Law libraries provide a host of data services. Law librarians help patrons find statistics for litigation briefs, draft regulations, and research papers. Academic law librarians collect citation counts, assess the factors driving these counts, and recommend additional channels for scholarly communication. Librarians build internal,  library datasets and create statistics for institutional reporting and inter-institutional benchmarking. Librarians assist patrons in generating original research data, including navigating university human subjects committees, or institutional review boards (I.R.B.s). Corporate and firm law librarians help patrons use data and statistics for competitive intelligence, or using data to increase an organization’s business advantage. Legal information professionals license data for individual projects and organization-wide use. Data and informatics work is a rich and growing specialty within law librarianship.

While data services are diverse, this chapter describes four common service areas: 1) Finding data and statistics; 2) Gathering bibliometrics and helping patrons understand citation counts; 3) Collecting and reporting internal, library data and statistics; and 4) Assisting patrons with original data collection.

Finding Data and Statistics

Law librarians help their patrons find data, statistics, and secondary statistical analysis (SSA), or writings that interpret statistics. Just as we are trusted to recommend treatises and loose-leaf publications, patrons have faith that we can find data on their topics. Law librarians can develop their data-finding skills by visiting data sites, employing data-finding questions, and learning data-purchasing processes.

First, law librarians can lay the foundation for data expertise by visiting data sites and empirical legal repositories. Monthly, law librarians can visit the U.S. government’s open data portal at data.gov. Quarterly, librarians can explore web-based government agency and nonprofit organization data and statistics repositories, such as the National Institute of Justice site, Bureau of Justice Statistics site, and National Conference of State Legislatures site. Librarians can then delve into sites that correspond with the predominant needs of their primary patrons, such as criminal justice data, economic trend data, or elections data. Routinely, law librarians can attend legal data vendor trainings, such as Bloomberg Law for corporate law data, UniCourt for case tracking and docket analytics, ALM for firm, hiring, and legal industry analytics, TRACfed for administrative agency and government employee data, etc. Librarians can find additional data sources and sites through empirical legal research guides published by dozens of law schools, including the University of Arizona College of Law and the University of Chicago Law School.

Second, law librarians can employ data-finding questions to discover more resources for their learning and to help patrons seeking data, statistics, and SSA. The three key questions to ask when seeking new data are:

  “1. Who would have an incentive to collect (raw) data about [the] subject?

    2. Who would maintain ready-made statistics about [the] subject?

    3. Where would we find secondary statistical analysis of this subject?” (Ryan, 2013, p. 32).

Data-collection incentives include government mandates, nonprofit missions, and profit motivations. For instance, federal statutes and regulations require many agencies to report spending data. Organizations such as the World Bank and Vera Institute, which focuses on reducing mass incarceration, collect and publish data and statistics in support of their work. Legal intelligence companies sell profitability data to law firms (e.g., ALM profitability per attorney reports, Thomson Reuters law firm profitability solutions). Once a law librarian identifies the type of entities that must, should, or would want to collect data, that information can drive better Internet searching for those entities and their free or proprietary data holdings. ?

While governments and organizations around the globe disseminate free data, the best data for some projects, including law firm profit maximization projects and large-sample empirical research projects, often needs to be purchased. Technical service librarians are well-versed in negotiating contracts and can assist reference librarians in knowing how to properly obtain a data, database, or statistical product license. Licensing is a legal process that requires the input of organizational fiduciaries and legal signatories. Even a text and data mining (TDM) license costing a few hundred dollars could require the review of commercial agreements officers, general counsel, fund managers, etc. Once a librarian finds a dataset or database (e.g., Wharton Research Data Services) to buy, the next questions should be: 1) Who will pay for the dataset/database?, 2) Who can authorize the licensing agreement?, and 3) Who will release the funds to purchase the dataset/database? The law librarian should allow several months for complicated licenses, particularly if the dataset/database will be purchased or used by multiple patrons or departments.

Regardless of whether data is found or purchased, law librarians should help patrons think critically about its limitations. For instance and as discussed further below, data about a group of people—such as Newark residents—should accurately reflect the characteristics of those people. When examining survey data on Newark residents, the patron must determine whether the total number of people surveyed is sufficient to permit statistical inference to all Newark residents, and if the numbers in subgroups—Asian residents, men, senior citizens—roughly match their proportion within the actual population. For this second concern, the patron can compare the survey data to demographic data from the U.S. Census or elsewhere.

The point is: law librarians can help patrons critically evaluate found and purchased data to determine its strengths and weaknesses. Librarians and their patrons can also critically appraise bibliometrics.

