, and high-quality service,” the disadvantages include inflexibility in the use of library staff, duplication of effort, lack of accountability, high cost, and reinforcement of the image of librarian as clerk.¹⁵

Tiered reference service has been described as an alternate to the traditional model, often dividing the reference desk into two or more service points. Initially made popular by the Brandeis model, the tiered reference model’s defining feature is that paraprofessionals or trained student assistants answer a majority of simple reference questions so that reference librarians may be reserved for handling more demanding research questions.¹⁶

Another influential model of reference involves the role of reference librarians in question-answering activities versus their role in providing instruction to users. With the liberal, or maximum, model, the librarian’s responsibility centers on delivering an answer in response to a user’s inquiry. The librarian does not attempt to educate the user in the process; rather, he or she puts all effort into finding accurate and credible information. Conversely, the ultimate goal for the conservative, or minimal, model is to train users to make use of the library independently, as the process of finding information is valued above the information itself. Despite the conflicting objectives of the liberal and conservative models, the approaches may coexist in practice.¹⁷

With the increased availability of technologies and the drive to increase library staff efficiency and reduce costs per reference transaction, some libraries have adopted the Call Center model.¹⁸ In this model library staff operate as “agents … tak[ing] calls at computer workstations where they … have ready access to databases, lists of frequently-asked questions and answers, prewritten scripts for particular situations, and other tools needed to deliver … information.”¹⁹

In academic libraries, a dynamic model of reference service has been termed “information commons.” On one level the phrase describes an “exclusively online environment in which the widest possible variety of digital services can be accessed via a single interface,” while concurrently denoting “a new type of physical facility … designed to organize workspace and service delivery around the integrated digital environment.”²⁰ This model reflects the ways in which academic libraries are responding to the demands for technology, combining information resources and reference assistance, and creating collaborative workspaces for acquiring and shaping knowledge.²¹ Most recently, Pomerantz et al. presented a model of the virtual reference process that highlighted five key question-handling functions: question acquisition, triage, answer formulation, tracking, and resource creation.²²

While each of these models is useful for thinking about reference services in the current information environment, what is missing from the literature is a model of the entire reference process, including both librarian and user behaviors, that takes into account the recent changes in the information environment. This study sought to fill that gap by creating a model that brings together librarian and user behaviors to provide a full picture of the impact of the electronic information explosion on the reference process.

Methods

Data for this study were gathered using two methods: focus group interviews and a town hall meeting. For the focus group interviews, a list of faculty who teach courses related to RIS and digital libraries was compiled based on a review of the websites of ALA-accredited master’s programs. Sixteen participants from thirteen universities took part in three focus groups held from January 2008 through January 2009 at the annual meetings of the American Society for Information Science and Technology (ASIS&T) and of the Association for Library and Information Science Educators (ALISE). There were four faculty members in the first focus group, five in the second group, and seven in the final group. Their experience teaching reference and digital libraries courses ranged from one year to more than thirty years. Six participants had been teaching reference for one to five years; six had been teaching for six to ten years; and four had been teaching reference for eleven or more years. They came from a range of academic ranks. Five held the title of instructor or lecturer; three were assistant professors; five were associate professors; two were professors; and one was an associate dean. Each focus group session lasted between one and one and a half hours.

The focus group guide was divided into five sections: (1) introductions, (2) curricular coverage of digital/remote reference, (3) barriers to teaching reference, (4) future of reference services, and (5) conclusions. The sessions were audio taped and later transcribed for analysis.

To triangulate the focus group data and to increase the number of participants providing data, the researchers collected additional data at the ALISE 2009 annual meeting from audience members at a town hall–style meeting of faculty members and doctoral students interested in virtual reference education. It was advertised through the ALISE conference program, and approximately fifty-five people attended and took part in the discussion. The session consisted of a panel-led group discussion of four topics related to virtual reference and information services: (1) current LIS curricula for digital reference, (2) barriers and obstacles to teaching digital reference, (3) future directions of digital reference services, and (4) ideas about how LIS education can best reflect the future direction of digital reference services. Several note takers were present to record the responses and discussion. The notes were compiled and transcribed for analysis.

