Changing the Venues but Not Changing Our Tune: Service Model Transition at a Music and Performing Arts Library

Kate Lambaria and Kirstin Dougan Johnson

Kate Lambaria (lambari1@illinois.edu) is Music and Performing Arts Librarian, University of Illinois at Urbana-Champaign, and Kirstin Dougan Johnson is Head, Music and Performing Arts Library, University of Illinois at Urbana-Champaign

The goal of this quantitative study is to explore the effect of consolidating circulation and reference functions at a single service point in a specialized setting, namely, a large academic music and performing arts library. It analyzes reference transactions before and after the merging of reference and circulation service points. Transaction trends are evaluated based on location within the library, type of question, and question complexity. The authors find that there are significant impacts on reference staff time due to consolidating desks, but the closer proximity led to an increase in activity and allows for referrals that are more successful.

Reference and circulation service models at academic libraries have evolved over time for various reasons—whether in attempts to meet changing patron needs and expectations, or in response to budget or staffing pressures. A frequent impetus for consolidating or closing service points is diminishing traffic, a trend that has been happening in academic libraries for over a decade.1 In best-case scenarios libraries have enough time to thoughtfully plan such a transition and assess the effects of the changes afterwards. This paper looks at changes in patron transaction trends at a large academic music and performing arts branch library before and after it merged two of its three first-floor service points. We hoped this reconfiguration would improve the level of service offered through more effective staffing and referrals. Data analyzed includes transaction frequency and type across the different service points. The goal of the study is not to argue whether or not an academic music library should have a stand-alone reference desk, but to explore the effect of consolidating circulation and reference functions at a single service point in such a specialized setting.

Background

The University Library at the University of Illinois at Urbana-Champaign is the largest public academic library in the country and was at one time composed of close to four dozen separate libraries.2 In 2008 the University Library began a coordinated series of projects known as New Service Models, which over the course of several years involved merging and closing some branches, less from a budget standpoint (although that was a factor) and more from making sure services were as efficient as possible.3 It was unofficially as part of this process that the Music Library (which already held music and dance materials) became the Music and Performing Arts Library (MPAL) by incorporating a portion of the theatre materials from what was then called the English Library. As of 2017, MPAL, which is housed in the Music Building, is now one of only seven subject libraries that exist outside of the Main Library building. MPAL serves the School of Music, the Department of Dance, the Department of Theatre, as well as faculty and students from elsewhere on campus, community patrons, and scholars from all over the world.

MPAL has almost 500,000 items in its physical collections, which include circulating materials and non-circulating Special Collections.4 All media materials are, for the most part, classed in Library of Congress Classification and are housed in two different closed stacks areas (but generally circulate). Circulating books and music scores are classed in either LCC or Dewey Decimal Classification, while non-circulating reference materials (books and scores) are classed in LCC; all of these materials are in open stacks. Although play scripts and journals are classified, they are shelved in open stacks by author/title for plays and by title for journals in their dedicated shelving locations. This duality in classification schemes and assortment of shelving arrangements leads to challenges for patrons and staff.

Service Desk History and Staffing

The service desk configuration has changed only a few times over the Music and Performing Arts Library’s 44-year history in its current space.5 When it originally opened in 1974, it had a two-station circulation desk near the entrance and a reference desk on the first floor, as well as a service desk at the top of stairs on the second floor. In the mid-1990s, the reference desk was relocated and incorporated into a new public computing area closer to the circulation desk, which was re-situated, but still close to the library entrance (see figure 1). While in close proximity to the desks, librarian and other staff offices are separated from the public computer area by a wall, making it difficult to see and hear what was happening at the reference desk. In 2003, the second-floor service desk, which primarily provided circulation functions and access to closed stacks media and special collections, was closed due to budget constraints. It was reopened in late January 2014.

Prior to the desk consolidation, paraprofessional staff and undergraduate student workers operated the two stations at the first-floor circulation desk during all of the library’s open hours. Librarians, select paraprofessional staff, and pre-professional graduate assistants from the graduate school of information sciences worked the reference desk, which was staffed 9 a.m. to 5 p.m. during the week, as well as during some evening hours until these hours were eliminated in 2008. Since its reopening in 2014, the second-floor service point, which is only open in the afternoons Monday through Friday, has been operated by a paraprofessional staff member. With four separate desks and different staffing profiles, staff observed that patrons were often required to repeat their questions multiple times to get the help or items they needed.

Impetus for Change and Planning Process

Starting after the fiscal crash of 2008, the University of Illinois faced ever-increasing budget pressures with late payments from the state of Illinois in some years and then when the state failed entirely to pass a budget for FY16 or FY17 (but it finally did for FY18). Therefore in 2015 the university, including the University Library, started planning ahead for expected loss of state income and likely permanently reduced budgets. The entire library was tasked with finding ways to save money. Many cost savings were realized in central administrative budgets and IT, but branch libraries had harder choices to make. MPAL’s budget is primarily divided between collections and personnel. Collections money is protected from budget cuts and the only part of the personnel budget that can be manipulated easily if there are no anticipated retirements or vacancies in a unit are funds to hire student hourly workers.

In the face of having to reduce its student wage budget, in Fall 2015 librarians in the Music and Performing Arts Library reviewed reference statistics over the preceding several years and noted an overall decrease in interactions with patrons. MPAL lost two music librarian positions and a senior paraprofessional staff position due to budget and other issues between 2007 and 2016, which decreased the amount of time the reference desk is staffed with librarians and paraprofessional personnel. We looked at service models at other libraries in the University Library system and at other music libraries and reviewed the literature on staffing and desk models. We also looked at MPAL’s historic student wage expenditures, which (somewhat surprisingly) had been routinely underspent in recent years. As a result, MPAL proposed that it would agree to a permanent 30% student wage decrease if Library Administration would make a one-time investment in rebuilding the first-floor circulation desk to house a circulation and reference station. The redesigned desk would better accommodate current day service activities and allow a reduction from three staffed stations on the first floor to two. Administration agreed and we began the process of redesigning the desk.

