[RSS] [Google]



contact us

Library Philosophy and Practice 2008

ISSN 1522-0222

The Reference Interview: Theories and Practice

Stephanie Willen Brown
Homer Babbidge Library
Collection Services
University of Connecticut
Storrs, CT 06269-2005 USA


This article has also been published in the journalBiblioletra, published by The National and University Library of Kosova.

  • I need some information on breast cancer
  • I'm writing a paper on colonialism in Shakespeare'sThe Tempest
  • Do you haveTimemagazine?
  • How do clouds form? Which clouds produce rain?

These are examples of the questions librarians receive at the reference desk. Sometimes questions are quickly resolved, while other questions seem easy but are actually quite detailed. These questions can be difficult for the new librarian to answer but become easier with time and experience. The reference librarian's fundamental task is to translate the patron's question into one that can be answered with the library's resources. I provide reference service to undergraduate and graduate students at the University of Connecticut Libraries, supporting the department of Communication Sciences; I also teach the introductory reference class to library school students at Simmons Graduate School of Library and Information Science at their South Hadley, Massachusetts campus. Librarians can use several strategies to best help patrons with their questions; I use these in practice and teach them in the classroom. The key element is to know what the patron wants, which can be more difficult than it sounds; the second element is to know what resources the library has and how to use them. Learning how to use the library's resources, and which other resources might be available is very important, but it is beyond the scope of this article. In my introductory reference class, I teach both how to provide good reference service and I also demonstrate over 300 essential reference resources. Students beginning to learn these resources in class, but the learning process continues throughout their career, both because new resources become available, but also because patrons present questions requiring new resources. This article will focus on how to determine what kind of information the patron needs through the reference interview.

About the Patron

First we have to know who "the patron" is. You will probably be familiar with the patrons in your library, but they could vary from library to library, and even within your library. For instance, you could work with a university or graduate student, or you could see a high school student. You could also see a neighbor or a governmental official. This could be in a public library, open to patrons of all kinds, or it could be in a school or university library. In the United States, you could also work in a special library (hospital, newspaper, Fortune 500 corporation). Regardless of patron or library type, you would approach each patron the same way, although you would provide them with different kinds of material. Let's think about the first question in this article: the patron requests "information on breast cancer."

The answer to that question will differ greatly depending upon who the patron is. For instance, if the patron is a biology graduate student studying cancer, his needs would be scholarly, scientific, current, and very detailed, while if the patron were a high school student writing a paper on various types of cancer, she would probably require less detailed material. You would take a completely different approach if the patron were the sister of a survivor, or if the patron herself had just been diagnosed with breast cancer. In that second situation, you would need to be more sensitive than the businesslike approach you might take with students, and you would likely provide less scholarly and more consumer-oriented health materials. Alternatively, you might be assisting a government official who wants to know the latest statistics about the disease: how many people are diagnosed each year, what is the survival rate, and what is the cost of treatment. From this one example, you can see how tricky it might be to help a patron with a question that sounds simple at the start.

The next element to consider is the time of the request in the patron's research process. In the United States, it is common for students to come to the reference desk the day before a paper is due and request detailed information to support their topic (colonialism in Shakespeare's playThe Tempest, for instance); while we can help these students at the last minute, we cannot provide them with the most appropriate material. At the University of Connecticut, we have some full-text databases which are "good enough" and will help students find articles with enough information to write their paper. However, if we were assisting a graduate student writing a dissertation on the same topic, "good enough" would not be sufficient. Ideally, we would work with the graduate student early in his research process and to teach him all of the sources which would be useful for his research, including databases and indexes on his topic, finding print material in our library, and requesting books from other libraries. Public librarians and archivists in the United States often work with patrons researching their family history. The genealogy patron often spends years on her research; she would not be satisfied with a resource that's "good enough."

The Patron's Real Question

We have just discussed two elements to consider about patrons when helping them answer their questions. This provides good background, but we also need to determine the patron's real information need. As we have seen, there are several different kinds of material a patron might want if she asked for "information about breast cancer." Let's consider another situation: a high school student asks the reference librarian at a public library for the most current issue ofTime magazine. The librarian could simply point "over there" and give the patron what he explicitly asked for.

