When Quantitative Analysis Lies Behind a Reference Question

Abstract

Statistics reference questions represent more than just a specialized retrieval challenge. When the fundamental point behind asking for data is to practice quantitative analysis techniques, reference librarians who are relative strangers to quantitative analysis and its vocabulary work at a disadvantage. To aid these reference librarians, here we present key quantitative analysis vocabulary. We will walk the reader through an elementary, fictitious example, and then a more complex, real example, of correlation and linear regression analysis. Integral use is made of statistical software (SPSS, version 8) and a dataset (Integrated Public Use Microdata Series) downloaded from the Internet, in order to instill in reference librarians a better sense of comfort and literacy with common quantitative analysis techniques using state-of-the-art methods. When quantitative analysis lies behind a reference question, anxiety and confusion can color even the most competent reference librarian's response. Searches for statistics arise frequently at reference desks, especially in academic library settings. In many cases, the completion of a statistics reference question also marks the beginning of the next phase of a student's work: performing techniques of quantitative analysis. Quantitative analysis plays such a role for students (and other patrons) because it is in the ascendancy in many of their fields of study, not only in the central social sciences of sociology, anthropology, political science, and economics, but also in disciplines often but not always regarded as social scientific, namely history and psychology.(1) Certain areas of applied learning, such as management, finance, education, and various health fields, also conduct research in which hypotheses undergo formal quantitative analysis. Reference librarians arguably do not prefer statistics reference questions. Consider the following four types of information--verses of poetry, a legislator's interest-group rating, the chemical structure of the benzene ring, and the regression of unemployment data run against minimum-wage-rate statistics. We suspect that a minority--perhaps only a small minority--of reference librarians would pick the fourth example as the type of query with which they are most at ease. Successful data retrieval is not at issue here; rather, the point is grasping what comes next, what the patron does with the numbers found, and what role the reference can best play. Why is that so? One might respond that reference librarians' unfamiliarity with quantitative analysis stems from the librarian demographic. The canonical profile of librarians as former English (or other humanities) majors, whose proportion of math phobics is at least as high as that of the general human population, might be at once an exaggeration and not too far off the mark. Alternatively, the statistics discomfort phenomenon can be traced to the nature of librarians' graduate school exposure to quantitative methodologies. Graduate library education has tended not to teach quantitative analysis as a core skill for the profession but as an optional side topic, useful only when conducting or publishing quantitative research. Why does this matter? To what degree is it necessary for reference librarians to grasp intellectually what inquirers do with the answers to reference questions? Patrons, after all, often judge the outcome of their reference questions successful upon the straightforward finding of a needed poem, a piece of physical science data, or a printout of citations or full text. The librarian's understanding of subsequent use is not always of importance. However, in the case of statistics reference, librarians' lack of grasp of the student's or other patrons' follow-up work, quantitative analysis, can actually lead to reference service shortcomings. The reference profession--probably in response to the challenges of statistics reference--has been assisted by a recent proliferation of published statistics in print and electronic formats. …

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 17 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.