MPAL librarians consulted with Library Facilities and Library IT who determined that we had to work within the constraints of the existing footprint since we did not have funds to remodel the entire shelving and staff area behind the desk. However, we made subtle changes to the layout of the desk to better accommodate reference work, incorporated shelving for ready-reference materials behind the desk, and angled the circulation station slightly to be more welcoming to people entering the library. The ready-reference collection was pared down significantly, and only the most-used items were moved to the new location with the remainder incorporated into the regular reference collection (see figure 2). The reconfigured service desk opened on January 12, 2016.

Staffing Implications, Configuration, and Training

MPAL reference staff have observed that student workers, and at times other staff, do not refer questions to reference staff when appropriate and often try to take a question too far. We hoped that by combining service points, reference staff would be physically situated to more seamlessly insert themselves into reference transactions without causing too much disruption to patrons or forcing them to repeat their questions multiple times. After the desk merger, the primary circulation station continues to be operated by staff and students, and the reference station continues to be staffed by librarians, select paraprofessional staff, and graduate assistants, who perform circulation functions when circulation staff are otherwise occupied with patrons.

Beginning in Fall 2017, MPAL added graduate assistants on the reference desk from 7 to 9 p.m. two nights a week and one two-hour weekend shift so that there would be some additional reference assistance available during more of our open hours. The other two evenings of the week are covered by the paraprofessional members who also serve on the reference team. While this changed the evening and weekend staffing profile, it did not increase the number of staff available, only the availability of staff with reference training. Other evening and weekend hours that the library is open, student workers or paraprofessional employees staff the reference station but don’t provide in-depth reference service. With reference staff now serving as backup for circulation, new and increased training was required. A single online guide for the service desk was designed for quick reference to information that would support both circulation and reference functions and incoming MPAL graduate assistants now receive training on circulation functions on an annual basis.

In January of 2016, around the same time the MPAL service desk was reconfigured, the School of Music completed a building renovation that included the installation of an information desk not far from the library’s entrance. School of Music personnel staff this desk and assist visitors to the building with directions and answer other questions from students and faculty. With the introduction of this new service point, we anticipated a decline in directional and other factual questions about the building and the School of Music. It is possible, however, that other questions asked at the information desk were now being referred to the library, increasing the number of transactions.

As part of our effort to assess the impacts the desk merger and other environmental factors such as the new School of Music information desk had on MPAL’s reference statistics, we examined reference transaction data for the three-year time period before the merger and the two years post-merger. These time periods were selected to give ample data for comparison pre- and post-consolidation. Data points for interaction locations, types of questions, number of questions, and question difficulty level were analyzed to show whether there were any significant changes in patron interactions or traffic patterns.

Literature Review

For decades, librarians have assessed the success of physical, dedicated reference desks and explored new reference and staffing models to meet changing user needs. The goal of this study is to explore the effect of consolidating to a single service desk in a branch library by reviewing transaction data before and after the desk merger, as well as to identify overall transaction trends over time. Therefore literature relating to reference service models, the merging of reference and circulation service points, and analysis of transaction data, in addition to literature specifically about music and the performing arts reference services, is most relevant.

Many authors have explored the various types of models for providing reference services in academic libraries. In a review of the literature regarding current trends in reference services, Bandyopadhyay and Boyd-Burns discussed the transformation of reference services in academic libraries (roving, tiered, and virtual) and the effects of library instruction, supplemental digital resources, and embedded librarianship on reference transactions.6 They also reviewed current trends in reference transactions and current trends in staffing reference services, finding that mediated reference services and human interaction are still important features of quality service. Frederikson and Wilkinson also conducted a literature review exploring the rationale of changing models, noting that these are usually either structural and budgetary or change-management endeavors.7 The benefits of changing service models cited in the literature include increased use of services, clarification of services, decreased frustration for users, and cross-training of staff, while challenges included the unexpected absence of cost savings and the difficulties in assessing success.

In 2000, Jackson surveyed the heads of reference in ARL libraries to determine if services were changing in response to a decline in reference activity and found that while reference services were changing, this change was not nearly as dramatic as many had thought.8 Twelve years later, Wilson surveyed heads of access services departments in 100 ARL libraries to determine the current composition of these departments and whether they were combining circulation and reference.9 Wilson found that “despite calls from and examples in the literature, the idea of combining reference with the Access Services organization is not prevalent in ARL Libraries.”10 Of the 63 responses received, only 8% included reference in their access services departments and only 14% had combined reference and circulation desks, with another 20% expressing a plan to combine desks. The trend of combining these functions at a single service desk was not exemplified in ARL libraries.

Various case studies have been published about merging of service points in libraries, including successful and unsuccessful outcomes. Crane and Pavy at the University of New Orleans described how their reference and circulation (and eventually media) service points were merged into a single service point, using reference librarians at the desk only during peak hours and developing an “on call” system.11 While there were concerns regarding the merger, the authors found many benefits to patrons and staff, including patrons being able to receive assistance from more broadly knowledgeable staff at one location due to cross-training. In addition, staff reported an increased level of job satisfaction due to having a greater variety of tasks and increased staff cohesiveness at the desk. Flanagan and Horowitz wrote about how MIT Libraries successfully integrated service points at one of the five main libraries on campus.12 While they found that there were no quantitative changes after their merger, reference statistics stayed consistent, surveys showed that patrons felt their needs were being met by the new model. Kiesling and Sproules provided a description of the merging of service points at the largest library at the University of Louisville after seeking user feedback through surveys and an advisory board.13 After assessing the results, it was determined that reference staff members should no longer staff the service desk; instead a consultation model was implemented to allow for in-depth research support. Hunter and Anderson described how the University of Missouri-Kansas City successfully combined reference and circulation services in a single service point.14 This helped eliminate previous patron dissatisfaction resulting from having to travel between desks. For patrons experiencing referrals in the new model, “it appears more like getting help from multiple people than being passed back and forth.”15 Abrams outlined University of California San Diego’s process of merging two libraries and an individual service point (including the Music Library) into one space.16

Not all mergers described in the literature were deemed successful. Fritch, Bonella, and Coleman conducted surveys, focus groups, and a quantitative analysis of transaction data to review the desk merge at Kansas State University Libraries.17 They found that, while there were many positive consequences, this desk merge was not an ideal solution, eventually re-separating reference from circulation.