Gathering Bibliometrics & Understanding the Factors that Drive Citation Counts

Most academic law libraries provide bibliometrics, or research citation statistics, support as part of their scholarly communication services. Bibliometrics support is especially important to junior faculty, as they can use citation statistics to demonstrate their output and impact during the tenure review process. Faculty can also use bibliometrics to select journals for their work, to make a case for their research impact in grant applications, and more. Student editors can use bibliometrics to assess the impact of their law reviews. Attorneys can use citation counts to argue the importance of a given secondary source in a legal brief.

Bibliometric support can be circular: a law librarian will access and discuss a legal researcher’s citation statistics with the researcher, then help that patron diversify or improve their scholarly messaging, then assess if citation counts are rising, and so forth.

For the scholarly messaging piece, librarians can assist scholars in establishing an ORCID iD (i.e., a scholar’s unique digital identifier); contribute research to repositories; build ResearchGate and SSRN pages; establish a wikidata author item and Wikipedia page; etc. Law librarians can also assist patrons in understanding the algorithms, data, and trends driving the bibliometrics produced by different scholarly venues.

4.4 How to create a faculty Wikipedia page (by Adrienne Kelish)

“How many times has an article been cited?” is a seemingly straight-forward question, but the answer will vary by database (Whisner, 2018a, 2018b, 2018c, 2018d). Legal scholars track their citations in a range of databases, including: HeinOnline, Lexis (Shepard’s), Westlaw (Key Cite), Web of Science, and Google Scholar (Watson, forthcoming; Whisner, 2018a). The citation counts produced by each database—for the same article—vary by what the databases index and how they retrieve citations. HeinOnline covers most U.S. law reviews but uses a Bluebook citation algorithm (e.g., number of articles citing to “52 Loy. U. Chi. L.J. 67”) to retrieve citations (Whisner 2018a). By contrast, Web of Science/Web of Knowledge employs a more flexible algorithm (e.g., including author name) but does not index many law reviews (Whisner, 2018a). Lexis and Westlaw, via their Shepard’s and Key Cite features (respectively), retrieve citations in Supreme Court briefs and other legal materials, but can miss social science citations. Google Scholar captures more interdisciplinary citations but misses Supreme Court briefs and other legal materials. SSRN tracks social media mentions and counts downloads (Watson, forthcoming); this complicates the meaning of “citation” because these references to the work are informal and not managed by editors. As a result, obtaining citation statistics for a given article requires searching across multiple platforms, assembling a master bibliography of citations, and reporting citations as types (e.g., research article, tweets).

“What factors drive citation?” is a different and difficult question, and the subject of countless studies (see, e.g., Ayres & Vars, 2000; Bornmann, & Daniel, 2008; Eisenberg & Wells, 1998; Shapiro, 2021; Shapiro & Pearse, 2011; Whisner, 2018b, 2018c, 2018d, Yoon, 2013). According to social scientists Lutz Bornmann and Hans-Dieter Daniel, seven factors influence the probability of being cited (2008): publication date, research field, journal, article, author, digital availability of the article, and technical issues.

First, publication date affects citation in at least three ways: authors tend to cite more recent studies (see also Ayres & Vars, 2000; Whisner, 2018c); authors cite highly-cited studies, or studies that have circulated long enough to be highly-cited; and authors have fewer choices when citing pre-1980s research because there were fewer journals and databases then (Bornmann, & Daniel; Lowe & Wallace, 2011), and some pre-Internet content has yet to be digitized.

Second, each field has its own citation practices and trends (Bornmann, & Daniel). Historians might be more open to citing older research. Small specialty fields have fewer researchers citing each other. Legal scholars tend to cite articles from law reviews at higher ranked law schools (i.e., per U.S. News & World Report rankings; Ayres & Vars, 2000; Whisner, 2018b), though articles at lower ranked schools can break through when they address unique areas of law (Whisner, 2018c). Law librarianship is a low-citation field; a typical law librarianship article will receive three or fewer citations (Togam & Mestri, 2021). Field differences matter when faculty and librarians are judged by interdisciplinary tenure, promotion, and awards committees.

Third, journals generate different average per-article citations (Bornmann, & Daniel), a phenomenon that drives journal impact factor. In legal research, articles appearing in the flagship journals of highly-ranked law schools enjoy more citations than most other law reviews (Ayres & Vars, 2000; Eisenberg & Wells, 1998; Whisner, 2018b). This holds true even for student-authored notes and comments in the Harvard Law Review (Whisner, 2018b). In Communication, Communication Monographs has a much higher impact factor than any other U.S. journal in the field (National Communication Association, 2022). A host of factors drive journal impact metrics, including editorial bias (Yoon, 2013), marketing, number of issues per year, and more (Bornmann, & Daniel).

