What is the best model of reference service?

ABSTRACT

REFERENCE SERVICE IS BASED UPON A SET of core functions that have remained valid since the earliest days of the public library movement. Inherent in those functions is a set of core values, including accuracy, thoroughness, timeliness, authority, instruction, access, individualization, and knowledge. Models of reference service that emphasize different aspects of those values take very different forms. In determining the best model for a specific library, the values of the community that the library serves must be taken into account. No one model is best for all libraries.

For anyone using libraries today, reference service is a standard feature. Regardless of the type of library, the size of its collections, or the demographics of its users, patrons expect to get help with everything from complex research projects to finding materials in the collections. Service has become almost synonymous with libraries.

This was not always the case. Reference service is a relatively recent development in library history, and the methods in which reference services are provided are still evolving. Administrators and reference librarians alike struggle to develop the most efficient and effective means of providing reference service to their users. During a time when change seems to be the only constant, librarians are seeking new models for providing service to their users.

Beginning around 1980, a number of innovative methods for providing reference service began to be created and promoted in the literature. From the Rethinking Reference project (Rettig, 1992) to the Brandeis model of tiered reference service (Massey-Burzio, 1992) to 24/7 online call-center-based electronic information services (Coffman, 1999), each new method has been promoted as the next revolution in the provision of service. Those who create these methods are often lauded for their innovations; while at the same time, those who retain older models often worry that they are becoming marginalized and outdated. Predictions of the demise of libraries in general--and of reference services in particular (Campbell, 1992)--have left many reference librarians wondering what the future holds in store. Some wonder if reference has a future at all.

Many see the values that attracted them to the field of reference as slipping away in a sea of electrons. Newly hired reference librarians often reinforce this perception as they show clear preferences for the Internet and databases over traditional reference works and the library catalog. At a time when fear seems to be replacing faith as one of the prime motivating factors among reference librarians, we need to reexamine the foundation of what we do and identify the core values inherent in reference services. By viewing change within those values, we can better understand how new models of service can enhance--or hinder--the provision of reference service.

LIBRARIES AND COMMUNITY

With very few exceptions, libraries are not independent organizations. Virtually every library serves a broader institution and is directly responsible to that institution. Public libraries serve the people living within specific geographic and political boundaries--most commonly a city or county. School libraries serve the students of a specific school or district. Academic libraries serve the students and faculty of a specific college or university. Special libraries serve the staff of a specific company, government agency, or organization. In each case, to be effective the library must meet the unique information and service needs of its own community. Expectations for information and service needs vary greatly from one community to another (even for the same kind of library), but the role of each library is to meet particular expectations of its own community. Although the function of a corporate library may appear to be very different from that of a public library, each serves as the information center for its own clientele.

In order to survive, the library must develop a symbiotic relationship with its parent community. Successful libraries serve their community and are rewarded for that service. The community benefits from the information and services provided by the library; the library benefits from receiving a strong political, economic, and cultural status. Both community and library grow and evolve together. A community that neglects its library will need to develop other means for satisfying its informational needs--or go without. If the library fails to provide useful information and services, the community will turn elsewhere for those functions. In each of these cases, the library will wither and die--and the community will also suffer.

Although libraries are highly complex organizations, their activities can be boiled down to three distinct functions. First, libraries select and collect information relevant to their community. Historically, this selection process has been the heart of what libraries do. Librarians sort through the universe of available information and bring together the portion that is most relevant to the parent institution. To be successful, librarians must know the needs of their community and reflect those needs in the library's collection. From ancient times, libraries have identified and collected information of value to their community.

The second function of the library is to organize the information within its collections. This process is not purely an intellectual exercise, but a necessity that enables librarians and patrons to locate needed information when it is needed. When a library consisted of a single shelf of books, it was relatively easy to browse through the titles to find the one or two pertinent sources. When the collection began to fill a room, indexes and catalogs were developed to guide users. When parts of the library became digital, new tools and organizational skills were needed. As the library became larger, more sophisticated organizational tools were developed, from cataloging rules to controlled vocabularies to MARC records. Each tool was designed to be more efficient and effective in guiding users to information. Regardless of how an individual library is organized, the tact that it is organized adds value to the information contained within its collections.