Several studies have assessed transaction data to provide insight into activity, determine appropriate staffing levels, and explore the need for or success of desk mergers. After reviewing data from the Academic Libraries Survey conducted by the National Center for Education Statistics and ARL statistics, Applegate found that while all types of libraries have seen a decline in reference transactions, ARL libraries and other doctoral institutions have seen more of a decline.18 Lenkart and Yu analyzed transactions that occurred at five specialized and two general reference service points.19 Their analysis showed that there was a strong correlation between question difficulty levels (as measured on the READ scale) and the time spent on transactions, which was especially true at the specialized service points. Compared to the main reference desk, these specialized service points also answered more questions with higher difficulty levels.

Ryan describes a project at Stetson University that analyzed transactions collected over several years to determine cost-effective staffing.20 These transactions were coded into directional, look-up, technology support questions, and reference, which were further categorized. Ryan found that only 11.3% of transactions were complicated enough to require librarians, stating that “this study included enough concrete data to prove that many of the transactions are simple enough to be addressed by trained staff, or even by students.”21 Raftus and Collins also assessed reference services at three of the University of Washington campus libraries to determine appropriate staffing.22 Similar to our study, they used Desk Tracker and READ to capture transaction data. They found that over 70% of transactions fell into READ levels 1–2 range that all staff should be able to answer, 27.5% fell into READ 3–4 range (typically ready-reference and basic reference queries), and fewer than 2% fell into READ levels 5–6. As a result of these findings, the libraries adjusted service desk and staffing models. Furthermore, the Engineering Library specifically decided to merge service points but kept librarians and LIS students at a nearby on-call desk because they often received specialized level 3 and 4 questions. Bishop and Bartlett from the University of Kentucky Libraries analyzed 1,852 reference transactions in their library system to better understand the content of transactions to inform staffing and develop a mobile library app.23 Their results showed that most questions were location-specific directional or policy questions, and that only 16.3% were subject-based questions that required professional help. McClure and Bravender analyzed reference activity at Grand Valley State University to determine whether consolidating service points would affect their reference service.24 After conducting a content analysis on recorded activity, they found that librarians answer more reference questions overall because staff at the service desk successfully refer appropriate questions, and that a single-service on-call reference model works for their institution.

Additional studies have examined transactions focusing on the point of first contact. Mosley analyzed 2,000 transactions at Texas A&M University’s Evans Library loss prevention specialist desk, located closest to the door and often the point of first contact for patrons, finding that many interactions still require a basic understanding of the reference interview and details of library information services and resources provided to the user population.25 Location is often a concern when considering merging service points to a single desk.

While authors have explored topics such as embedding music reference services and music virtual reference,26 the contemporary literature directly related to describing and assessing reference services physically occurring in music libraries is limited. In 2004, Hursh surveyed music libraries about whether they had a dedicated reference desk and their reasoning.27 Hursh found that dedicated reference desks are not standard for music libraries, but many respondents expressed an interest in establishing such desks. To address changing patron needs and expectations, Dougan outlined various assessment tools available to music libraries including the READ scale and the Wisconsin Ohio Reference Evaluation Project (WOREP).28 Newcomer and Hursh used Durrance’s “Willingness to Return” methodology and evaluated services through patron surveys and an analysis of reference activity statistics in a music library.29 They found that the existence of a dedicated reference desk did not increase the likelihood of patrons seeking expert assistance elsewhere when the desk was unstaffed. Newcomer and Hursh also found that while librarians at the reference desk were the preferred source of assistance, patrons preferred immediate assistance overall, regardless of source.

Methodology

Many libraries and other service points that make up the University Library at the University of Illinois have used Desk Tracker software since 2008 to track patron interactions. MPAL began tracking patron interactions using Desk Tracker in 2008 and incorporated the READ scale in 2013. READ, the Reference Effort Assessment Data scale, is a six-point scale that helps libraries measure the effort and knowledge required on the part of the library staff and the amount of instruction involved.30 In general, READ levels 1 and 2 are used for questions that staff at every level should be able to answer; anything that requires specific reference training or subject knowledge should be tracked as level 3 or higher, with levels 5 and 6 requiring a significant amount of time and skill (see figure 3).

MPAL locations where data has been tracked over the years include Circulation 1, Circulation 2, Reference Desk, Office, Off Campus, and Second Floor. This information is recorded when staff log in to Desk Tracker. While most of the location options refer to a single desk, Office can be used in any staff office in the Library and Off Campus can be used by any staff member conducting work out of their office. Prior to the merge, Circulation 1 and Circulation 2 were equally responsible for circulation and basic directional questions. Since the merge, the Reference Desk and Circulation 2 are the same workstation, although staff who do not provide reference services will log in to DeskTracker using Circulation 2 (see figures 1 and 2).

MPAL’s current policy states that any person positioned at the primary first-floor circulation station (Circulation 1) should focus on performing circulation functions and answering basic directional questions, and to refer questions requiring reference skills to reference staff when present. The exception to this is if they are one of two paraprofessional staff members who also serve on the reference desk. Therefore there should be very few reference transactions tracked at circulation locations, including Circulation 1, Circulation 2, and the second floor.