Fourth, article factors drive citations (Bornmann, & Daniel). The first article in an issue tends to receive higher citations (Bornmann, & Daniel). Letters, book reviews, and other short pieces tend to receive fewer citations (Bornmann, & Daniel). Article length also drives citations (Bornmann, & Daniel) though legal scholars see diminishing returns for length once a law review article has met a typical threshold (i.e., 40-70 pages; Ayres & Vars, 2000).

Fifth, authorship characteristics affect citation frequency. In many social sciences, co-authored articles are more heavily cited than solo-authored articles (Bornmann, & Daniel). In the law, that correlation is weaker (Ayres & Vars, 2000), in part because law reviews publish far more solo-authored articles than social science journals (Tietz, Price, & Nicholson, 2020). Historically, men received substantially more citations to their work than women (Bornmann, & Daniel), and White scholars received more citations than scholars of color (Bell, 2007; Nunna, Price II, & Tietz, forthcoming), though Ian Ayres and Frederick Vars found that women and non-White authors fared better when they published in the Harvard, Stanford, or Yale law journals. Law professors are cited more often than law students (Whisner, 2018d). Social networks and scholarly connections drive citations too, as authors favor people they know personally (Bornmann, & Daniel), and law professors acknowledge and cite authors from their own law schools more frequently than scholars at other schools (Ayres & Vars, 2000; Tietz, Price, & Nicholson, 2020).

Sixth, availability of articles in digital and open-access formats positively correlates with citations (Bornmann, & Daniel). Seventh, technical issues affect citation, including misspelled author names and publication information in metadata and reference lists (Bornmann, & Daniel; Watson, forthcoming).

Law librarians should understand how to retrieve citation counts and how to explain the factors driving those counts. Law librarians should also care about bibliometrics because they affect our scholarly communication and professional advancement. Similarly, law librarians should be conversant in law library metrics and assessment because these statistics affect our individual and collective professional strength. Librarians can approach library assessment as an opportunity to learn about research design, data collection, dataset construction, statistical analysis, and reporting.

Collecting Library Data, Creating Datasets, Analyzing Trends, and Reporting Information

Law librarians collect data and report statistics to numerous stakeholders. For instance, U.S. academic law libraries submit library statistics to the Association of College & Research Libraries (ACRL), Association of Research Libraries (ARL), U.S. News & World Report (USNWR), their law school deans, and others (McAllister & Brown, 2019; NELLCO, 2021; Panella, Iaconeta, & Miguel-Stearns, 2017). Prior to 2017, the American Bar Association (ABA) also collected extensive information from academic law libraries (Panella, Iaconeta, & Miguel-Stearns). In 2014, Teresa Miguel-Stearns, then director of the Lillian Goldman Law Library at Yale Law School, led the charge to consolidate data reporting into a single system that would enable library directors to pull reports for different stakeholders and benchmark their collections and services against peer libraries’ collections and services (Panella, Iaconeta, & Miguel-Stearns). Today, the national reporting and benchmarking system, known as ALLStAR, is maintained by the NELLCO consortium (NELLCO, 2021).

For reporting and benchmarking, library directors collect data from every department. Technical services librarians provide usage counts for databases and eResources (McAllister & Brown). Reference librarians aggregate data from patron interactions and reference questions. Access Services librarians gather circulation data. All departments report spending and personnel data. Commonly reported data are organized into 10 categories in the ALLStar 2021 survey (Figure 4.5). Because information services are boundary-crossing, ALLStAR items sometimes appear in non-obvious categories. For instance, library directors reported some circulation statistics in the Services section of the 2021 ALLStAR survey (NELLCO, 2021).

While library benchmarking can involve the whole staff in data collection and dataset creation, it can also allow empirical librarians to practice their trend analysis and reporting skills. For instance, librarians have been challenged to estimate post-pandemic eResource usage (Lowe, 2020). This estimation was difficult in the early phase of the pandemic because there was insufficient data for a pre- and post-emergency comparison. Now that we have years of pandemic-era data and visitor restrictions have relaxed, we can better predict future eResource usage. Empirical law librarians can produce data charts and research write-ups to help administrators effectively budget for eResources.

Assisting with Original Data Collection/Creation

While many legal research questions can be answered with existing demographic, legal system, or other data, some require the collection of new—or “original”—data. Empirical legal researchers can collect data via a host of methods and instruments, including interviews and surveys. Original data collection/creation requires a systematic process and ample planning. Data collection begins with the articulation of hypotheses and research questions.