Librarian Basics: The Reference Interview

This is the first in a series discussing things that librarians do.
Willen Brown
pointed me to this hilarious video from UT Arlington.

Actually, the other librarian's reference interview isn't the model of
perfection, either, but we'll talk about that.

The purpose of a reference interview is for the librarian to understand
the patron's information need - what information will be useful to them
to resolve a problem or learn about something or whatever.
 When done right, the patron can go from a sort of general
unease (anomalous state of knowledge[1]) to information in hand/on
screen, confident [2], and off to make the world a better place. The librarian and the patron converge on a shared understanding of the problem, that can be translated into a query or a jaunt to the shelf.

 So this is pretty complicated right?  Taylor lays
out the steps to get from this unease to actually putting terms into a
search engine.

1. the actual, but unexpressed need for information (the visceral need);

2. the conscious, within-brain description of the need (the conscious

3. the formal statement of the need (the formalized need);

4. the question as presented to the information system (the compromised
need) [3].

Research has found over time, that librarians only answer about 55% of
the questions correctly, completely, and with a citation [4].
 Part of the reason for this is lack of a decent reference
interview.  Like in the video - when someone asks for China -
(you're hearing it not reading it) - do they want china, China, Chyna?
 What about China?  This is also why a TREC track
with a librarian in the loop beat other systems for interactive
information retrieval.

The state of Maryland requires a 24 hour class for all public library
librarians and library associates who work the reference desk.
 Librarians are subsequently coached and evaluated on modeling
these behaviors the remainder of their time in the system.
 Other libraries don't tend to discuss this so much, I've
found. Here are their steps, which are well proved.

First - be welcoming and open!  Smile!

Second - open questions like: tell me more about what you need?

Third - probing questions, closed questions, repeating or rephrasing
the question to make sure you've got it

Fourth - find the information

End by following up with the customer whether you completely answered
their question, or gave them enough information to get started. This
step is super important - I've gotten all the way to this and had the
customer say, well no, actually, I really wanted x instead of y.

Some things that sort of depend more on the context:

  • provide instruction or give answer?  university
    and school libraries probably do a lot more instruction whereas public
    libraries do more answering but this is all situation specific, too, so
    I'll ask or I'll just talk through all of my search steps and show them
    what I'm doing on the screen to sneak in a little instruction
  • ask why they need the information?  depends on the
  • walk them to the shelf?  give them a map? e-mail
    something?  in Maryland you are required to walk the patron to
    the shelf but in a large academic library, you might not be able to
    truck up to the 6th floor or something.

Online via chat or e-mail adds in some additional problems -
particularly if your patron e-mails you 3 words.   There has
to be some back and forth, and sometimes this back and forth really
helps the patron, too, because it helps them articulate their problem.
 It's easier getting to common ground if you've worked with
that particular patron before, but you must not fall into the trap of
thinking you know what the person wants based on some stereotype or
past interaction.  What's funny, is when you get a reluctant
teen who
starts out all "whatever" and then ends up unloading angst - and the
parent thinks you're a miracle worker because their child hasn't said
three words to them for 2 years :)

[1] Belkin, N. J.
(1980). Anomalous states of knowledge as a basis for information
retrieval. Canadian
Journal of Information Science
, 5, 133-143.

[2] Kuhlthau, C. C. (1991). Inside the search process: Information
seeking from the user's perspective. style="font-style: italic;">Journal of the American Society
for Information Science, 42, 361

[3] Taylor, R. S. (1968). Question-negotiation and information seeking
in libraries. College
& Research Libraries
, 29(3), 178-194. (I actually
just copied that off of my review of this article for href="">EBLIP)

[4] Hernon, P. & McClure,C.R. (1986, April 15) Unobtrusive
reference testing: the 55 percent rule. style="font-style: italic;">Library Journal,111(7),

More like this

Perhaps the favorite part of my job as a corporate librarian is the reference interview. It's how I get the real understanding of what I need to do -- and where I can figure out what the requester already knows, which gives me a good starting place.

Questions come in to us electronically through email or a ticketing system and I always try to do a real-time ref interview before getting started. Doing it thru email or IM just doesn't get the results. Sometimes I have to start work on a research request before being able to conduct the interview because it's a tight turn-around and/or the person is in a distant geography. I feel in those instances that I put in more work with less confidence that I am meeting the person's need.

My manager likes to say that what we do is "improve the question."

And in my situation, it is absolutely valid to ask why they need the info. Is it so they can make a point on a presentation slide? Or will the data be used to make a decision about developing a new product? And who needs the info? A strategic planning team? Or Timmy the Intern finishing up his summer project?