Desk Tracker is used for all interactions with the exception of regular circulation transactions, which are materials that patrons bring to the desk, and with all patron types from undergraduates to visiting scholars. The only circulation transactions at MPAL that are tracked using Desk Tracker are those that are considered “paged items,” such as course reserves, closed stacks media, loanable technology, or listening room/carrel keys, which must be retrieved by staff. The online Desk Tracker form tracks location, patron type, status of staff, whether it was a scheduled appointment, mode of communication, time spent, question type, READ level, subject (or academic discipline), whether the question was referred to a specialist, if government information was used, and a description of the question (see figure 4).

The “Question Type” field has to do with the nature of interactions, rather than their subject (music, dance, theatre, etc.). Options included in this field include “Data Assistance” (which should rarely occur at MPAL), “Database/eJournal, SFX Access Problems,” “Directional/Hours,” “Finding Specific Library Materials,” “Library Policies and Services,” “Other,” “Ready Reference,” “Research Assistance,” and “Technical Issues (printers, scanners, software).” While many of these categories are self-explanatory, it is important to point out that the category labeled “Finding Specific Library Materials” includes finding a known item (e.g., a recording of Verdi’s Macbeth or parts for a Beethoven string quartet) in the catalog, but not finding an item on the shelf when a call number is already known, which would instead be “Directional/Hours.” Known-item queries are a common type of transaction in most music libraries in part because of the inherent difficulties in searching for music materials in library catalogs.31

The Library Assessment Coordinator generated reports from Desk Tracker for READ and Question Type information cross-tabulated by location for each calendar year from 2013 to 2017, allowing a comparison of transaction trends overall. It also allowed for a comparison of the time spans from 2013 to 2015 and from 2016 to 2017, before and after the desks merged. We also compared MPAL data to University Library Desk Tracker data overall for the same time period. Data pulled from Desk Tracker is not a complete representation of all transactions occurring at MPAL service points because it doesn’t include standard circulation transactions, but also because it is dependent upon individuals using Desk Tracker regularly and filling the form out completely. Capturing consistent data is difficult due to evolving standards and the complex working environment, but the data extracted is still useful for analysis and can provide insight into trends occurring at the desks.

Findings

The goal of this study is to show any significant changes in patron interactions based on amount or type of questions occurring at specific locations (i.e., reference desk vs. circulation desk). While trends over time are necessary for context, the chief concern is whether there are any changes in patron interaction data that can be tied to the consolidation of the reference desk with the circulation desk.

Trends by Transaction Location

Patron interactions, not including generic circulation transactions, can best be analyzed in aggregate by their location (see table 1). Patron transactions decreased from 2013 to 2014 by 40% across all locations. From 2014 to 2016 transaction levels were quite stable, with a slight dip from 2014 to 2015 and a slight rise from 2015 to 2016. In 2017, transaction levels rose significantly (by 12%) over the previous year.

When looking at activity occurring at specific locations, activity at the first-floor circulation stations dropped 50% in 2014. However, after this large drop, activity at “Circulation 1” increased steadily between 2014 and 2017. Overall, activity at “Reference Desk” dropped by 45% from 2013 to 2015 (by 19% from 2013 to 2014 and 32% from 2014 to 2015), but increased by 75% in 2016, which is not completely unexpected, since post-merger it now serves the dual functions of the second or backup circulation point and the reference desk. This is also why activity logged at “Circulation 2” dropped to almost nothing in 2016, since that station is now usually signed in as “Reference Desk” in Desk Tracker, with the exception of nights and weekends when reference staff do not occupy the station. “Circulation 2” and “Reference Desk” transactions combined decreased from a total of 1,621 transactions in 2015 to 1,590 transactions in 2016, which is only a 2% decrease. In 2017 transactions logged at these two locations totaled 1,816, which is a 14% increase over the previous year. However, the three first-floor stations, when taken in aggregate, decreased by 17% from 2013 to 2017.

Activity tracked in staff offices (“Office”) declined noticeably in 2015 from the previous year (44%), likely in part because one of the two librarians on staff split their time between MPAL and another campus library from August 2015 to May 2016. In addition, one librarian retired in May 2016 and was not replaced until June 2017. However, despite this, “Office” activity increased 16% in 2016 and 48% in 2017. While the stabilization of librarian staffing levels likely played a large part in this increase, it is also possibly due in part to the increase in research appointments made with librarians, as it is more difficult to conduct longer one-on-one sessions at the combined desk.

The second-floor service point was not reopened until January 2014 and is only open Monday–Friday afternoons. The use of Desk Tracker was not implemented immediately, meaning data from 2014 are not necessarily complete. Traffic for 2015 and 2016 was fairly consistent, and the large drop seen in 2017 was primarily due to technical difficulties that prevented the individual from using Desk Tracker at that computer.

Trends by Type of Question

Another informative way to look at overall trends in patron transactions is to view them by type of question asked (see table 2). This analysis is based on a smaller set of data, since “Question Type” wasn’t a required field in the library’s Desk Tracker until July 2015 and was recorded less frequently, especially at the circulation stations, before that time. “Directional/Hours” questions declined only slightly in 2016, which is unexpected given the installation of the new School of Music Information Desk across the building lobby from MPAL. Activity in “Library Policies and Services” declined in 2014 and 2015, but increased significantly in 2016 when the desks merged, and by 2017 was again approaching 2013 levels.

While “Ready Reference” and “Research Assistance” activity declined significantly from 2013 to 2016 (by 80% for “Ready Reference” and 74% for “Research Assistance”), they both increased from 2016 to 2017 (by 7% and 14%, respectively). “Finding Specific Library Materials” decreased by 21% from 2013 to 2014, they were mostly stable between 2014 and 2016, but increased significantly from 2016 to 2017 (13.5%). When looking at the breakdown of MPAL’s transactions as a whole, “Finding Specific Library Materials” interactions account for 27% of MPAL’s transactions from 2013 to 2017, or 40% of all transactions excluding those in the “Other” category (because the majority of “Other” transactions are not true questions but are paged item retrieval).