The focus group transcripts and the transcribed town hall meeting notes were analyzed using the constant comparative method, the most common method for analyzing qualitative data. The constant comparative method involves repeatedly reading through a set of data, grouping and regrouping individual pieces of data into categories to create a coding scheme that addresses the research questions.²³ Typically, analysis of the resulting coding scheme leads to the creation of a typology, model, analogy, or other method of presenting the data.

Findings

The resulting coding scheme included six major themes relating to the current and future state of virtual reference: (1) the convergence of multiple modes of reference, (2) the burgeoning range of information resources, (3) reference as a collaborative process, (4) reference in the library 2.0 mode, (5) the shift from librarian-as-searcher to librarian-as-evaluator, and (6) the possible demise of ready reference. These six themes are explained below with supporting literature from previous studies where applicable.

The Convergence of Multiple Mode of Reference

Rather than viewing face-to-face reference, chat reference, IM reference, e-mail reference, telephone reference, and so on as unique services, each one separate from the others in staff and resources, the trend seems to be to view them all as subsets of “reference and information services,” or “question-answering services.” As one focus group participant explained, “the trend is toward thinking how to integrate.” While the use of each mode of delivery might require some variant skills and techniques, they are all part of what one focus group participant labeled “the reference continuum.”

One of the town hall meeting participants pointed out that the traditional practice was for librarians to respond using the medium that the user had chosen to contact the librarian. Over the past decade or so, many librarians have been moving toward responding using the most appropriate medium for answering the question, regardless of the initial medium of contact. For example, if a user calls a library to ask a fairly complicated question, the librarian might choose to research the question and respond via e-mail to create a written record of the response, instead of automatically responding via the phone.

This trend toward the convergence of reference media seems to be occurring in both public and academic libraries. In speaking about academic libraries, one focus group participant said that:

the trend in a lot of academic libraries is consolidation of service points, so at every level we’re having to anticipate that our students will be involved in all these modes of delivery. They’re not going to be able to specialize in digital reference in most settings, or in telephone, or face-to-face.

The other group members agreed that the same is true for public libraries.

The convergence of reference modes is creating multitasking problems for many librarians, and as a result, today’s reference librarians must multitask on a nearly continual basis. As one of the focus group participants said, “If you’re on the desk, you’re also answering the phone. You may be doing chat while you’re also having to monitor e-mail … and there’s no opportunity to handle just one patron at a time.”

Print version (Adobe Reader required)
This article compares the concept of the “problem patron” in the library and information science (LIS) and nursing literatures as the basis for developing a new framework for use in LIS. The trend in the LIS literature has been to identify either the patron or the patron’s behavior as the problem. The nursing literature uses interactionist theories to contextualize the so-called problem within a larger framework that includes, among other things, the nurse, hospital-related norms of behavior, the patient care environment, the philosophy of care, and the patient’s own life experiences. This paper examines theories of stigma, deviance, and labeling, among others, as they have been used in the nursing literature to examine the process and effect of labeling.

I argue that the work on labeling found in the nursing literature provides the foundation for a new framework to think about the “problem patron” in LIS. In the proposed framework, I define problem behavior at three different levels: the community, the library, and the individual. Using this framework is helpful for thinking about solutions because it encourages us to respond to the “problem” at the level where the behavior is labeled as deviant. This framework is used to explore solutions offered in the LIS literature for the problems that can be identified at each of these different levels.

