Notes from Circulation Services Session


Facilitator: Trevor Dawes, Head, Circulation and Support Services Department, Access Services Division, Columbia University Libraries

Recorder: Karen Green, Supervisor, Milstein & Reserves, Butler Library: Columbia University Libraries

Discussion Topics:

  • Recruitment & retention of student assistants
  • Performance standards and measures
  • E-services - how electronic are they?

I. Student Staffing

[What problems do you encounter in staffing desks with students?]

DARTMOUTH: Because we are on a quarter system, we need to staff our desks year-round. This results in constant coming and going, and constant retraining.

RUTGERS: Our worst scheduling problems come during exam sessions and Summer Session. Our best bet is with international students, who tend to stay on campus during the summer.

DARTMOUTH: Our students tend to be nocturnal! We can't count on them during the morning hours. We have two night supervisors who stay until midnight, covering the five service points that we maintain.

We coordinate all student staff under one supervisor: they are then divided into four teams, each team having a student supervisor. The student supervisor handles payroll, training, and scheduling, and also deal with no-shows, which has been effective. Management works directly with the student supervisors. The whole student supervisor system, which has been in place for eight months now, has been very successful.

YALE: There has been a noticeable drop in student interest in library jobs. Why? The economy is good, there is parental pressure to study, and they have extensive extracurricular interests. And we don't pay enough money to be competitive.

[This elicited murmurs of acknowledgement throughout the group.]

YALE (but different speaker): Money is not the problem we think it is. There are other hourly jobs available to students that pay even less.

PRINCETON: The new campus center at Princeton has offered jobs at significantly higher pay than other university venues offer, and that has raised the bar. The library has had to raise pay scales commensurately. Students' primary concerns in whether to take a job or not seem not to be money as much as: How many hours a week? What opportunities are there to study? Will I be working with friends?

NYU: At the ILL desk, there is no studying allowed at all, but there is a very high retention rate among student staff. That is because ILL recruits from the campus "stars."

There are also perceived benefits to working the ILL job, which requires a lot of bibliographic searching. These benefits include: learning to use the library better - the layout, the LC call-number system, the OPAC, advanced searching skills. These are used as recruitment points in the ad used during hiring season. The ad is placed on the Web before the Financial Aid letters are sent out, so that students unsure of whether they will have aid or not will see it when they begin their search for income alternatives. The Financial Aid letter also directs students to the web-site where the ad is located.

YALE: The library's Human Resource office, which organizes the Job Fair and provides a staff member for hiring, has proved ineffective.

PRINCETON: There are seven levels of student jobs, plus a managerial coordinator's position. The Human Resources office did a study, assessing the student job descriptions with comparable jobs elsewhere, and they found that Princeton's student wages were low.

COLUMBIA: The problem with hiring practices like the Job Fair is that it is inevitably scheduled too late to be useful: the desks are already up to full operating speed when the fair occurs. Another alternative is posting flyers, but that is labor-intensive and results in a stream of walk-in candidates whom staff must be on site to deal with.

[A question was asked about the job levels at Princeton.]

PRINCETON: The lowest paying job is a "sitting" job, which deals mostly with checking-in and checking-out, and emptying the bookdrop. These students are guaranteed 50% of their time can be used for studying, and they are paid $5.25/hour.

Another level is the "working" level, where students also handle checking in and but, but which includes shelving, fine settlement, shelf-reading, e-reserves scanning, OPAC searches, etc. This level pays $8.25/hour.

The highest level is the managers, who do the student schedule, handle claim returns, cover the desk in the event of an emergency, handle the student payroll and training, and are the liaison between librarians and the students. This job pays $10/hour.

COLUMBIA HEALTH SCIENCES: Given our student base, it's unlikely that we can get any student help in staffing our access desks. Where do other libraries turn under such circumstances?

CHICAGO: The University has a "temp pool" of 6-month employees. The library can hire these temps on a quarterly basis. It's true that a lot of training is involved, but it makes it very easy to get rid of troublesome or incompetent employees: you simply don't renew the contract!

YALE: We also have a stratified student job scale, paying a premium for post-midnight hours. This still isn't enough to deal with shortages, though, so we turn to hiring part-time locals. This can significantly raise the training factor.