Transactions logged as “Other” increased significantly in 2015 over the previous year (by 679%), in 2016 (by 102%), and again in 2017 (by 21%). One of the common types of transactions logged as “Other” consist of items paged by staff members. In 2015, 87.5% of transactions tracked as “Other” were also tracked as paged items. This trend continued in 2016 with 89% and in 2017 90.5% of “Other” transactions being paged items. Many of the types of items paged by staff were added to the circulating collection during the time period covered by this study, including iPads, CD/DVD drives, and listening room keys. Since they were added to the catalog in August of 2014, the six listening room keys have circulated more than 3,800 times. The increase in “Other” could also be in part because Question Type wasn’t a required field until July 2015 and individuals may not have had enough training or reminders as to how to designate question types, even though it had been encouraged in use at MPAL since beginning to use Desk Tracker in 2008.

Trends at Specific MPAL Location by Question Type and Year

At a more detailed level, it is possible to examine trends in the types of questions asked at each location from year to year, again with the caveat that Question Type was not always a required field. In order to uncover any changes in interaction patterns at the various locations after the desk merger, the data was analyzed for shifts in patterns from 2013–2015 against data from 2016–2017.

Before the consolidation, the most common type of transaction at the “Circulation 1” location was “Finding Specific Library Materials” followed by “Other,” and “Directional/Hours.” After the merger, “Other” became the most frequent interaction type, followed by “Finding Specific Library Materials” and “Directional/Hours.” At the “Circulation 2” station, the most common types of interactions before the merger were “Finding Specific Library Materials,” followed by “Other,” and “Directional/Hours.” After the merger “Other” and “Finding Specific Library Materials” traded spots.

Prior to the merger, the most common interactions at “Reference Desk” were “Finding Specific Library Materials,” “Technical Issues,” and then “Directional/Hours.” The old reference desk was closest to the printers and scanners, which is why it fielded so many technology-related questions. Post-merger, “Other” moved into first place, followed by “Finding Specific Library Materials” and “Directional/Hours.” For comparison, in 2016 “Finding Specific Library Materials” were down at “Circulation 1” from the previous year by 14% but up at “Reference Desk” by 44%, and in 2017 were up from 2016 by 21.5% at “Circulation 1” and by 29% at “Reference.” In 2017, “Ready Reference” and “Research Assistance” were again up at “Reference Desk” (by 44% and 54%, respectively) and again down at “Circulation 1” (by 20% and 57%). This indicates the likelihood that in addition to an overall upturn in “Finding Specific Library Materials” activity, there is an increase in proper referrals from the circulation station to the reference station.

The most common interactions in staff offices (“Office”) are “Finding Specific Library Materials,” “Research Assistance,” and the “Library Policies and Services.” The most common interactions logged “Off Campus” are “Finding Specific Library Materials,” followed by “Ready Reference.” We did not assess transactions at the second floor service point due to technical issues that caused tracking difficulties at that station.

Trends by READ Scale Levels and Location

Between 2013 to 2017 the percentage of transactions at all MPAL service points with a READ level assigned has not changed dramatically from year to year, with 40%, 50%, 41%, 37%, and 43% of transactions assigned a READ level each year, respectively. By raw totals, the number of transactions assigned a READ value mostly follows the pattern found in all transactions, with the highest numbers in 2013, decreases from 2014 to 2016, and an increase 2017 (see figure 5).

In comparison, the percent of transactions with a READ level assigned occurring at “Reference Desk” has changed significantly (see figure 6). During the pre-merge period, the percentage of transactions with a READ level assigned was 89% in 2013 and dropped slightly to 84% in 2015. After the merge in 2016, the percentage of transactions with a READ level assigned dropped dramatically to 54% but rose slightly in 2017 to 61%. This change may be reflective of the increase in paged item transactions reference staff members are now handling that do not require a READ level (although they could be assigned one) since types of transactions do not necessarily require reference skills, knowledge, or training. In 2017, there were 3,985 transactions across all locations, of which 1,786 occurred at “Reference.” From total transactions, 1,696 were assigned a READ value, 1,089 of which were at “Reference,” which shows that the preponderance of interactions assigned a READ value occur at the reference desk.

Of more importance are any statistically significant changes in the levels of questions asked at “Reference Desk” and at the first-floor circulation points (“Circulation 1” and “Circulation 2”) before and after the consolidation of service desks (see figure 6). While transactions at READ levels 1 and 2 can be answered by all levels of staff, looking more closely at trends can illustrate changes in the work done by reference staff now located at a multi-function service desk. At the first-floor circulation points, READ level 1 decreased by 18% after the desk merger in 2016, while there was a 50% increase in transactions assigned this READ level at “Reference Desk” and another 80.8% increase in 2017. Prior to the merger, READ level 1 at “Reference Desk” was most often used to describe “Directional/Hours” and “Technical Issues” type questions. After the merger, it is most often used for “Other” (mostly paged items), followed by “Directional/Hours” and “Technical Issues” questions. The overall increase in this level of transaction in addition to “Other” being the most common Question Type, demonstrates the significant increase in reference staff supporting circulation functions by paging items.