The final, and most recent, function of the library is to provide direct information services to members of the community. As libraries became larger, even the organizational tools developed by librarians were not sufficient to help patrons find the specific information that they needed. When members of the community required help, they turned to the most knowledgeable resource--the librarian. As faithful servants of the community, librarians took on the new responsibility of helping users find what they needed. This is what we now call reference service.

The Birth of Reference Service

Reference service as we know it today is a direct outgrowth of the nineteenth-century American public education movement (Garrison, 1979). Before universal public education, the vast majority of the American public was illiterate; libraries only benefited the elite minority. Before the 1850s, libraries were limited to colleges, government, and subscription libraries that served small segments of the entire community. During the late nineteenth century, the relationship of the library to the community changed significantly.

The primary goal of the public education movement was to produce a literate working class. Employers felt that having a literate workforce would be more productive on the job; but literacy would also create a better society for those employees off the job. One direct consequence of the public education movement was the development of true "public" libraries. Unlike subscription libraries, which were open only to paying members, the new public libraries were available to everyone. Communities that established public libraries were often surprised at how well they were received. Not only did the elite patronize the library, but many of the formerly disenfranchised working class took advantage of their new education. There was only one problem--they did not know how to use the library. Thus reference service was developed to solve that problem.

The beginning of reference service is generally attributed to Samuel Swett Green, who in 1876 published the first article on helping patrons use the library. While it is doubtful that Green actually invented the idea of reference service for library users, he was the first to speak publicly about the concept and was the first to discuss it in writing. In both his speech to the first meeting of the American Library Association and his Library Journal article, Green discussed the need for librarians actively to assist members of their communities in using library resources. While the term "reference" did not evolve until several decades later (Rothstein, 1953), the publication of Green's article helped to popularize the new concept of reference service. That patrons appreciated such service is evident from the fact that virtually all modern libraries still offer some version of "personal relations between librarians and readers."

COMPONENTS OF REFERENCE SERVICE

In his seminal article Green (1876) introduced four prime functions for the reference librarian. These functions remain the basic components of reference service today. Briefly stated, the reference librarian:

* Instructs patrons how to use the library

* Answers patron queries

* Aids the patron in selecting resources

* Promotes the library within the community.

The first function came about as a direct result of public education. The newly educated members of the community wanted to gain access to the information resources of the library, but had no idea how a library was organized or functioned. Teaching the community to use the library became an extension of the overall education process.

As community members used the library, they often wanted information about specific topics. Naturally, they turned to the librarian for expert assistance. By applying knowledge of the library and its organizational tools, the librarian was frequently able to find that information. Over time, the reference librarian as question-answerer was born. For many library users, this remains the most visible image of the reference librarian today.

However, not every patron came to the library looking for specific factual information. Many came simply wanting something to read--and, after reading one work, they often wanted to read something similar. In the early days, the librarian was charged with leading readers to the "good works"--those that emphasized the benefits of honesty, hard work, and strong moral character. As times and tastes changed, this guidance process evolved into the Reader's Advisory services that we know today.

Finally, Green saw that the library had to demonstrate its value to its parent community. The best way to do this was for the librarian to be highly visible and to become an integral part of the research process. Providing direct service to users would lead those users to appreciate, and by extension to support, the library. Green advocated having the librarian mingle freely with users. While the simple act of mixing with patrons has evolved into today's polished outreach and marketing campaigns, the idea remains the same--to make the library prominent in the minds of its community members.

Many changes have taken place since the publication of that first article. Samuel Rothstein (1955) detailed the growth and development of reference service from the earliest times until the mid-twentieth century. Others have followed changes in reference service through the 1990s (Tyckoson, 1997). Many new models, new tools, and new ideas have been discussed, implemented, and accepted into practice. However, despite all of these changes, the basic functions of reference service have remained essentially constant.