NYU: We had a hard time getting stack staff from the students, so we recruited students from high schools and other local universities. There is a supervisor for each floor on the library, so control is tight. High school students receive around $5.25 to start, with a $0.50 increase each semester they remain. A nice side-benefit for hiring high school students is that they are more likely to go on to college after finishing their high school career.

II. Performance Standards & Measures

[How do you measure performance?]

YALE: This is actually a very big project. We are at 101% of shelving capacity, which makes our job all the more difficult. The process of shelving was broken down into segments, so we could see how much time was being devoted to each aspect of the task.

COLUMBIA - LEHMAN: Our statistics also include the time it takes to sort and prep the book-trucks. This has allowed us to schedule more effectively. Circulation staff downtime is often used to pre-sort book trucks from the return bins.

[This last remark elicited an extremely positive response from attendees!]

DARTMOUTH: We keep comprehensive statistics and have found that they are time-consuming and not always useful.

It seems that it would make sense for all libraries to count the same things in our statistics, which would allow us to make useful comparisons.

RUTGERS: Our student supervisors also train the new full-timers, using a training checklist: this guarantees service consistency.

MIT: A lot of our staff complain about keeping performance statistics. They say that logging statistics takes a lot of time, and they don't see what we do with the results.

YALE: After getting the same complaint, we modified the statistic form to include the time it took to log the statistics. We then gave the results at a weekly staff meeting. When the statistics showed certain changes were desirable, we implemented them

An example of the changes in service: We had a "Request Card" program for On order/In process items. There was an inconsistent response time for these cards. We tracked the cards that were returned (i.e., if no book) and tracked unretrieved books. The result was that we found that the "Request Card" program was seeing an 80-90% fill rate in 2-3 days, providing proof that the program was more successful than had been thought.

COLUMBIA: Is anyone using the data that is or can be generated by automated on-line systems?

CHICAGO: We use Horizon. It scans for browses, and breaks the results down into call-number categories. We used to count browses by hand.

COLUMBIA: We use the on-line statistics to check hand-counted statistics for quality control.

COLUMBIA - HEALTH SCIENCES: Student Assistants are not allowed to shelve books. We use the Shelve-It program to test shelving skills. A staffer who passes it five times is allowed to shelve.

CHICAGO: We have about 60 students doing our shelving, and four full-timers. The full-timers conduct random floor checks. They also conduct shelf checks: they will note down 5 call-numbers at random from a book-truck and then go to the shelf later to see if the book has been shelved correctly. The students are not given specific ranges that they are responsible for - they prefer the flexibility of shelving different places at different times.

YALE: All students must come in one hour per week to shelve. Their work can be checked by a log that notes call-number range, time, and signature.

NYU: We log by the first and last call numbers shelved. A full-timer will follow student staff on the same day and check random call numbers.

COLUMBIA: We use the "Claim Return" statistics as a measure of quality control. Only one staff member handles these claims, and can come back to the staff and say, "This call-number was not checked in!"

[Any other thoughts on performance standards?]

UPENN: We've launched a customer service initiative. Every staff member should be aware of smiles and eye contact. Training was provided for helping patrons with special needs. The customer service training was customized for individual service units. There's no denying, though, that the customer service initiative was first met with staff cynicism.

COLUMBIA: It's not that standards don't exist; it's that they're so difficult to measure.

RUTGERS: Sometimes it's more successful merely to have certain service expectations in place than to have qualitative standards. An expectation, perhaps, such as, "What is the satisfactory outcome of this service transaction?"

CHICAGO: We have had customer service sessions for the student workers. Staff expectations are often much different from student expectations: faculty, for example, are looking for a more formal interaction, while students at the desks are often expecting a more informal interaction because they are being waited on by their friends. Full-timers as a rule treat patrons more formally.

RUTGERS: We've been trying to make a turnaround in the "flaming emails" we receive.

COLUMBIA: One way to measure accuracy - when does the student staff say, "I don't know"? How often are they willing to refer the things they're unsure of to the full-timers?

CHICAGO: One way to increase accuracy is to have staff cross-train with other departments. This way, for example, the circulation staff sees the reference staff in action, and can better understand when they should refer patrons there. It also gives an added dimension to the needs of the patron, who is seen as more than just a circulation transaction.

NYU: We used to have an anonymous generic account for circulation sign-ons, but now each staff-member, student or full-time, has an individual password-protected account. This allows management to track problem to specific staffer.

