What is data services?

How many reference librarians start to inwardly cringe when they hear the word statistics or data? We can’t be experts in everything, that’s why we refer in-depth questions to subject specialists. Special formats (government information, for example) have their own particular quirks that necessitate specialist librarians, as well. Data services falls into this category.

Machine-readable data files for secondary analysis came onto the research scene in a big way in 1962 with the founding of the Inter-University Consortium for Political and Social Research, one of the world’s premier social science data archives. The road to mainstream academic use of datasets follows the path of advancements in computing technology. Now students are trained in data analysis using statistical software programs and, in the process, are often introduced to public use datasets like the General Social Survey. Demand for datasets is set to keep pace with the progressively important role of quantitative research methods.

At MSU, requests from researchers for numeric data were previously handled on an ad-hoc basis, with individual subject librarians providing assistance with varying levels of expertise. This stretched busy librarians beyond their knowledge base of available resources—if researchers even thought to go to the library for data in the first place. While questions for statistics are commonplace, numeric datasets have not always been associated with traditional library collections. Academic departments have historically worked independently to provide their faculty with access to data with varying levels of success. The status quo has long caused isolation, duplication, and confusion. Librarians saw this as an opportunity to fill a void and bring the same level of service to data as they already bring to other formats of information.

With the creation of the data services librarian position, one individual retains responsibility for this specialized format across disciplines. Researchers can now expect increased attention to collection development of data resources and reference support. However, since the MSU Libraries never provided this level of support for data before, researchers must be informed that they have access to this service.

Over two years, two models of outreach were followed based on promotion of different areas of emphasis for the data librarian: collection management and reference. The outreach model for reference services addresses the need to integrate data services into the library system and promote data services to the academic community at large, whereas the outreach model for collection management is a case study for establishing a partnership with an academic department.

Emphasizing data reference services across departments

Introducing new students and faculty to—and reminding current students and faculty of—library services is an ongoing process in any academic library. There is no need to reinvent the wheel when it comes to promotional activities in your library; piggyback on to existing efforts as much as possible. However, if you are new to the organization, you will need to do legwork to find out what the existing programs are and who to talk to about bundling data services into the mix. The key point here is that the first place to promote and create buy-in for data services is within the library itself.

Fostering friendly working relationships with colleagues is important to the development and success of any service. This is particularly true for data services because of its interdisciplinary nature. As data is a format and not a subject, there is not a specific departmental liaison relationship built into the job like other traditional bibliographer positions. The data librarian should work cooperatively with colleagues in all outreach efforts. This should be a two-pronged effort: first, educate peers on your expertise, and second, ask them to share their expertise.

First, as a new data librarian you represent a freshly minted service that may not be well defined. Presenting your service plan and sharing data reference strategies at departmental meetings will help to clarify your role. Also, if you are new to the organization, the act of simply getting to know your coworkers will facilitate much of the process. Fellow librarians are sure to be a major source of referrals, but it may not always be clear-cut as to whether a data question belongs to a subject bibliographer or data services. Suggest shared referrals, which create the opportunity for collaboration and may ultimately lead to a co-worker who is busy or unsure of data resources happily handing the referral to data services.

Second, ask colleagues to share their expertise. For reference purposes, the data librarian must become conversant in multiple subject domains and may need to draw upon others knowledge. Direct outreach efforts to departments simply cannot be done without working with the liaison, both to keep the data librarian from committing unwanted turf creep and to avoid duplication of efforts. As long as the data services role has been clearly defined as complementary to (but not the same as) the work of liaison librarians, most colleagues should be willing to share outreach activities.

Reaching out to campus centers and programs with related missions is also beneficial. At MSU, the Center for Statistical Training and Consulting (CSTAT) is a service unit that provides training and consulting in statistics. Researchers who need assistance with statistical software and analysis can use CSTAT’s services. This is distinct from library data services, where the focus is on access to resources but not the mechanics of statistical analysis. Since researchers need data to perform statistical analysis, reaching out to CSTAT was a natural fit. After meeting with the director of CSTAT, the opportunity for mutual referrals was established.

Another way to interact with potential data users is to target classes with a quantitative analysis component that may involve the use of numeric datasets by sending an introductory e-mail to instructors. Auditing a relevant course is another option to get you close to your potential users. It is time-intensive but provides the opportunity to make contacts among students and faculty, find out how classes are using numeric data, and “talk the talk” of your users. Any chance you have to make personal connections with faculty and students is especially important.

Opportunities to meet in-person with faculty and students are the most direct form of outreach, but indirect strategies should also be pursued. It is important to establish a Web presence that identifies data services as a unique library program and also seam-lessly integrates data services and resources into the overall structure of the library Web site. One way to accomplish this is through the creation of a separate category of data resources and by posting research guides that will help answer ready reference questions and serve as a reminder to refer in-depth questions to the data librarian.

Making and distributing a flyer or brochure is also a useful investment. A brochure can be bundled in with other library promotional items at resource fairs, get handed out as a referral tool, and distributed to departmental mailboxes.

Overall, promotion of data services involves cultivating relationships both inside and outside of the library. The ideas covered above represent an effort to raise awareness of data services and increase referrals to the data librarian. It is not an overnight process, but takes time and sustained effort. The next promotional model, a project to integrate the library into data collection management for purchases made by the Eli Broad College of Business (COB), is the result of almost a full year of outreach work. It represents a directed outreach effort that targets relationship building with a single department.

The College of Business and data collection management

A beginning goal as data librarian was to talk to faculty about data and how they use it. It was known from colleagues at the Gast Business Library that faculty at the COB were prolific users of data. Therefore, it was decided that the business faculty would be an excellent first user base to cultivate.

The objective was to create a central list of data purchased, created, or used by faculty by talking to anyone with knowledge or interest in data. Armed with business cards and a notebook, a number of months were spent knocking on doors and talking to faculty. It also became valuable to work with the business librarians because their combined expertise was useful, and it helped all parties make inroads with faculty, including being asked to attend department meetings.

The breakthrough came when a discussion with a faculty member revealed the existence of a research committee with a prior interest in data that was able to provide a draft list of data purchases. Around the same time, COB administrators had a strong interest in investigating purchases and potential cost-saving measures. Administrative support is essential and puts a force behind expeditious action taking.

In addition, a top figure in the COB saw librarians as silo crossings, or people in a unique position to see the big picture across campus, while departments and colleges are typically are more focused on their own interests. For example, a college might duplicate resources or unwittingly pay for a database that is already available from a campus association membership. This expensive lesson shows that cross-campus communication and resource sharing is essential.

After sustained searching and question asking, the data librarian created a spreadsheet using information provided by the research committee adding additional information like description, alternative resources, and location. Library data resources were also added to raise awareness. Any duplicate data were noted. The data and business librarians were invited to a meeting with administration and department chairs to share the findings and to discuss the potential for a continuing relationship between the library and college.

The COB and library agreed that data purchases would go through the library. The library assists with purchases, examines license agreements, and suggests alternative resources, as necessary. The COB posted the spreadsheet on their intranet Web site for easy access.

In addition, faculty are encouraged to and do contact the librarians for data assistance. The spreadsheet is updated regularly. All of this encourages a continuing and valued relationship between the library and the COB.