INHERENT VALUES OF REFERENCE SERVICE

Each of Green's four functions of reference service is built upon a set of core service values. Rarely mentioned or studied, these values underlie each of the primary activities of the reference librarian. An examination of the core values of reference service is essential to understanding the basis for defining that service and for understanding the conflicts that arise around it.

The first of Green's functions involves teaching patrons how to use the library. In more modern terms, we call this "bibliographic instruction" or "information literacy." The act of instruction carries with it three distinct facets: the subject content being taught; the process of using that information in the research process; and the ability to critically evaluate information. To be successful, the librarian must teach the patron a specific knowledge base (such as the fact that Education books within the Dewey Decimal System are in the 370 call number, or that the ERIC database indexes education journals and reports), the context and process for using that knowledge (the Dewey system divides knowledge into ten discrete categories and like subjects are categorized together in a hierarchical manner; or searching the ERIC database is most effective when using Boolean combinations of controlled language subject descriptors), and the ability to evaluate the information found in the search (this book is out of date or this ED document is directly relevant to my research project). In this respect, the skills implicit in teaching patrons how to use the library are factual knowledge, understanding the research process, and critical thinking.

Factual knowledge of a different kind is also valued in answering patron queries. In teaching patrons how to use the library, the librarian is transferring knowledge. In answering questions, the librarian is seeking knowledge from the broader world for the patron. In the first example, the librarian knows the process and teaches it to the patron. In the second, neither the patron nor the librarian know the answer in advance; they seek it together. The values associated with answering questions are not related to the process of answering, but to the answer itself.

The most common value associated with answers is accuracy. Regardless of how well the process of a reference query is handled, a wrong answer will almost always result in a poor satisfaction rating by the patron. For precisely this reason, accuracy has been frequently studied as a measure of reference success. Hernon and McClure's 55% rule (1985) and other similar unobtrusive studies are based solely upon measuring the value of accuracy. These studies have been heavily criticized by reference librarians precisely because accuracy is not the only value associated with answers to patron queries.

Thoroughness is another such factor. An answer may be accurate but incomplete. When working on a question, a reference librarian often keeps searching for a better answer even after finding one or more possible solutions. By consulting additional sources, the librarian is able to verify the accuracy of the initial response and to determine if the answer may have changed in some respect. For example, a question with an apparently simple answer (Who is the U.S. ambassador to the United Nations?) may be found in any of a number of different directories. Depending on when that specific question is asked and when the source was written, the answer may have changed. To be certain that a published answer is accurate, the librarian might check the answer found in a government directory with other sources such as supplements to the original source, Facts on File, or the Internet. Finding an answer is often simple, but verifying that it is completely correct can be very difficult. Because a reference librarian usually cannot continue work on one single question indefinitely (what with other patrons or priorities to attend to), a good professional librarian has an intuitive feel for when to stop working on a question.

Timeliness is another important value. Often patrons need an answer quickly and the reference librarian does not have sufficient time to conduct a thorough investigation of the topic. Some patrons (such as those working on genealogy) may be willing to wait days, weeks, or even months for an answer. Others (such as those who need data for a presentation, paper, or meeting) have a definite time limit, and information delivered after that limit is useless. The first patron wants accurate information regardless of the wait. The second will accept a quick answer even if it is not thoroughly researched. Accuracy and timeliness are both valued by these patrons, but in much different ways.

Finally, reference librarians value authority when answering questions. During the time when most answers came from published reference sources, the authority of the answer was most frequently reflected in the identities of author and publisher. For example, an answer from a newspaper citing an unnamed source was not considered as authoritative as an answer written by a university professor and included in an encyclopedia published by Macmillan. Similarly, an answer found in the Statistical Abstract had the authority of the United States Census Bureau behind it, whereas data included without attribution in a magazine article had only the validity of the article's author. Reference librarians learned which sources and publishers were most likely to produce authoritative material and tended to use those sources to answer patron queries. With the advent of the Internet, it has become very easy to find unauthoritative sources, but finding an answer that is believable can be quite difficult. The value that sets reference librarians apart from other services that answer questions is their dedication to finding authoritative information.