Librarians experiencing difficult interactions with patrons are not alone. Many professionals who work with the public can recount tales of challenging interactions. As librarians we can learn from reflecting both on the literature of our own profession and on that of other disciplines. This paper looks at LIS and nursing literature from the late 1990s to the present to see how it approaches the “difficult patron” interaction. The unique contribution from the nursing literature comes from its use of symbolic interactionist theory to explore the “difficult” patient interaction. The label of symbolic interactionism covers a “relatively distinctive approach to the study of human group life and human conduct.”¹ In exploring the literature of each field, we can identify some emerging themes that encourage us to reevaluate how we approach challenging patron situations. Using insights gained through reviewing the literature of both fields and reflecting on symbolic interactionist theory, I suggest a framework for thinking about patron behavior that might offer insight for problem solving. This approach identifies three levels at which behavior can be labeled as a “problem.” Understanding the level at which the behavior is defined as a problem is a key factor when searching for a solution. Through this framework, we can explore solutions and resources found in the LIS literature to aid librarians working through or preparing for challenging interactions.

Searching the Library and Information Science Literature

Searching LIS databases shows us that the term “problem patron” is used in the LIS field. In the Library and Information Science Abstracts (LISA) database, searching for the descriptor “problem patrons” with the keyword “librar*” from the late 1990s to the present yields an average of two to four articles each year, except in 2002, when there were twenty-seven articles listed, and 2003, when seven were listed. Many of the 2002 articles were simultaneously published in the journal The Reference Librarian and in Sarkodie-Mensah’s Helping the Difficult Library Patron: New Approaches to Examining and Resolving a Long-Standing and Ongoing Problem (2002).² A search of the Library Literature database using the subject term “problem patrons” from the late 1990s onward shows a consistent presence of one to three articles on the topic for most years, with a higher number (four to twenty-one articles) for some years from 2000 onward. The phrase “problem user”‘ is not used as a descriptor or subject term in the LISA or Library Literature databases, and keyword searches using the term provides limited and mostly irrelevant results. The consistent number of articles over the last decade on the topic of the difficult patron, along with the more recent spikes in articles by more than thirty authors, suggests that dealing with challenging situations is a very real issue within libraries.

Searching the Nursing Literature

Searching health care databases shows us that nurses experience challenging interactions with their patrons as well. The Cumulative Index to Nursing and Allied Health Literature (CINAHL), an important database for nursing, uses the phrase “problem patient(s)” as a subject heading and major heading. Using the subject search terms “problem patient” and “nurses” in CINAHL finds seven to eight articles for most years between 1996 and 1999. In 2000, twelve articles appeared, with a decrease to three or four articles in the following years, until 2004, when between five and six articles per year became the average, continuing to the present. In addition to CINAHL, I searched Medline and PsycINFO using a variety of terms reflecting the concepts of “problem patient” and “nursing” with limited results. The consistent pattern of articles suggests that the topic of the problem patient is an ongoing one in the nursing literature.

What is the Problem?

While exploring the LIS literature does not provide a clear definition of the problem patron, we can see that a clear dichotomy in thinking exists. We talk about problem users and problem behaviors. Some definitions of the problem patron focus on categories of problem users in which the problem is a person or group of people with certain undesired characteristics. The literature shows us that mentally ill or homeless patrons are often categorized as problem patrons.³ Vandals, criminals, angry patrons, and unattended children are a few other categories of people identified as problem patrons.⁴ While the literature offers examples of people being categorized as problem patrons, is this the best way to view the situation?

The second common approach to viewing the issue is to focus on a specific behavior as the problem. Chelton highlights that as early as 1994 there was the recognition of a shift in thinking from “categories of ‘problem people’ to categories of ‘problem behaviors’ regardless of the people doing them.”⁵ Blessinger notes “a problem patron could be defined as someone who infringes on others’ enjoyment of the library by displaying behavior that is deemed destructive, criminal, bothersome, offensive, or otherwise inappropriate to the norms of behavior in libraries or society.”⁶ Redfern highlights a number of “problem patrons,” including those who are disruptive, difficult, mildly irate, violent, or mentally deranged, as well as those who monopolize librarians’ time, are suspected to be under the influence of drugs or alcohol, are emotionally disturbed, and are sex offenders.⁷ Looking at these definitions and lists we can see that behavior is the focus.

