Analyzing Collections for Decision Making

(L-R) Diana Plunkett, Angie Miraflor, Lutgarda Barnachea

Being Lean But Not Mean: Weeding Print Collections to Improve Services

Lutgarda Barnachea, Assessment Coordinator, University of Maryland Libraries, described a comprehensive analysis of library collections at the University of Maryland. The goals were to reduce the footprint of the print collection and improve services to the community. They wanted the project to go as quickly as possible; it is now in its  second year. The main library is the focus of the weeding program; some branch libraries are also participating.

Motivations for weeding the print collection were:

Entering the building had become like entering a warehouse of books with the result that the collections were becoming disorganized. “Lost” materials were not really lost but were likely somewhere on the shelves. The result was a slowing of foot traffic in the library. The concept of the library as a space is changing.  In a series of interviews, students listed their needs and wants:

The research that was conducted had three phases:

1. An ethnographic study with interviews, drawing sessions asked students for their ideas on what the ideal space would  look like,
2. Users were observed inside the library. It was interesting to note that they wanted both quiet and noisy areas as well as flexible areas. They also wanted to be able to consult with library professionals. And the most surprising thing was that they wanted to  keep the books.
3. The survey data was given to a graduate class in architecture who designed the “ideal” library building. They also consulted with an architect to assess current spaces and see which ones could be easily changed or reorganized.

Content analysis: They looked at big picture using OCLC WorldShare to see the uniqueness of the collection and found that only about 15% of the items held are truly unique to the library.  Studies of publication dates and subjects were also conducted.

Usage: ALEPH reports generated charts of loan counts by call number, as well as shelf list reports by other criteria. Because of the wide variety of available reprots, it is important to have descriptive names for them. Collection development librarians marked items to be removed. Here are some strategies for weeding.

Other opportunities to reduce the collection footprint include moving items to a remote storage facility about 2 miles away from library. Here are others:

The results of this exercise were that a makerspace was created, rooms with laptops were set up. Some rooms have furniture that can be rearranged by users.

Lessons learned. Don’t be hindered by limited money.  Listen to users’ concerns.

Angie Miraflor and Diana Plunkett described tools for data-driven decisions at the Brooklyn NY public library (BPL). 37.5% of the population of Brooklyn is foreign born which raises challenges for the library. With 58 branches, BPL is the 5th largest library system in the U.S. It serves more than 15 million residents and circulates more than 30 million items in 15 languages. It has the largest free Wi-Fi network in the borough.

BookOps is a shared technical service organization serving both the New York Public Library  and BPL. It does collection management for both organizations and researches technical services. It processes 800,000 items in-house. Because it is a combination of two services, constant communication is important. Libraries watch their communities and what they are checking out and communicate that to the BookOps staff. BookOps works with vendors and does negotiations and makes budget recommendations based on a system-wide view.

Collection HQ software is used to produce reports. BPL produces a list of “grubby” (well worn) and dead items, top authors being checked out, and a popular subject summary to detect areas were more items should be obtained. This is evidence-based decision making.

Future projects include a Grubby items refresh to transfer copies between branches and Floating rebalancing to know what branches are overstocked and understocked. It is important to be thoughtful and careful about when you make changes. Another future project is to create staff-friendly training to produce local experts.

Knowledge of the collection is very critical. Dashboards have been created in Tableau. Details of any item can be created in several formats. It is possible to compare monthly data for this year vs. last year, for example. The library’s historic data goes back to late 1800s. Here is an example chart showing  hourly visit data.

This shows that Wednesdays at 3 PM are a popular time, which helps figure out staffing levels needed, when to do programs, etc.