View source for Studies:Cornell University Library

{{Case Study
| type=academic
| type2=Academic Library
| library=Cornell University Library
| contact=Zsuzsa Koltay, Director of Assessment and Communication and Linda Miller, Research and Assessment Librarian
| study=
====The Cornell Challenge: Entering and Querying Statistical Data for Multiple Units====
Cornell University Library is a large and complex library system with multiple unit libraries. Collecting statistics centrally from disparate data contributors to comply with ARL and NCES requirements as well as meet the needs of their own institution has been an ongoing challenge.

In the past, Library Research and Assessment staff used Excel spreadsheets to compile a 60-80 page statistical report annually. Managing and reporting from the data was complicated by the fact that units sometimes report counts together, (e.g., 3 units share a gate count, 2 units share staff, etc.). Responding to requests for custom reports was difficult and time-consuming, especially reports that cut across measure groups, units and years. For example, in the current economic climate, the library is reviewing its organization structure. Providing statistics to support this review has been a labor-intensive manual process because the data has been in annual Excel files, grouped by measure area (e.g., collections, circulation, reference, expenditures).

Library Research and Assessment staff needed a database solution with web-based data entry and querying. The goal was to compile and report statistics more easily, and to open access to the data to staff from across the library. Such a system would provide more timely evidence for better decision-making and reduce the reliance on a limited number of staff for generating reports. The library would move from reporting statistics “just in case” to providing access to needed data “just in time.”

“People expect to find databases online that they can query,” says Zsuzsa Koltay, Director of Assessment and Communication. “We have complex and rich data that we want people to make better use of without making every query a big project. Our annual statistical report was ‘just in case’ with all our numbers out there for people to page through as opposed to ‘just in time’ as a database that people can query based on what their question is.”

====The LibPAS Solution: Just in Time vs. Just in Case====
The library selected LibPAS from Counting Opinions to provide them with the centralized database they needed. Although still in the early stages of implementation, library staff can see that they are on the path to streamlining the way they collect, manage and report data.

To effectively use LibPAS, library staff needed to evaluate what they wanted to count and how they wanted to count it. Implementing LibPAS has also given Cornell the opportunity to continue to examine which data needs to be collected centrally. For example, does a performance indicator such as number of laptop computers checked out by graduate students need to be tracked in LibPAS or is it sufficient to have this queryable through the ILS?

In reviewing performance indicators as part of implementing LibPAS, library staff want to emphasize statistics that demonstrate the impact the library has on the university community. According to Koltay, there is a shift in their culture toward evidence-based decision making. “Delivering data will be much more efficient with LibPAS,” she says. “For people to be able to query LibPAS, and extract the data they need, will result in better decisions.”

Library staff believe LibPAS is taking them in the direction of making data more readily accessible to those who need information prior to decision-making. Koltay and Linda Miller, Research and Assessment Librarian, point to the following key benefits Cornell will realize through the use of LibPAS:
* '''Ability to readily produce unique sets of data needed for decision-making''' (including, once retrospective data is entered, longitudinal data sets).
* '''Open efficient reporting to staff across the library.'''
* '''Simplify the data collection for data providers.''' For example, performance indicator definitions will be available in the database rather than having to look them up separately, and footnotes can be directly entered as appropriate.
* '''Once formatted, easy importation of centrally collected or retrospective data.'''
* '''Ability to quickly create reports and switch between available report formats and graphing displays.'''

They also see a potential for benchmarking against similar institutions and the possibility of matching annual data with qualitative data collected through LibSat, which they have yet to investigate.

====What Comes Next: Evidence-Based Decision Making====
Library staff have been working with the Counting Opinions development team on enhancements to benefit their library as well as other ARL institutions. The first of these projects was completed before Cornell’s deadline of July 1, 2010, the start of their 2009/10 annual data collection. The project focused on making it easier to extract data when unit libraries report some but not other performance indicators together. Koltay says they knew implementing a system under development would be a little rocky. “We had a timeline and we had this huge issue. We appreciated Counting Opinions’ flexibility and willingness to work with us to develop their system to meet our needs.”

A second development project will involve making it possible to use SUSHI to directly import data from local library systems. This would eliminate the multiple manual steps needed to import local systems-held data into LibPAS.

Miller and Koltay are pleased with their partnership with Counting Opinions. They recognize they are pioneers but their belief in the benefits of LibPAS outweighs the challenges of working with a system in development. “Someone needs to go first,” Koltay says. “Only by using the system are some of the needs identified for complex ARL institutions like Cornell. We benefit from this relationship by helping create the system we want and that other institutions like ours will need.”

Looking ahead, Koltay says they believe that use of LibPAS will drive itself. Once training is complete she sees LibPAS taking the library in the direction it needs to go. “Data becomes something that is in front of everyone in the organization and readily accessible to answer their questions prior to decision making. We believe LibPAS is the right direction for institutions like ours.”

====About Cornell University====
With almost 14,000 undergraduates and over 7,500 graduate students, Cornell University is the largest and youngest university in the Ivy League. Its campus includes 14 colleges and schools: seven undergraduate units and four graduate and professional units in Ithaca; two medical graduate and professional units in New York City, and one in Doha, Qatar.

Cornell University Library (CUL) is one of the largest academic research libraries in the United States, with unit libraries on Cornell’s Ithaca campus, as well as libraries at the Weill Medical College in New York City and the Agricultural Experiment Station in Geneva, New York. CUL holds approximately 7.7 million printed volumes, 8.5 million microforms, and 76,600 linear feet of manuscript materials. It subscribes to about 99,000 journal and other serial titles, about 70% of which are available electronically to the Cornell community. In addition, the library provides access to over 650,000 electronic books and other networked resources including 54 locally created digital collections, such as The Making of America, that are used by people around the world.
}}