People are multifaceted, and perhaps public service interactions are changing in response to awareness that behavior is a more effective area to target to produce change. Labeling a problem patron as homeless does not accurately reflect the challenge that person brings to the library. Focusing instead on behavior allows the librarian to define the same expectations for all patrons. This behavioral focus gives a librarian a starting point for problem solving, or a concrete goal to work toward.

All the definitions above reflect the idea that the problem lies within the patron. As Chatoo explains, “The word ‘problem’ when used as an adjective, as it is in the phrase ‘problem patron,’ denotes the meaning of dealing with a problem of conduct or social relationship and ‘difficult to deal with.’”⁸ In looking at both the problem patron and the problem behavior approaches to challenging interactions, we could conclude that “problem” is not only the patron’s behavior but is a reflection of the librarian’s inability to deal effectively with the situation. This idea of the problem as something beyond the patron or the behavior is interesting. In this range of approaches to defining the problem patron, we can see that in most of them the librarian judges or labels the patron. The term “problem patron” appears broad in scope, and a good and socially appropriate person can still be labeled a problem patron as a reflection of their request.⁹ It is important to recognize that “problem” is an assigned label and isn’t reflective of the individual as a whole.

“The Difficult Patient”: Lessons and Insights from Nursing

Nursing, like LIS, is a profession that involves daily interaction with members of the public. While the term “difficult patient” is used in the nursing literature, and specific behavior is connected to the concept, we can see that the nursing literature clearly shows an understanding of the problem as something arising from an interpersonal interaction. Insights have been gained by using theory including deviance and labeling, stigma, power, and grounded theory. This has led to the understanding that not only is the label subjective, but also that the act of applying a label has an impact.

Print version (Adobe Reader required)
This paper presents a theoretical framework for understanding the process of generating and storing knowledge from online reference service transactions. Terminology for this study has not been sufficiently developed in scholarly work, so this paper uses the phrase “online reference knowledge base” (ORKB) to denote a place for storing knowledge generated from online reference services. In addition to reevaluating the present role of ORKBs through interviews with experts and practitioners who are most closely linked to online reference services, the paper attempts to offer a blueprint of adequate principles and guidelines for the future development of ORKBs.

Online reference services emerged during the mid1990s as an extension of existing reference services in libraries. Today, online reference services represent a standard library service in a networked environment. They have added archival value because they provide a way to store communications between the reference librarian and the user, preserving generated knowledge and its repeated use in a search for new information. This archive of users’ answered questions is often referred to as the “knowledge base” of the service, which is alluded to by the name of the largest online reference service in the world—QuestionPoint (www.oclc.org/questionpoint). That is why I use “knowledge base” here; however, the literature has not sufficiently explored this term. Neither the very concept of an existing online reference knowledge base (ORKB) nor the possibility of its development has been explored thoroughly. Therefore the purpose of this article is to question the current role of knowledge bases as an archive of users’ answered questions in the context of online reference services, with the basic goal of establishing appropriate principles and guidelines for ORKBs’ future development.

At the beginning of the article, I outline the concept of ORKBs as an archive of users’ questions and answers. I will talk about frequent question card files, which is the precedent to the modern ORKB, after which I will discuss the current theoretical accomplishments in this area. In the practical sense, I will refer to QuestionPoint, and to the Croatian public libraries’ Ask a Librarian service (“Pitajte knjižničare”), which I had at my disposal for analysis and research through their administrative interfaces.¹ I explore the online reference services’ standardization as well the presentation of the Digital Reference Electronic Warehouse (DREW) project promoted by R. D. Lankes and S. Nicholson, both online reference services experts. I conducted in-depth interviews with online reference service expert s; I attempted to ascertain the experts’ positions on the possibility of the development of ORKBs. On the basis of the conducted analyses and research, I will propose principles and guidelines for developing efficient ORKBs. Finally, I will suggest further research in this area.