The number of transactions assigned a READ level 2 did not change dramatically after the desk consolidation. READ level 2 at “Reference Desk” is most often assigned to “Finding Specific Library Materials” and “Technical Issues.” In the case where there is a known title for the item (i.e., a monograph title), “Finding Specific Library Materials” should be assigned a READ level 2. However, a significant number of questions posed in music libraries are finding printed music and recordings of musical works. These “Finding Specific Library Materials” questions are considered a READ level 3 because of the added difficulty and subject knowledge often required. The prevalence of “Finding Specific Library Materials” with a READ level 2 designation and not 3 may be due to the “Quick Guide” for the READ scale on the U of I Desk Tracker form that lists known item searches as READ level 2, which is true in many other campus libraries (see figure 3). Transactions assigned READ levels 3 and 4 require reference knowledge and experience with the reference interview, and as such are the types of transactions that reference staff and not circulation staff should be handling. After the desk consolidation in 2016, transactions assigned READ level 3 at the first-floor circulation stations decreased by 48.4% and transactions assigned READ level 4 decreased by 29.3%. This decrease may mean that circulation staff are less likely to take on reference questions now that reference staff are in closer proximity.

READ level 5 and 6 transactions are more in-depth questions that take extended amounts of time and do not normally resolve at the desk. The decline in these levels of questions occurring across all MPAL locations is to be expected as more students are scheduling time for research consultations when they have difficult questions. While there was an overall decrease in level 5 transactions, part of the reason for this at “Reference Desk” specifically might be because reference staff are now also providing circulation support and do not have the time to dedicate to longer, more in-depth questions at the desk.

MPAL vs. Overall Desk Tracker Participants in the University Library

In comparing MPAL’s data to the Library as a whole, we found that patron interactions at MPAL are more alike than different across the various libraries, with the exception of the “Other” category (see table 3). The use of “Other” at MPAL is high because of the number of items held in closed stacks (media, special collections, loanable technology, etc.) that staff must retrieve for patrons. Excluding the “Other” category, “Finding Specific Library Materials” was the most predominant interaction type across all libraries and “Directional/Hours” was second.

Discussion

It can be difficult to identify clear cause and effects reflected in transaction data due to the subjective manner in which it is collected and the numerous environmental variables at play. The context of the local setting must be considered, as well as more qualitative factors such as impacts on reference staff.

Environmental Impacts

It is not possible to determine what exactly caused the large change in MPAL patron transaction totals from 2013 to 2014, but one possible factor is enrollment decreases in the areas MPAL serves. In 2014 there was a 3.7% decrease enrollment in the School of Music, Department of Dance, and Department of Theatre across all degree types, with a 4.3% decrease in the School of Music. There were increases in each of 2015 and 2016 across all three areas and by 2017, there was a 1.6% overall increase in enrollment across the three areas (1.1% in the School of Music) over 2013 levels.33

Another possible factor in 2014’s large transaction decrease could be the addition of a subscription to the audio streaming tool Naxos Music Library in late July 2013. This acquisition could have caused changes in media circulation and course reserves use patterns (course reserves and media are tracked as paged items in Desk Tracker and would contribute to Desk Tracker totals). Additionally, patrons frequently use the reference desk to ask for help finding call numbers for CDs and course reserves since they are shelved in closed stacks. The trend of collection formats impacting service transaction levels will likely continue in MPAL as additional streaming media subscriptions and other e-resources and e-journals were added in 2018. Furthermore, the technology that connects patrons to the library’s electronic resources (such as the link resolver SFX) has increasingly better functionality, which may decrease patrons’ need to contact library staff for help finding journal articles online or on the shelves.

Although there was a significant drop in “Technical Issues” questions at “Reference Desk” post-merger, the library’s printers were relocated in January 2018. The service desk is now situated between the printers and the public computer and scanner area, with the reference station closest to the printers, so there may be an increase in those questions at “Reference Desk” in the future. Also, a new printing payment system was launched in 2018, which could increase MPAL’s statistics, at least during the initial phase of rollout.

A possible factor in the upsurge in transactions in 2016 and 2017, as previously mentioned, is the circulation of loanable technology and group room keys. We expect to see this trend of increased interactions in this area continue, as the rooms were upgraded in 2018 with new displays. In addition, a new room reservation system was launched, which may increase the visibility and usage of the rooms. Additional loanable technology items are also under consideration. We also anticipate seeing an increase in transactions at the second floor service point now that technology issues have been resolved at that circulation workstation. Had tracking been possible there in 2017, we might have seen an increase, as a marketing campaign was launched in November 2016 to remind people that our LPs circulate even though they are located in the closed stacks behind the second floor service desk.

Impacts on Reference Staff

The Desk Tracker data shows that reference staff have become busier at the desk since the merger, and this could be due to a variety of reasons. Reference staff are answering a broader range of questions now that they are also fulfilling circulation functions, including paged item transactions and other low-level questions, as evidenced by the increase in Level 1 questions at “Reference.” There is a notable increase in “Finding Specific Library Materials,” “Ready Reference,” and “Research Assistance” activity at the Reference station with corresponding decreases at “Circulation 1,” which illustrates that referral practices have improved. This growth in traffic for reference staff has a concrete impact on their abilities to do project work or to offer in-depth help at the desk, both of which must happen in “Offices” and traffic for this location has increased.

This specifically impacts the librarians, who are primarily the ones who track interactions in the “Office” location. Whereas before, queries that arrived via phone and email could be saved to work on while a librarian had a shift at the desk, now much of that work needs to happen in offices because there is not time due to the upturn in activity at the service desk. Additionally, with two librarians since mid-2017, there is more capacity and availability for librarians to work with patrons who have been referred from the service desk in their offices. The introduction of research consultation scheduling software also makes it easier for patrons to arrange a meeting in advance. This raises the question of whether it is worth having “expensive” librarians at the reference desk when graduate assistants or undergraduate student assistants might be trained to do this work and refer questions appropriately. MPAL already makes use of graduate assistants at the reference desk and finds it successful, but the recurring training that would be required for undergraduates due to turnover is currently unrealistic.

Limitations in Data

The primary limitation to this study is the data itself. As others have noted, there will always be difficulties with collecting accurate data at busy service desks.34 We know that staff are not always tracking completely or correctly based on some of the entries we see (Level 6 questions at all, Level 5 anywhere other than “Office,” any use of “Data Assistance,” etc.) and that there will always be some level of human error. A lot of the information tracked is based on personal interpretation, and while efforts are made to realign these interpretations, it will never be perfect. Also, quantitative data only provides part of the picture about what occurs at a library service desk.