The third function of the reference librarian is what we now call Reader's Advisory. Public libraries offer reader's advisory services in order to guide patrons to books that they might like to read. Reader's advisory services in public libraries are most commonly associated with fiction collections, where patrons read one author or genre and come to the librarian in order to find similar works. For example, a patron who has read works by the author Rita Mae Brown might be referred by the librarian to Lillian Jackson Braun, since both authors involve cats in their mystery novels. Academic librarians do a great deal of reader's advisory service as well, but they do not usually call it by that name. In academic libraries, reader's advisory often takes the form of referring students to sources or databases relevant to their research. In academic libraries, the question "Where should I look for information on my topic?" is really a request for reader's advisory service. The student is seeking not an answer to a specific factual question, but sources that can lead to further information on the topic. As libraries have come to rely more upon electronic databases, the number of reader's advisory requests has risen dramatically.

The key value involved in reader's advisory is knowledge: knowledge of the reader; of the literature; and of the collection. A librarian who is good at this service must understand the needs of each patron. The librarian must understand the general interests of the community served by the library and must also relate to the specific needs of each patron. Reading level, language skills, and educational background all factor into good reader's advisory service. The librarian must also be aware of the universe of sources from which to recommend selections. Most reference librarians are also involved in collection development, and the subject knowledge gained from building the library collection transfers over to reader's advisory. Finally, the librarian must know the local collection. Suggesting sources that the reader cannot use is not only unhelpful, it is downright irritating. For example, it does little good to suggest that a patron search a database that the library does not offer. It aggravates the user, and it reflects poorly on the library, since the student will inevitably wonder why the library lacks the recommended source. Reader's advisory services depend on librarians who value and understand the relationship of the reader, the literature of the discipline, and the local collection.

The final function of Green's reference librarian is to promote the library within the community. While this function is certainly self-serving, it ensures that the other functions will continue to be available. Without support from the community, the library itself would fail. The implicit assumption is that the librarian is promoting the library on a one-on-one individual basis. According to Green (1876), "One of the best means of making a library popular is to mingle freely with its users and help them in every way" (p. 78). Reference is not something that is packaged and marketed to the masses; rather, it is a service that treats every library patron as an individual with unique needs. The library is promoted because each patron feels that the librarian is working specifically for him or her. For many community members, the library is one of the very few social institutions where they receive such service. Access and individualization are highly valued aspects of reference service.

Although much has changed in society over the past 125 years, these basic values remain at the foundation of reference service. Today's technology allows us to offer information services to anyone in anywhere. We now have more information available than at any other time in human history. Despite these advances, the basic values of reference service are virtually unchanged. The tools with which we work are very different from those in Green's time, but the process in which we are engaged is very similar.

CONFLICTING VALUES AND MODELS OF SERVICE

While the individual values upon which reference service is based remain unchanged, the emphasis among those values can vary greatly. Different models of reference service exist because each model emphasizes a different set of basic values. When a model that is based upon one value is evaluated on the basis of another, misunderstanding and conflict arise. Even within a single function of the reference process, such as answering patron questions, values of accuracy, timeliness, thoroughness, and authority are frequently in direct conflict with each other. To guarantee that an answer is thorough and accurate, the librarian may have to compromise on timeliness. For an answer to be timely, the librarian may not be able to guarantee authority or accuracy. The model of reference service chosen by an institution is an indication of which values are considered most important to the library--and to the community that it serves.

Debates on models of reference service are nothing new. Not long after reference service became common in libraries, a conflict developed between the values inherent in the first two functions--offering instruction and answering factual queries. Some librarians emphasized the instructional aspects, whereas others emphasized question answering. Throughout the twentieth century, librarians debated these two models of service.

