Why aren't they using what we've got?

Interlibrary Loan requests for locally held items
Why aren’t they using what we’ve got?

Interlibrary Loan (ILL) is intended to supplement a library’s collection, but often patrons submit requests for items already owned. An analysis of a large academic library’s ILL requests shows that in most cases the cataloging of bibliographic records is correct, but patrons need instruction in bibliographic searches and ILL policies and procedures. An unexpected result of the study was a realization that patrons are confused by the way items are shelved in the various branches.

Interlibrary Loan (ILL) is intended to provide an additional library resource for patrons. The National Interlibrary Loan Code for the United States, 1993 (Code) takes this into consideration where it states, “The purpose of interlibrary loan as defined by this code is to obtain, upon request of a library user, materials not available in the user's local library (emphasis added). Thus one of the first items of business for an ILL office is to check the local catalog to ascertain if the library holds the requested item in any form. If the item is locally held and the patron still orders it, the questions then become: Why was this request made? Is the catalog record correct, did the patron accurately search the catalog, or does the patron understand or know ILL policy?

The most frequent reason, 41% in 1998, for canceling ILL requests at the University of Florida (UF) is “UF owns” (i.e., patron requests an item which UF owns in some form). To find out why such requests are made and to seek strategies for reducing the number of "UF owns" requests, we thoroughly examined a selection of requests made to the borrowing unit at UF. We wanted to test the quality of the catalog and the ability of patrons to use the OPAC (called LUIS at UF). "UF owns" requests for the first quarter of 1998 were scrutinized and compared to the OPAC records for the requested materials. We assumed that errors in catalog records might be the culprits. We concluded that in most cases the cataloging of bibliographic records is correct, but that patrons need instruction in bibliographic searches and ILL policies and procedures. An unexpected result of the study was a realization that patrons are confused by the way items are shelved in the various branches.

The most comprehensive article to date on the subject of requesting locally held items was by Scott Seaman, who examined the problem of “Online Catalog Failure” in terms of patrons’ and catalog error. In this study of requests submitted in 1988-89, he found that patrons’ errors were largely due to use of stoplist words (.9%), spelling errors (7.3%), unclear abbreviations (9%), incorrect information (40%), and inexplicable failure to find the online record (this was the largest category at 42.7%). The catalog errors were due to incorrect truncation during search (9.5%), analytics (16.4%), punctuation (21.6%), file too large (24.1%), and corporate word order (28.4%). Most of these cataloging errors were due to lack of enough access point in the online catalog. His conclusions were that keyword and Boolean searches might have reduced the number of catalog errors and that the largest category for “Catalog Failure” resulted from the patrons’ failure to search the catalog at all. The present study augments Seaman's findings by examining "Catalog Failure" in a different library setting and in a keyword-Boolean environment. The current research also offers a chance to evaluate changes in searchers' sophistication over the intervening ten years.

Seaman’s “An examination of Unfilled OCLC Lending and Photocopy Requests” and articles by David Everett, Christine Guyonneau, Jo Anne Bell and Susan Speer, and Sue Medina and Linda Thorton touched on the matter of local catalogs, but they dealt more with local bibliographic verification and failure on the lending library’s part to supply an item. A review of ILL literature indicates that the present study and others which include and examination of patron behavior are needed.

We decided to look at the 770 cancelled ILL requests for the first quarter of 1998, of which 300 were locally held, or “UF owns.” Each request was analyzed for the following:

From the list of cancelled ILL requests for the first quarter of 1998, 300 were identified as locally held, or “UF owns.” A student assistant made a copy, front and back, of each cancelled request. It is the custom of the ILL office to make notes on the back of each request to reflect its history and any deviation from regular ILL procedure. Thus the cards yielded a wealth of information which was supplemented by interviews from some of the ILL staff.

The copies of the cancelled requests were divided among the researchers. Another student assistant searched the local on-line catalog, LUIS, for the bibliographic record that matched the item requested and made copies of each record. The researchers then analyzed the information to supply the figures needed for the specified categories of information. We inserted this information into an Excel spread sheet, which allowed us to manipulate the data.

We found that 17% of the requests for the this time period were cancelled and of those 39% were “UF Owns.” (fig. 1) The patron status mix was reflective of our overall numbers with graduate students using ILL services in greater numbers than other patrons. (fig. 2)

	1998	First Quarter
Total requests	16,921	4,476
Total Cancels	3,286	770
Total "UF Owns"	1,343	300

Total Patrons	198	Total Cancels	300
Faculty	33	Cancels	48
Graduate	123	Cancels	194
Undergraduate	21	Cancels	31
Staff	21	Cancels	27

In opposition to ILL policy, 11% of the patrons knew that UF owned the item but did not wish to use it in its format (microform) or did not want to wait for a recall or a shelf search.

The areas of interest of these patrons were evenly divided between Humanities & Social Sciences and Sciences, with 44 departments in Humanities, 106 in Social Sciences and 150 in Sciences. The branch location of the requested materials echoes this division (fig. 3). The language of the vast majority requested materials was English (271 requests) and a distant second was German (15) followed by French (8), Spanish (3), Czech (2), and Arabic (1).

Architecture and Fine Art	10
Documents	1
Judaica	2
Education	17
Journalism	1
Latin America	3
Legal Information	2
Music	1
West (Main)	100
Total H & SS	137

Health Center	30
Herbarium	1
Science	116
Total Science	147

Special Collections	8
Storage	2
Total Mixed	10

Of the 300 cancelled requests 196 of them were for copies of articles, chapters, or papers and 104 were for the loan of materials. This partition mirrors the fact we average more requests for copies than loans.