Next Steps and Further Research

Since reviewing this data, new training has been implemented regarding use of Desk Tracker to ensure that all individuals are comfortable with using the form and the READ Scale. This was done in a two-part approach, with staff members filling out an anonymous training exercise online and then once the authors had reviewed the results, having a joint meeting with all staff to go over the questions and most appropriate responses. While student staff participated in the online exercise, it was not possible for them to attend the meeting, so a brief summary was shared. Moving forward, we plan to encourage all staff members to use Desk Tracker in their offices to track any patron interactions that might happen there. This makes us consider whether there is other data we should or could be tracking in Desk Tracker. An option for tracking reproduction requests was added in late 2017 at MPAL’s request, since this is something that is a frequent occurrence given its rich and unique Special Collections but that is tracked in no other way. Perhaps purchase requests from patrons, fine and billing questions, and requests to put materials on course reserves should also be tracked through Desk Tracker—essentially any interactions with patrons regarding services. This will require a conversation with the Library Assessment Coordinator about overall practices and philosophy of tracking across the University Library.

Another consideration going forward is the potential to implement alternate staffing models. MPAL is somewhat different than other libraries at the University of Illinois (and perhaps from other music libraries) in that librarians staff its service desk. For example, the Main Library’s Information Desk and Social Science, Health, and Education Library also use this model, while most other branches staff their desks with paraprofessional staff and graduate and/or undergraduate students only. However, removing librarians completely from the MPAL service desk except on an on-call basis would not work very well. This would require each of the two librarians to be on call for twenty hours a week, which would severely impact their ability to do service and other work. Many academic libraries have gone to a scheduled consultation model for reference assistance, which is essentially what MPAL already has for most in-depth research queries. Further research could include a survey of music librarians to see what service models they use and how much of librarian time goes to desk shifts or reference work off desk, depending on their role.

It is difficult to ascertain why transactions post-merger have increased, as it could be due to a number of factors. For example, it could indicate that patrons now receive better and/or more efficient service and therefore come to the desk more often. Or it could indicate that not enough instruction is occurring at the desk and patrons return because they have not learned how to find the information themselves. Alternately, it could be some other variable as yet unidentified. Human interactions and patron intention and satisfaction are impossible to measure with this type of tool. MPAL previously did a study to measure service effectiveness and patron satisfaction and could re-run that study or a similar one.35 Reference staff are now in a better position (literally) to ask if patrons found everything they were looking for when acting as circulation backup and to interject in an interaction in which a circulation staff member needs assistance. We plan to continue to review Desk Tracker statistics annually in conjunction with all other data available such as circulation data and our space usage tracking efforts to help us understand how our library is being used and how it may need to evolve to meet patron needs. It will also help us to see where training in using Desk Tracker and READ may continue to be needed.

The data in this study does not speak to how (or if) patrons benefit from having subject-specialist librarians at the service desk. Further research can be done to measure whether it insures the shortest amount of time between a patron posing a question and receiving a satisfactory answer given the complexity of music materials. As Poparad noted when discussing the benefits of librarians at the desk, “Hearing the students’ questions firsthand at the desk informs how we teach in the classroom, in consultations, and through our online guides and tutorials.”36 We feel there is still a considerable benefit to having librarians scheduled at the desk since transactions there have led to instructional and collection development opportunities. This allows the librarians to see directly the interactions between patrons and collections, leading to a more holistic approach to service writ large.

Conclusion

MPAL’s service delivery and transaction profile is very different than it was five years ago, and it is hard to know what it might look like in another five years. The elephant in the library has long been whether having a separate reference desk matters, and there are increasing discussions over who should provide reference services. Libraries and music libraries (or at least this music and performing arts library) are seeing a decrease in number of reference assistance and other queries at the desk. As staff, resources, and patron needs continue to change, MPAL will need to continue to review available data and assess services to make service hours and staffing decisions. While the current configuration appears to be a successful change, this may prove to be less so in the future, or more effective models may arise.