What if the student's underlying question were different? What if the student was doing a paper on the war in Iraq and wanted articles describing the United States ' presence in Iraq, the participation of other countries in the war effort, and current opinion (both domestic and international) of the war? In the student's mind, magazines likeTime publish articles about such issues and would have articles on the topic. But in the library reality, browsing through the current issue ofTime would not necessarily yield useful articles. In fact, browsing through current issues of magazines is not an efficient way of searching. Furthermore, the student's professor might stipulate that students cannot quote from "popular" magazines such asTime; the professor may require students to read articles from scholarly journals such asForeign Policy or the Journal of Conflict Resolution.

Another possibility is that the student remembered hearing about a series of articles inTime (or maybeNewsweek) about exercise and health. The student doesn't want to appear confused and so instead of presenting the unformed query to the librarian, he just asks for the current issue ofTime. In fact, the articles the student heard about were inNewsweek three months ago. Pointing the student to where the current issue ofTime sits may not answer the student's question, especially if the library shelves the most current issue of a magazine in a different location from issues that were published three months ago.

There is, therefore, a fundamental conflict between what the student asks and the material that would really answer his question. The student may not know what his real question is; he may believe that readingTime will help him with his paper. He may know thatTime is not the right source, but thinks that browsing throughTime will magically lead him to other sources that will help him. He may know thatTime is the wrong source but is afraid to appear dumb to the librarian and so asks a question that he knows won't make him feel stupid. Finally, it may be too difficult for the student to articulate precisely what he needs (he may not even know!), so he asks a question which he believes will get him close, and hopes that he can figure it out when he gets there. An analogy for me is buying wine. I don't know how to buy wine: I don't know the technical terms for the taste of wine, I don't know how to select a wine that I will like. I know what I like or don't like - but only when I taste it! I am intimidated when I have to select a wine for dinner. I am reluctant to ask questions because the owner might think I am stupid, and (in a worst-case scenario) will refuse to sell me wine. So instead of asking for help, or learning more about wine (I don't want to learn more; I just want to buy some wine that I like), I just pick a bottle at random and hope. My process of deciding what wine to buy is similar to how patrons feel when they contemplate asking the reference librarian for assistance with their question.

I argue that it is the reference librarian's job to anticipate this potential conflict and ask the student "clarifying questions" about his initial request during a reference interview. The Online Dictionary for Library and Information Science (ODLIS) defines "reference interview" as

"The interpersonal communication that occurs between a reference librarian and a library user to determine the person's specific information need(s), which may turn out to be different than the reference question as initially posed. Because patrons are often reticent, especially in face-to-face interaction, patience and tact may be required on the part of the librarian. A reference interview may occur in person, by telephone, or electronically (usually via e-mail) at the request of the user, but a well-trained reference librarian will sometimes initiate communication if a hesitant user appears to need assistance." (Reitz, 2007)

Conducting a reference interview is the most important work a reference librarian does, because this enables the librarian to match the patron's question to a relevant and useful source.

The Reference Interview

The Reference and User Services Association (RUSA) of the American Library Association has created and updated "Guidelines for Behavioral Performance of Reference and Information Service Providers" which list five important elements, or stages, of the reference interview:

  1. Approachability
  2. Interest
  3. Listening/ Inquiring
  4. Searching
  5. Follow up (RUSA 2004)

Approachability is a matter of good hospitality: you want to make the patron feel that she or he can ask you a question. In the United States, it is important to make eye contact as a sign of respect when the patron approaches. It is also useful to stand up, incline your head toward the patron, or otherwise indicate that you are willing to answer their question. I teach my students to develop standard greetings to start the reference interview such as "Can I help you?", "Do you have any questions", or "Are you finding what you need?" You don't want to appear too busy, or the patron will be afraid to interrupt you. This seems like good common sense, and it is - but there are numerous accounts (published and anecdotal) where patrons felt that the librarian was too busy or otherwise unwilling to answer their question. As librarians, then, we must be aware of the image we present to our patrons at their first encounter with us.

The next element of the reference interview is a continuation of approachability; you want to appearinterested in the question that the patron asks. I often explicitly tell patrons that their question is interesting (this is usually true - as a librarian, I delight in patrons' varied information needs and often find their queries fascinating). For me, this is a way of indicating my interest in their topic, but it can also buy me some time while I think about appropriate resources or questions to ask as we proceed through the reference transaction. For instance, I recently helped an environmental engineering graduate student who was investigating the growth of clouds and ways to predict if the clouds are going to generate precipitation. This topics was both intriguing and challenging, as I am not familiar with the literature in the topic of cloud study. So I indicated to the patron a genuine interest in his topic which offered me a few extra seconds to consider how to proceed with his research.

