America's Star Libraries, 2010: State-level Views on Data
LJ Index of Public Library Service 2010: State-level perspectives on LJ Index ratings - America's Star Libraries - best libraries - top-rated libraries By Ray Lyons & Keith Curry Lance Oct 1, 2010
|
Our hope in publishing the LJ Index is that seeing the reported data on which scores and star library ratings are based and discussing the results will lead public library directors and others to share their data collection methods as well as their exemplary practices. Then public libraries will identify models of practice—and data collection —to use in improving their own statistics out of a genuine desire to represent accurately what they deliver to their communities.
Whether or not that happens depends in part on the efforts of people at the state level: members of Chief Officers of State Libraries Agencies (COSLA) and their State Data Coordinators (SDCs). Local libraries do not complete a federal questionnaire; they report statistics to their state library agencies, and IMLS compiles them nationally. We asked veteran representatives of the state level groups to reflect on the challenges they face in collecting complete and accurate statistics from public libraries.
Of the four LJ Index per-capita measures—visits, circulation, program attendance, and Internet computer use—which do you believe local libraries have the greatest difficulty reporting well, and why?
Texas state librarian Peggy D. Rudd—who is also the chair of the COSLA Research and Statistics Committee, a member of the IMLS Library Statistics Working Group, and a former SDC herself—conferred with her SDC, Stacey Malek. They believe that "most Texas libraries have automated counts of circulation and visits; however, program attendance is determined by the old-fashioned head count" [while] "most libraries [are] left to their own devices" in counting "computer users/uses in the library."
Alaska SDC Patience Frederiksen agrees about the difficulty of counting Internet computer usage, and suggests a likely explanation. "Internet computer usage is frequently a non-staff-mediated service," she says. "Staff actually check materials out to patrons, they are frequently sitting near the exit, so they have a sense of how many people are [visiting] the library, and staff actually present programs. People more frequently expect to do-it-themselves when it comes to sitting down in front of a computer. They do not need staff help with this service, so that makes Internet computer usage statistics and surveys more difficult for library staff."
Kit Keller, a longtime (now former) Nebraska SDC, identifies some of the specific variables that challenge the comparability of this statistic from library to library. "Libraries with time limits (sometimes one hour, sometimes half-hours) might count each slot as one use; others might count each unique user as one use. This can create quite a variance—someone might sit for several hours, counting as one, whereas those with stricter rules and higher demand might have three, four, or five. Smaller libraries might allow a 'regular' to sit at length. So the variables of demand, time-slot length, and rigor of counting process all affect this count."
Kentucky SDC Jay Bank expresses concern about the comparability of the statistic in which most would place the greatest confidence: circulation. "I have discovered that just about every type of library activity has been added in with circulation totals by one library or another: computer use, in-house use, programs, OPAC access, and more." He also expresses a rarely-mentioned concern about program attendance, though it must be a problem in most states that survey for a July-June fiscal year—that, as he puts it, "summer reading, the largest program, is often in progress when the reporting period ends."
What kinds of assistance does your state library agency provide to help data reporters overcome these challenges?
Our four state library contacts report a variety of assistance to data reporters. All four report providing assistance via individual consulting contacts via email and telephone and in person.
Rudd (TX) and Bank (KY) indicate relying heavily on the definitions, help screens, and IMLS edit checks built into their Bibliostat Collect online surveys.
During her tenure in Nebraska, Keller reports, she relied upon regional workshops "designed to work through the reporting form. Librarians were told to bring their information with them… and each started completing the form during the session. This way I was available to respond to questions, which presented a teachable moment for everybody present. When the librarian is sitting alone at a terminal trying to interpret a question, she might 'just answer' it without taking the time to read the instructions or make a phone call to the state library. However, if she can call out her question in a group setting and get an explanation on the spot, she's more likely to do so."
Keller also promoted local data use by various means. "I observed over time that the more the librarians realized the value of having quality, comparable data for their own use," she says, "the better both the response rate and the quality of the responses on individual annual reports."
Alaska's Frederiksen also relies upon her agency's annual "librarian boot camps" as well as "two audioconferences the month before the report is due."
What kinds of assistance with data collection do you wish you could provide from the state level, given more resources?
Rudd (TX) shares Keller's view about the connection between data quality and data use, calling for "more on-site training with emphasis on using the data—the direct link between good data collection and having reliable data to use for a variety of purposes must be constantly stressed." She adds, "Once library staff are successful [data] users, in my view, they become more conscientious data collectors."
Malek (TX) notes that staff turnover in general in smaller libraries is problematic and "director turnover is a real problem in data collection, particularly where there are not professional librarians." Frederiksen reports that she "makes a point of getting in touch with a new librarian and offering help or suggesting they attend the audioconference, which is a simple Q & A [about the Alaska survey]."
In an ideal world, she suggests, "a 'new librarian' training program would include an introduction to the data collection process, possibly with prepared templates for accurate monthly/quarterly data collection. This could include discussions of many job aspects, such as budgeting, circulation, weeding, and collection management, in conjunction with a process to document these activities."
What national solutions to data collection issues do you believe would be most effective?
Bank (KY) recommends "an ongoing re-evaluation of the data elements [because] ... the types of materials, access, and library use seem to be progressing at a greater speed than our ability to categorize and capture them."
Keller (NE) believes "the annual SDC conference [goes] a long way towards establishing consistent processes, definitions, and responses to librarians," and believes it is a key mechanism for maintaining an ongoing re-evaluation of changing data needs.
There is consensus among our state contacts, however, that there are no simple national solutions to many of the challenges faced in collecting national public library statistics. Rudd (TX) is, she says, "tempted to say all data solutions are local."
Frederiksen (AK) concurs. "I doubt a single method can be developed to gather any of these numbers." That being the case, she adds "it is helpful to learn how other SDCs solicit good statistics from their libraries."