References

1. Kay Vyhnanek and Christy Zlatos, Reconfiguring Service Delivery, (Washington, DC: Association of Research Libraries, 2011).
2. Shaneka Morris and Gary Roebuck, ARL Statistics 2015–2016. (Washington, DC: Association of Research Libraries, 2018).
3. Much of this process is described in several of the chapters in Difficult Decisions: Closing and Merging Academic Libraries, ed. Sara Holder and Amber Lannon (Chicago: Association of College and Research Libraries, a division of the American Library Association, 2015).
4. This does not account for the thousands of MPAL items now at the University Library’s off-site high-density storage facility.
5. Prior to the construction of the Music Building, where the Library now resides, it was in the original home of the music department, Smith Hall.
6. Aditi Bandyopadhyay and Mary Kate Boyd-Byrnes, “Is the Need for Mediated Reference Service in Academic Libraries Fading Away in the Digital Environment?” Reference Services Review 44, no. 4 (2016): 596–626, https://doi.org/10.1108/RSR-02-2016-0012.
7. Linda Frederiksen and Brandon Wilkinson, “Single Service Points in Libraries: A Review,” Journal of Access Services 13, no. 2 (2016): 131–40, https://doi.org/10.1080/15367967.2016.1161522.
8. Rebecca Jackson, “Revolution or Evolution: Reference Planning in ARL Libraries,” Reference Services Review 30, no. 3 (2002): 212–28, https://doi.org/10.1108/00907320210435482.
9. Duane Wilson, “Reenvisioning Access Services: A Survey of Access Services Departments in ARL Libraries,” Journal of Access Services 10, no. 3 (2013): 153–71, https://doi.org/10.1080/15367967.2013.795430.
10. Wilson, “Reenvisioning Access Services,” 160.
11. Janet Crane and Jeanne A. Pavy, “One-Stop Shopping: Merging Service Points in a University Library,” Public Services Quarterly 4, no. 1, (2008): 29–45, https://doi.org/10.1080/15228950802135707.
12. Pat Flanagan and Lisa R. Horowitz, “Exploring New Service Models: Can Consolidating Public Service Points Improve Response to Customer Needs?,” Journal of Academic Librarianship 26, no. 5 (2000): 329–38, https://doi.org/10.1016/S0099-1333(00)00139-7.
13. Bruce L. Keisling and Claudene Sproles, “Reviewing and Reforming Library Service Points: Lessons in Review and Planning Services, Building Layout, and Organizational Culture,” Library Management 38 no. 8/9 (2017): 426–36, https://doi.org/10.1108/LM-02-2017-0020.
14. Diane Hunter and Mary E. Anderson, “A Single Service Point,” in Reimagining Reference in the 21st Century, edited by David A. Tyckoson and John G. Dove (West Lafayette, IN: Purdue University Press, 2015): 207–13.
15. Hunter and Anderson, “A Single Service Point,” 211.
16. Leslie Abrams, “A Case Study in Transformative Renovation and Organizational Change at the University of California, San Diego Arts Library,” Art Documentation: Bulletin of the Art Libraries Society of North America 30, no. 2 (2011): 64–69.
17. Melia Erin Fritch, Laura Bonella, and Jason Coleman, “Nothing is Permanent but Change: The Journey to Consolidation and Back,” College & Undergraduate Libraries 21, no. 1 (2014): 2–18, https://doi.org/10.1080/10691316.2014.877724.
18. Rachel Applegate, “Whose Decline? Which Academic Libraries Are “Deserted” in Terms of Reference Transactions?” Reference & User Services Quarterly 48, no. 2 (2008): 176–89.
19. Joe Lenkart and Jen-chien Yu, “Specialized Reference Services at Illinois: Reference Transactional Analysis and Its Implications for Service Providers and Administrators,” Reference & User Services Quarterly 56, no. 4 (2017): 268–76, https://doi.org/10.5860/rusq.56.4.268.
20. Susan M. Ryan, “Reference Transactions Analysis: The Cost-Effectiveness of Staffing a Traditional Academic Reference Desk,” Journal of Academic Librarianship 34, no. 5 (2008): 389–99, https://doi.org/10.1016/j.acalib.2008.06.002.
21. Ryan, “Reference Transactions Analysis,” 395.
22. Deb Raftus and Kathleen Collins, “Does the Reference Desk Still Matter? Assessing the Desk Paradigm at the University of Washington Libraries,” in Rethinking Reference in Academic Libraries: Innovative Developments and Future Trends, edited by Carrie Forbes and Jennifer Bowers (Lanham, MD: Rowman & Littlefield, 2015), 153–66.
23. Bradley Wade Bishop and Jennifer A. Bartlett, “Where Do We Go from Here? Informing Academic Library Staffing through Reference Transaction Analysis,” College & Research Libraries 74, no. 5 (2013): 489–500.
24. Hazel McClure and Patricia Bravender, “Regarding Reference in an Academic Library: Does the Desk Make a Difference?” Reference & User Services Quarterly 52, no. 4 (2013): 302–8, https://doi.org/10.5860/rusq.52n4.302.
25. Pixey Anne Mosley, “Assessing User Interactions at the Desk Nearest the Front Door,” Reference & User Services Quarterly 47, no. 2 (2007): 159–67, https://doi.org/10.5860/rusq.47n2.159.
26. See Lisa Woznicki, “Transposing the Tradition: Providing Embedded Reference Service to Music Students,” Music Reference Services Quarterly 20, no. 2 (2017): 69–90, https://doi.org/10.1080/10588167.2017.1309933; Gerald Szymanski and Mary Alice Fields, “Virtual Reference in the Music Library,” Notes: Quarterly Journal of the Music Library Association 61, no. 3 (2005): 634–58; and Cassidy R Sugimoto, “Evaluating Reference Transactions in Academic Music Libraries,” Music Reference Services Quarterly 11, no. 1 (2008): 1–32, https://doi.org/10.1080/10588160802157124.
27. David Hursh, “Calling All Academic Music Library Reference Desks,” Music Reference Services Quarterly 8, no. 3 (2004): 63–81, https://doi.org/10.1300/J116v08n03_04.
28. Kirstin Dougan, “Delivering and Assessing Music Reference Service,” The Reference Librarian 54, no. 1 (2013): 38–54, https://doi.org/10.1080/02763877.2013.734759.
29. Nara L. Newcomer and David Hursh, “Calling All Academic Music Library Reference Desks: A Follow-Up Study,” Music Reference Services Quarterly 11, no. 2 (2008): 101–29, https://doi.org/10.1080/10588160802143405.
30. Bella Karr Gerlich, “The READ Scale,” http://readscale.org/.
31. Mary Du Mont, Alan Green, and Beth Christensen, “Taking Note: Assessing the Performance of Reference Service in Academic Music Libraries: A Progress Report,” Notes: Quarterly Journal of The Music Library Association 58, no. 1 (2001): 39–54, https://doi.org/10.1353/not.2001.0127.
32. Gerlich, “The READ Scale.”
33. “Division of Management Information,” University of Illinois, http://dmi.illinois.edu.
34. Bishop and Bartlett, “Where Do We Go from Here?,” 496.
35. Dougan, “Delivering and Assessing Music Reference Service.”
36. Christa E. Poparad, “Staffing an Information Desk for Maximum Responsiveness and Effectiveness in Meeting Research and Computing Needs in an Academic Library,” The Reference Librarian 56, no. 2 (2015): 89, https://doi.org/10.1080/02763877.2014.977212.