MIT: At the circulation desk we have photos of each staff member with their names on them, so that patrons can pinpoint point-of-contact staff. This eliminates the problem of patrons unable to identify who gave them misinformation, or checked something out incorrectly.

CHICAGO: Has anyone experimented with self-checkout machines?

CORNELL: We had self-checkout machines, but we had to stop using them once we got our NOTIS update, because they no longer worked.

III. Electronic Services

[What electronic services do you currently use? Examples: E-renewals, online recalls/holds, online fine payments, checking circulation records, etc.]

NYU: No recalls or holds are possible in our system. We send out no email notices.

CORNELL: We have Voyager, and we handle everything electronically. This has proved tremendously popular. When the new version of Voyager came out, it compromised our ability to send electronic notices and we had to return to sending out all notices on paper. This got possibly the highest negative reaction of anything we'd done! We found that we were spending up to eight hours a day folding and stuffing these notices.

We also have Iliad for ILL and document delivery from Offsite. This has also proved extremely popular. We have also found that we can better deploy our staff now that their time isn't taken up with paper notices.

UPENN: We also have Voyager, and we turn on the various electronic services as the patron demand initiates them. We were able to eliminate one clerical position already, not to mention the savings on stationery and postage.

Our system allows us to verify correct patron email addresses via bouncebacks, which go into a file for paper notices. We also up date the address database daily from the registrar's files.

CORNELL: We also coordinate our database with files from the registrar and from Human Resources. We've found that while there is some bounceback with bad email addresses, the return rate is lower than it was for print.

We have eliminated the option to receive print notices.

MIT: We have Jake Advance and are going to Ex Libris. We provide a 3-day courtesy notice: three days before books are due patrons receive a notice reminding them of the due dates. This was written by the Systems Office. It has reduced overdue notices, and gets books back on the shelves.

COLUMBIA: One worry we've had is whether electronic renewals cause us to relinquish our control over the collection.

CHICAGO: We use Epixtech Horizon. We limit our renewals as a way to guarantee that the patron still actually has the book and hasn't lost it, and to prevent books for staying out for one student's entire academic career. Electronic renewals are tricky for non-students because they need a PIN. Faculty has no due date associated with their loans.

UPENN MED: We have a 2-renewal limit, and all our loans are for 28 days. We allow 3rd-renewals on a case-by-case basis. This prevents students from checking out their textbooks instead of buying them.

RUTGERS: We have SIRSI and Unicorn. We allow for unlimited renewals, but if certain titles seem to be staying out too long, we'll assess in a one-on-one discussion. Our system requires a 4-digit PIN for electronic services such as renewals, ILL, Offsite recall - many of these services only seem electronic, but are actually staff-mediated.

Email notices are preferred but not the only option. Patrons can provide their email for electronic notices if they like. Currently, we send out about 50-50 email and paper notices.

UPENN: We've been piloting a program called BorrowDirect with Columbia and Yale. It makes available books that are not in the collection, that are checked out or on reserve. There's about a four-day turnaround for delivery to the patron. It's faster than ILL, and there's already a good bib record available.

COLUMBIA: We're interested in expanding BorrowDirect to more libraries, but at the moment the software isn't stable enough to incorporate another large OPAC.

CHICAGO: Returning to the question of paper versus email notices: we've found that our patrons are happier with email notices because that's the form of communication that they've become most comfortable with. Plus, they can settle problems without ever having to come in and deal with someone in person. For the staff, it's faster to deal with an email response because they're not put on the spot. They can wait to answer until they've done all the investigating they need to do.

COLUMBIA: We have NOTIS and it requires a lot of staff mediation for electronic services, although that is invisible to the patrons. What we've seen as a result is that often the staff handling the electronic queries are not the same staff that were previously dealing with the public, so there are some potential inconsistencies in how such transactions are conducted.

CORNELL: We have Voyager, and also have unlimited renewals. We also deal with high-repetition renewals on a case-by-case basis, because they represent such a small percentage of our whole. This way we don't have to have a one-size-fits-all policy set up just to address the 1% of problem cases.

COLUMBIA: One drawback of electronic fines notices is that they are sent out automatically at the moment that the fine is incurred. Sometimes this fine is in error, and is corrected within a short period of time, but the patron is still going to come to us in anger at an unjust fine.