Finally, at the third stage, we are ready to ask patronclarifying questions which will get us to the heart of his reference need andlisten to his responses. My engineering student needed material on predicting whether clouds were going to generate precipitation. My next task was to find out what kind of information he needed, and what he had already retrieved. If he were at the beginning of his research project, I would perhaps explain library databases to him and starting with a basic search in his topic. If he wanted to fine-tune research that he had already conducted, I would start at a more advanced stage in the process. In this case, he was quite knowledgeable about his topic and had even published an article about it. However, he had done research only in a very broad way, by searching in a general science database and looking at citations at the end of articles he liked. He hadn't done any targeted research in engineering-specific databases.

So I asked him some questions about where he had searched. The RUSA Guidelines give some suggestions on what questions to ask the patron, such as "What have you already found?", "What type of information do you need (books, articles, etc.)?", and "Do you need current or historical information?" My student explained what he had already found, telling me how he conducted his research; he added that he needed current articles on the topic. However, since he had not searched the engineering literature, I suggested he might also need some older articles, as I wasn't convinced that he'd done a thorough search for his topic.

While the task of asking such probing questions appears easy, it is harder than it seems. I provide my library school students with a set of questions applicable to many different reference situations. Some of these include open-ended questions like "What have you done so far?" and "Can you tell me more about that?" Open-ended questions are often better for the patron because they encourage him or her to elaborate in detail rather than providing you with the yes or no answer that you would get from a "closed" question. Getting patrons to elaborate is very helpful when you want to find out more about their topic. When I am unfamiliar with the patron's topic, I often say "How interesting! Can you tell me more about that?" This indicates to patrons not only that I am interested in their topic, but it also encourages them to elaborate on their topic so that I can be most helpful. A related question, particularly useful for students doing technical or scientific research, is "can you spell that?" If a student indicates that she is looking for a particular book, I will ask a closed-ended question like "have you looked in the catalog?" because I want to ascertain whether or not they have searched our catalog before proceeding to search it together. Other than that situation, however, closed questions are not nearly as useful for determining the patron's need as are open-ended questions.

This leads us to RUSA's fourth stage of the reference interview, the search itself. It may be odd to consider that RUSA does not recommend starting the actual search until the fourth step of the process, but if we think back to the "Time magazine" question, we can understand why the literal answer to the question ("We keep the current magazines over there") might not be the best response for the patron. Instead, in stage three, we determine that the patron is really looking for current information about the war in Iraq, and so we can conduct a search in a database which will retrieve articles from general periodicals to answer his real, underlying question. At this stage, I showed my "cloud" patron an essential database for engineering. Before we began the search, I explained the value of this database to him and stressed that as a student of environmental engineering, he needed to be familiar with the material in this database.

The final stage in the RUSA Guidelines for a good reference interview isfollow up. When the reference session is winding down, I often ask the patron "does that make sense?" or "will this be enough to get you started on your paper?" This gives me a sense of whether more instruction is needed or if the patron has enough information. If so, we keep going, and I ask those questions again to determine how confident they feel about what I've just explained to them. When I am finished, however, I always tell them how they can contact me if they have additional questions. It is not sufficient simply to point the patron in the direction of the best resource to answer his question. Instead, we must also offer a way for the patron to get in touch with us. It is often relatively easy for patrons to follow along with what we're showing them when they are with us, but sometimes they get confused when they try to complete the search on their own. Or perhaps they've forgotten an important element of what we explained to them. Or they have been successful with that initial query and now would like to move on to more sophisticated searching. Whatever their later information needs are, it is important for us to provide a mechanism whereby they can contact us with additional questions.

I most frequently tell the patron, "I'm on the reference desk until 3, so come back if you have any more questions." Patrons do stop back at the desk to ask me more questions or to tell me how their library research worked out, and this is useful information for me. I am happy to know when things worked out well, but I'm also willing to clarify how to use a database or where a particular book is located if they have had trouble. If the patron's question is very detailed, or if it is in my area of specialty, I will give them my business card and invite them to email, Instant Message, or otherwise ask additional questions. Students always seem pleased to have my contact information for later, although they rarely contact me.