The Concept of Knowledge Bases in Online Reference Services

Developing the Concept of Knowledge Bases in Reference Services: Frequent Question Card Files

As far back as the nineteenth century, reference librarians’ need to facilitate knowledge exchange between themselves led them to use special cards or card files, which were created and maintained mostly by volunteer reference librarians with the goal of storing information related to users’ questions and librarians’ responses. These librarians organized the card files either alphabetically or by subject and made them available to other reference librarians to answer similar requests in the future. The annotated information usually related to the discovered sources, that is, answers to certain questions, especially those repeatedly asked or those that demanded in-depth research. In addition, Anglo-American expert terminology often referred to frequently-asked-question card files by various names, such as “quick reference file,” “query file,” “useful reference file,” “information file,” “file of answered questions,” “vertical file,” etc. However, the names that became most commonly used are “frequent question card files” and “ready reference card files.”²

Although there is evidence of the existence of such card files in every library for more than a hundred years, their purpose, structure, and contents have not been sufficiently processed or represented in library-related literature.³ With the advent of information technology in libraries, some librarians have attempted to automate these card files.⁴ Perez characterizes frequent-question-andanswers card files as beta-test knowledge bases.⁵ There is no doubt that the frequent-question-andanswer card files can be viewed as a concept that precedes the emergence of ORKBs more than a century later.

The Concept of ORKBs As an Archive of Users’ Answered Questions

The usual understanding of an ORKB is of an archive that contains answers to users’ past questions. In the beginning of ORKB development in the mid-1990s, online reference services were no more than an ordered archive with questions and answers along with information regarding the time of query and the time necessary to give an answer. In the example of the Ask a Librarian service offered by the Croatian public libraries, it is clear that these are not knowledge bases but in fact “answered questions bases,” or rather, “questions and answers bases.”⁶ However, in the case of the QuestionPoint service, the archive of answered users’ questions is explicitly referred to as a “knowledge base.”

According to Peter Armenti, one of QuestionPoint’s founders, an ORKB is a “searchable electronic archive of question and answer pairs as a support to the users’ information needs.”⁷ First, a knowledge base is built to save time. Questions that have already been answered are not answered again; instead, the answers are forwarded to the user automatically from the knowledge base. The other reason to develop such a base, according to Armenti, has to do with distributing reference librarians’ knowledge as they participate in the project to answering users’ questions.⁸ In this way, ORKBs begin to incorporate the purpose of another kind of knowledge base found in the field of knowledge management: knowledge distribution. Apart from that, it is worth mentioning that future explorations of the economic effects would be beneficial because setting up a questions-and-answers archive will surely contribute to significant savings in the business of a library.

Apart from being an electronic pair archive (question and answer), a knowledge base should also contain information regarding how a certain question was answered, including information about research techniques, strategies, sources, etc. Armenti offers the following advice to consider before building a knowledge base for an online reference service:

• Identify the target user group who will be using the knowledge base.
• Determine the goal or purpose of the knowledge base (e.g., saving time in answering previously answered user questions).
• Identify metadata types, which will make up the transaction entry in the knowledge base.
• Create suitable documents that will contain instructions on how to search through the knowledge base.
  • Constantly check every new idea brought to the reference librarian related to the design of the knowledge base and that can stem from everyday use of the existing questions and answers.⁹
  • In a 2005 work, Chen demonstrates best practices in the use of web technology to build a web-based system in which reference librarians can record their reference experience.¹⁰ This work is significant because the author treats such a knowledge base as not only a key component for online reference services as a separate system, but also as an aid in the reference librarians’ work in the context of traditional reference services. Any reference service provider can find the reason for building such a system in everyday work with users, where users often ask questions that have been previously answered, but the provider is unable to remember the answer. In that sense, Chen sees particular advantages in building such knowledge bases for the following reasons:
• Time is saved for the reference librarian.
• The knowledge base becomes an apt tool for training new reference librarians.
• The knowledge base also becomes an archive containing the answered users’ questions that is available to other users.
• The knowledge base thus develops into a significant component of the online reference service.¹¹