This debate has been well documented by Rothstein (1955, pp. 75-79) and is summarized in the Encyclopedia of Library and Information Science (Galvin, 1978). The model of service that emphasizes education over information has been termed the "conservative" or "minimum" model, with the ultimate goal being to train patrons to use the library independently. The process of finding information is valued above the information itself. This model has long been the predominant approach in school and academic libraries, where the mission of the broader institution is clearly educational. Its many variations culminated in the concept of the "teaching library" (Guskin, Stoffle, and Boisse, 1979), in which bibliographic instruction becomes the central role of the entire academic library.

The "liberal" or "maximum" model emphasizes information over process. Under this model, when a patron asks a question, the librarian finds the answer and may even deliver it to the patron. The librarian does not attempt to educate the patron in the process, but puts all of the effort into finding accurate and authoritative information. Proponents of this model believe that libraries are highly complex organizations and that attempting to train all patrons in the skills necessary to navigate this organization is impossible. Rather than spending time and effort to create independent library users, the librarian serves as a subject expert who is to find and analyze the information in the collection. Used heavily in corporate and government libraries (where the librarian is viewed more as a research colleague than as a teacher), this model places a high value on accuracy, thoroughness, timeliness, and authority.

Obviously, neither of these extremes is the "right" model for all reference departments. In fact, most libraries offer services that fall between these two extremes. Reference librarians tend to value both the teaching aspects of reference work and their ability to answer specific questions. Both models can even exist simultaneously within the same institution. Some large universities have general, undergraduate-oriented libraries that emphasize the teaching side of reference work, and small, discipline-oriented departmental libraries that emphasize factual knowledge. Even within the same reference department, librarians may practice one model with one group of users and another model with a different group. In some libraries, it is standard policy to show students how to conduct research (the conservative/minimum model) while librarians will actually conduct extensive searches for faculty (the liberal/maximum model). Departmental guidelines on how much work to do for any given patron or how much time to spend with a single user stand as attempts by reference librarians to regulate the relative strengths of these two opposing viewpoints.

TECHNOLOGY AND CHANGE

While technology is not the only factor that has changed in libraries, it is most certainly a driving force. In an interview published shortly before his death in 1985, Hugh Atkinson (Alley, 1985) predicted that technology would spark a revolution in reference service by the end of the twentieth century. His prediction came true, as reference librarians explored and adopted an ever-growing number of innovations for providing information content and delivery. Before the mid-1980s, information technology had been applied primarily to the technical and organizational sides of the library, mostly to develop large centralized catalogs and databases. As the power of computing became more decentralized and universally available, librarians used it to enhance the service models of the past.

In most reference departments, the first encounter with information technology was through mediated online database services such as Dialog and BRS. Such services became widely available in the mid-1970s through the mid-1980s. Since most libraries were unable to absorb the costs associated with this kind of database searching, patrons usually paid for some or all of the direct costs. Although the librarian worked with the patron to develop search strategies and review results, it was the librarian who understood the process and who had access to the technology. To reduce costs, the librarian actually did the work while the patron observed the process. In many ways, the economics of mediated searching dictated that librarians follow the liberal/maximum model of reference service.

From the mid-1980s to the mid-1990s, there followed a period of unmediated searching of some of those same databases on CD-ROM. Librarians purchased databases that members of their community could use; while the librarian still controlled the possible sources, it was up to the patron to perform the search and evaluate the results. Since the cost of the database did not depend on usage, a library could purchase the database and make it available to all community members at no charge. Not surprisingly, the total number of database searches rose dramatically. The librarian's role shifted from that of performing the search to that of teaching patrons how to do their own searches. The model of service shifted along with this change in technology, from the liberal/maximum model toward the conservative/minimum model.

What neither Atkinson nor any other visionary could predict was that information technology would be directly adopted by library users in their homes and offices. With the advent of the Internet in the 1990s, the public gained the ability to find information virtually anywhere. Information that once would have required retrieval by a reference librarian was now in the hands of anyone with a computer and a phone line.