Another good way to provide reference service is "roving" reference. This process can generate a reference interview - only it is the librarian who approaches the patron rather than the patron who comes to the reference desk. Roving reference is important because patrons can be afraid to approach the librarian at the reference desk, fearing that they are interrupting us; if we approach them when they appear lost in the stacks or are staring quizzically at the computer screen stuck on a page in our online catalog, they are often grateful for the opportunity to ask a question at their point of need. Some opening questions in the first stage would be "are you finding what you need?" (this question is handy for many situations!) or "can I help you find something?" If they say "no, I'm all set," I would move along. But often patrons will respond positively and ask a question, leaving you at RUSA's stage two, indicating interest in their topic; you can move right into stage three and ask probing questions about their topic. At stage four, you can show patrons how to do a search or find material at their workstation, which is often more comfortable for them. Finally, at stage five, give them contact information so they can get in touch with you later if they have additional questions. A nice element of roving reference is that it is easy to follow up with patrons: you can see that the student you worked with an hour ago is still sitting at the computer, and you can ask "how are you doing? Are you finding good material?" The RUSA Guidelines provide useful questions that you can put into your reference toolkit and pull out when needed. Ross (2003) is also a good source of information on the reference interview

Here's a concise view of the five stages of the reference interview, based on my earlier search examples of "information about breast cancer;" this will help us see how these five elements of the reference interview would work in practice. I could start by making an initial assessment as the patron approached me about what level information she might need: if she were 15 years old and I was in a public library, I would assume she wasn't writing a dissertation. I would make eye contact and indicate that I was interested in the topic of "breast cancer". At stage three, I would ask some clarifying questions like "what kind of information are you looking for?" If the patron didn't understand my question, I'd follow up by saying "did you want scholarly material or perhaps something easier to understand?" Regardless of the answer to this question, I'd probably then ask "tell me what kind of information you already have" so that I didn't repeat any steps, and thereby frustrate the patron. Then I would think of some useful databases to search which would meet the patron's need - this might be the online catalog to find books, PubMed to find scholarly medical articles, or a general search database like Academic Search Premier to find broad articles suitable for an undergraduate paper or for the non-medical scholar. I would get the patron a few useful articles and ensure that she knew how to continue her search. Finally, I would end the interview with my contact information if I felt that she might have additional questions; if I were at the reference desk, I'd tell how long I'd be there and encourage her to ask me or other reference librarians if she had additional questions.

Problems with Patron Questions

Bill Katz, long-time library science educator and editor of the reference text bookIntroduction to Reference Work offered a set of rules for librarians trying to help patrons. Two of these rules provide essential clues to providing patrons with a good reference experience:

  1. Be skeptical of the of information the patron presents
  2. Try different search terms (Katz 1992, p. 82)

Human memory can be a tricky thing, so it is important to be wary of specific details that the patron provides. For instance, I once helped a student look for an article that his professor suggested he read. The student said that the article was in an issue in the journalTheBehavioral and Brain Sciences published in the late 1980s. The student asked me where the journal was shelved so he could browse through it. Knowing that the date was likely to be wrong (and hoping that the name of the journal was correct), I told the student to browse many issues, not just the ones in the late 1980s. I was curious to know how this search turned out, so I asked the student to return and tell me where he had actually found the article - and if he didn't find the article, I said that I would help him locate it another way. (Note this example of follow up, as suggested by the RUSA Guidelines.) The name of the journal was correct, but the article appeared in a volume published in early 1990. Date problem occur repeatedly in my work as a reference librarian, and I urge my students to be wary of specific details that patrons provide: start with the information they give you, but be prepared to expand dates, check spelling, or change some words.

Related to this, Katz suggests using different search terms in the search itself, both in the book catalog and in online databases. We need to take the patron's search term and think of synonyms or alternate terms for their topic. I recently helped a student research "gender identity" for a sociology class. We searched that phrase in the sociology-specific database called Sociological Abstracts and retrieved approximately 1,000 articles. This sounds like a lot, but my sense of the topic was that there should be a great deal more. I noticed a subject heading on an especially useful article that said "sex role identity." The patron agreed that the second phrase sounded like a term similar enough to be useful and when we searched it, we retrieved over 2,500 articles - over twice as many as for her original search. In this particular example, the patron would have found material that was "good enough" with her original search, but for her to do a comprehensive search, she needed to modify her search term. I explain this concept to patrons by giving a language analogy: patrons know what their search term means, but the database may have a different terminology for the topic. Once they figure out what the database calls it, they can do more thorough research.

The reference interview is an intermediate step between the patron's question and the ideal resource to answer that question. In their reference textbook "Reference and Information Services in the 21 st Century," Kay Ann Cassell and Uma Hiremath say:

"The reference interview is more an art than a science, an ever-changing practice that requires responsiveness to context rather than the application of a predetermined set of skills. While librarians should learn the elements of a good reference interview, they must also recognize that these steps must be adapted to match each situation." (Cassell and Hiremath, 2006)

This practical comment is both wise and true. In practice, I don't always ask probing questions at the start of the reference interaction; I occasionally take the patron's question at face value. But I must be willing to modify my approach, because some patrons would have been unsatisfied with the answer to their first question. For instance, a patron recently asked if we had a book called "Excel VBA." I began to look for the book he suggested, rather than performing a reference interview. One of our branches had a book with that title, but it was checked out. I then began the interview process by asking the patron if he would be interested in electronic books on the topic. I also suggested he request general books about Microsoft Excel (expanding his original topic) through Interlibrary Loan. This patron would not have discovered these additional resources if I had not suggested them. It is important to adapt the reference process to the situation and to be prepared to ask the patron clarifying questions when they ask for assistance. RUSA's "Guidelines for Behavioral Performance of Reference and Information Service Providers" offer several good suggestions for how to interact with patrons to ensure that they find the most relevant and useful material for their question.

Reference Saves the Time of the Reader

The notion that the reference librarian is an important part of the research process is not a new one, nor is it confined to the United States. Massachusetts librarian Samuel Swett Green (1876) spoke of "Personal Relations Between Librarians and Readers" back in 1876, suggesting that librarians "mingle freely with . users, and help them in every way." Green offers several ways to improve "[p]ersonal intercourse and relations between librarian and readers" such as enthusiasm (what RUSA terms "interest"); he recommends that librarians "[g]ive [inquirers] as much assistance as they need, but try at the same time to teach them to rely upon themselves and become independent." This is consistent with RUSA's fourth guideline, where librarians begin the search and teach the patron to use valuable resources.

S. R. Ranganathan (1931), the "father" of librarianship in India, outline the Five Laws of Library Science :

  1. Books are for use.
  2. Every person his or her book.
  3. Every book its reader.
  4. Save the time of the reader.
  5. The library is a growing organism.

All of Ranganathan's laws are relevant to the reference service, and Law Four is most applicable to the reference interview. The truth of his Laws is especially striking if we substitute "library resources" for "books." Library resources are to be used; we purchase, catalog, and link to our materials so that patrons will use them. There is a resource for every patron, and every resource has a patron who needs it. Frequently the process of finding material is so complex that finding resources to be used - Law Two - is very difficult. This is where reference librarians become important: we closely follow Law Four by saving time of the reader. Patrons often come to the reference desk and say, "I've looked everywhere but can't find what I need!" Generally, they have spent quite a bit of time looking "everywhere," and when we show them the most relevant resource for their research, we can often save them time and teach them to be more efficient searchers. Finally, Ranganathan argued that the library is a growing, changing organism; he wrote these laws 76 years ago, but Law 5 is clearly still true with the advent of the Internet, digital collections, and electronic mail. The reference librarian's role is as important as ever with the proliferation of these digital resources; we must hone our reference interview skills to both save the time of the reader and provide her or him with the most appropriate resources.

Works Cited

Cassell, Kay Ann and Uma Hiremath,Reference and Information Services in the 21st Century : An Introduction. New York : Neal-Schuman Publishers, 2006, p. 15.

Green, Samuel Swett, "Personal Relations Between Librarians and Readers."American Library Journal. 1.1 (1876): 74-81. Reprinted inLibrary Journal 118.11 (1993): pS4-5.

Katz, William A. Introduction to Reference Work. Volume II: Reference Services and Reference Processes. 6th ed. New York : McGraw-Hill, 1992. p. 82.

Ranganathan, Shiyali Ramamrita.The Five Laws of Library Science. London: Edward Goldston. 1931.

Reference and User Services Association. MOUSS Management of Reference Committee (2004). "Guidelines for Behavioral Performance of Reference and Information Service Providers", updated 2004. Accessed from the American Library Association web site on 4/30/07 athttp://www.ala.org/ala/rusa/rusaprotools/referenceguide/guidelinesbehavioral.htm

Reitz, Joan M. "Reference Interview" from the Online Dictionary for Library and Information Science (ODLIS). Accessed from Libraries Unlimited on 5/1/07 at by http://lu.com/odlis/odlis_r.cfm#refinterview.

Ross, Catherine Sheldrick. "The Reference Interview: Why It Needs to Be Used in Every (Well, Almost Every) Reference Transaction."Reference & User Services Quarterly. 43, no. 1 (2003): 38-42.



contact us