As information technology became available in the household, new predictions began to emerge. Some said that the reference librarian--indeed, the entire library--would no longer be needed. Others felt that the new technology required a new means of delivering reference services, especially for those patrons who were not physically present in the library. These predictions often cited the declining statistics of reference use in libraries of all kinds in all areas of the nation. A variety of new models of reference and library service were proposed and touted as the way of the future. In addition, a number of commercial information services and dot.com companies attempted to take on some roles formerly reserved for the reference librarians. In a matter of a few short years, there arose a feeling that reference librarians had lost control over their chosen profession (Tyckoson, 1999).

NEW MODELS, OLD VALUES

The key to understanding and evaluating the new models of reference service is to examine them in light of the traditional values. The new models of reference service do not dismiss the traditional values; they simply emphasize different combinations or aspects of those values. The most effective way of evaluating a new model of reference is to compare the values associated with that model to those of the community that the model is designed to serve. Some examples of this analysis follow.

Traditional Reference Model

The traditional reference service is closest to the liberal/maximum model. In the traditional model, the reference librarian works at a desk or counter and handles all types of queries, from directional questions to in-depth research. The role of the librarian is primarily to answer patron questions and secondarily to provide reader's advisory services. Patrons receive individual attention and service, although they may have to wait in line when the library is busy. The traditional model of reference service emphasizes the values of personal service, access to information, knowledge of the discipline and collections, accuracy, and timeliness. Traditional service deemphasizes the values of instruction and thoroughness.

The Teaching-Library Model

The teaching-library model represents the extreme opposite of the traditional model and is a primary example of the conservative/minimum approach. The role of the librarian is not to answer questions, but to provide instruction in the research process. Reference librarians working under this model often work with patrons in groups in classroom settings rather than individually at a reference desk. Patrons do not approach the librarian when they have an information need, but are introduced to the librarian before such a need arises. The teaching-library model highly values all of the educational aspects of librarianship, including factual knowledge, the research process, and critical thinking. This model also values authority and thoroughness over accuracy and timeliness. Personal service is not as important as in the traditional model. Knowledge of the discipline and collection is also considered subordinate to knowledge of the research process.

Tiered Reference Services

In tiered reference services, different librarians or staff answer different kinds of questions. Initially made popular as the Brandeis model of reference service, tiered reference has subsequently developed several different variations. Their common feature is that support staff or students answer the majority of the simple queries and that reference librarians are reserved for answering in-depth research questions. In some cases patrons are required to make appointments for research consultations, whereas in others they are simply referred to a different desk or area of the library. One of the primary goals of tiered-reference service is to allow reference librarians to make better use of their subject and research skills.

Tiered-reference service places very high value on a librarian's knowledge of the discipline and of the collection. Since its primary goal is to allow librarians to spend more time with patrons, this model also values accuracy, authority, and thoroughness. Tiered reference also values personal assistance, but places less value on access and timeliness. It is more difficult to reach a librarian, so patrons who need answers quickly may not take advantage of the benefits. Tiered service does not place a high value on the educational aspects of reference service, although those patrons who receive in-depth service may also receive personal instruction in the research process.

Virtual Reference Services

One of the more recent models of reference service is the virtual reference desk or online reference service. These services are designed to help patrons using technology--especially when those patrons are not physically within the library. Using email, chat, and call-center-based software, librarians help patrons in real time over computer networks. Virtual reference is advertised as being able to reach patrons at any time of day or night at any location in the world. As such, virtual reference service places the highest value on access, accuracy, and timeliness. Virtual reference also values personal service, although without having the librarian and patron meet in person. Values associated with the educational and reader's advisory aspects of reference service are not as important in virtual reference service, since the focus is usually on providing answers rather than instructing users. Interestingly, some commercial information services that provide virtual reference service confuse reader's advisory with answers to queries. Instead of providing information in response to a patron question, they recommend sources (usually web pages). These services often end up disappointing and confusing their customers, who generally expect the quality of service provided by library reference departments.

THE "RIGHT" MODEL OF REFERENCE SERVICE

Each model mentioned above is based upon the same set of core service values, but each emphasizes a different subset of those values. For any model to be successful, there is one value that must be emphasized above all others: knowledge of the community that the library serves. A library staff that understands its users and their needs will develop a service model that best supports those needs. Models that work well in one library will not necessarily apply to another that serves a different type of community. For example, a library serving a high school in which every student is given a laptop and an Internet account may wish to set up a virtual reference desk to support homework and after-school programs. However, this model would not be appropriate for a public library serving a low-income community where few families have computers at home. Similarly, a teaching-library model may work very well in an undergraduate liberal arts environment, but it would be totally out of place for a corporate library.

No single model of reference service applies to all libraries. Many new models have been proposed over the past two decades, and many others will be proposed in the future. These models receive attention in the professional literature precisely because they are innovative. However, the fact that they receive attention does not imply that they must be adopted. The needs of the community must remain foremost in the minds of reference librarians, who should not feel pressured to change models of service simply for the sake of change. Just as CD-ROM replaced mediated searching and the Internet is replacing CD-ROM, good new ideas will continue to be incorporated into the mix of library services. Traditional reference service remains the predominant model in libraries today not because reference librarians lack initiative or are stubborn or resist change, but simply because that model still meets the needs of many communities that libraries serve.

Each model of reference service should be measured against its ability to support the values upon which it is based. A model that emphasizes instruction should not be criticized for failing to provide accurate, thorough, and timely answers to patrons' questions. Conversely, a model that values factual information should not be criticized for failing to teach the research process. Most important, any model should be measured against community values. As new models are proposed, librarians should study those models to determine if they offer advantages over existing systems. Over time, communities change, libraries change, and models of reference service change. Determining the best model of service for an individual library should be a process of evolution, not revolution.

REFERENCES

Alley, B. (1985). A conversation with Hugh Atkinson. Technicalities, 5(4), 3-7.

Campbell, J. (1992). Shaking the conceptual foundations of reference: A perspective. Reference Services Review, 20(4), 29-36.

Coffman, S. (1999). Reference as others do it. American Libraries, 30(5), 54-56.

Galvin, T. (1978). Reference services and libraries. In Kent, A., Lancour, H., & Daily, J. E. (Eds.), Encyclopedia of library and information science. Volume 25. (pp. 210-226). New York: Marcel Dekker.

Garrison, D. (1979). Apostles of culture: The public librarian and American society, 1876-1920. New York: Free Press.

Green, S. W. (1876). Personal relations between librarians and readers. American Library Journal, 1(2-3), 74-81.

Guskin, A. E.; Stoffle, C.J.; & Boisse,J. A. (1979). The academic library as a teaching library: a role for the 1980s. Library Trends, 28(2), 281-296.

Hernon, P., & McClure, C. R. (1985). Unobtrusive reference testing: the 55% rule. Library Journal, 111(7), 37-41.

Massey-Burzio, V. (1992). Reference encounters of a different kind. Journal of Academic Librarianship, 18(5), 276-286.

Rettig, J. R. (1992). Rethinking reference and adult services. RQ, 31(4), 463-466.

Rothstein, S. (1953). The development of the concept of reference service in American libraries, 1850-1900. Library Quarterly, 23(1), 1-15.

Rothstein, S. (1955). The development of reference services through academic traditions, public library practice, and special librarianship. Chicago: Association of College and Research Libraries. [Reprinted by Gregg Press, 1972.]

Tyckoson, D. (1997). What we do: reaffirming the founding principles of reference service. Reference Librarian, 59, 3-13.

Tyckoson, D. (1999). What's right with reference. American Libraries, 30(5), 57-63.

David A. Tyckoson, Head of Reference, Henry Madden Library, California State University, 5200 N. Barton Ave., Fresno, CA 93740

DAVID A. TYCKOSON is Head of the Reference Department of the Henry Madden Library at California State University, Fresno. Prior to coming to California in 1997, he held positions at SUNY/Albany, Iowa State University, and Miami (Ohio) University. His article in the May 1999 American Libraries, "What's Right with Reference," presents his view on the state of reference services today.