The formats of the incoming requests were 59% in the traditional card form and 41% in an electronic form (Email and through First Search). These figures are a little skewed from the formats for all requests submitted in this time period (69% by card and 31% electronically). The more ILL requests that are done online the higher the danger of patrons seeking instant gratification of ordering an item from ILL while in a database rather than checking the local OPAC or actually searching the shelves.

An alphanumeric sorting of Class Numbers did not divulge any helpful information to spot a trouble area since a wide range of Dewey Decimal Classification (DDC) and Library of Congress Classification (LCC) Class Numbers were represented and there were few repeats.

The results were not quite what we expected in that they did not point out as many cataloging errors as anticipated. Several of the categories we examined turned out to be dead ends. The questions we asked about the bibliographic records: Is the titled abbreviated? Does it start with an article? Is the requested item a conference? Has this item ever been declared missing? Does the record have a 210 field? Is the date of the item significant? The overwhelming answer to these questions was NO.

Other lines of inquiry became apparent as a result of this study. We found that the actual physical setup of the main and branch libraries might be confusing to a patron particularly when it comes to periodicals. Some of the branches shelve periodicals alphabetically by title, while most shelve by call number and date. Thus in the main library (West) periodicals might be four places on one floor depending on call number (DDC or LOC) or by physical state (bound or unbound). The date of an item might also have sent it to off campus storage, which is evident in the LUIS record but not marked on the stacks. The Science Branch (MSL) adds the confusion of another floor to these categories. The authors’ experience with the periodical reference desk and retail convinced them that many patrons take the line of least resistance in a search and believe that if it is not in the first place they look, it must not exist. Bibliographic instruction is difficult to give patrons who are in their last month of their senior year and have never used the library.

An interesting aspect of this study was a look through the various “Where was this cited?” answers that ILL received. ILL requested this information in case the bibliographic record is difficult to find or the results obscure. Twenty-six gave a specific print source and 77 specified an individual database, but 11 references were to umbrella databases like First Search that were as useless as no answer at all. Twenty-seven listed LUIS as a source of information and 8 consulted other catalogs. The most unfortunate references were the 16 that became contenders for the title of “Most Useless.” The answer ‘a book’, ‘a dissertation‘, ‘an article’, or ‘a database’ certainly does not narrow down ILL chances of ever finding the reference. The favorite response was still that small, confined, controlled, and localized reference source: the “World Wide Web.” These last citations are part of a problem in a busy ILL department, in that staff members who wish to aid in the research aspect of the library are forced to act as clerks who regurgitate what little the patron supplies and then have to cancel requests when no bibliographic record or lender is found.

In Seaman’s study of “Online Catalog Failure,” he found that approximately 9% of the borrowing requests for the period studied were for locally-held titles. Our study revealed a smaller percentage, 7%. A striking difference between Seaman’s conclusions and ours was that we found very little of what Seaman calls “catalog failure,” which includes punctuation, analytics, corporate word order, truncation, and “file too large.” We found that the catalog records by and large contained no errors which would prohibit retrieval. One implication is that the quality of catalog records have improved since 1988. Another conclusion is that Seaman was correct when he surmised that the availability of KWB searching would aid patrons’ ability to find items in the catalog.

The fact that 11% of the requests were from patrons who demonstrably knew that the library owned the item was interesting. The people who took the time to submit these requests apparently did not realize that local ILL policies prohibit the borrowing of a title simply to supply a preferred format or to protect the patron from waiting for a recall. The large percentage of such requests points out the need to better educate library users about ILL policies.

More active communication between ILL and Reference staff would be beneficial. When the ILL staff do their work away from the public eye, Reference is usually the conduit between the patron and ILL staff. A better liaison between ILL and Reference would keep staff appraised of upcoming changes and new sources of information. Ongoing communication would help insure that Reference staff remain informed of ILL policies so that they can appropriately guide patrons. We also propose that a handout, to be kept at the reference desk and given out at all library instruction classes, would assist in this colloquy.

Though the catalog record was correct in the majority of cases, some improvement might come from attempting to provide more alternative title access on serial records. MARC fields 222 (Key Title) and 210 (Abbreviated Key Title) should be used to include the ISSN Center’s abbreviated titles whenever possible. Field 246 (Varying Form of Title), which may be used for titles which are not standardized and which do not appear on the item, is a repeatable field appropriate for abbreviated or spelled-out variations under which a patron might search. All alternative title fields, including 222, 210, and 246, should be indexed as both titles and as keywords. Our advice is to be liberal with the provision of alternative title access.

The results of the study suggest several actions which might reduce requests for locally-held items. These actions can be encapsulated as follows:

This study offers hope that the reasons for ILL requests for locally-held items can be discerned and that steps can be taken to reduce such requests. Communication, between ILL and Reference staff and between librarians and patrons, is key. Careful signage and the inclusion of Interlibrary Loan policies in routine library instruction hold promise for reducing the stress which superfluous requests create for library staff. The real winners will be the patrons, who will be protected from needless paperwork and will be connected with needed library materials more quickly. Maintaining fine communication within the library takes effort. To expend energy in order to improve service to our patrons is always worth the effort.

Bell, Jo Ann, and Susan Speer. “Bibliographic Verification for Interlibrary Loan: Is it Necessary?” College and Research Libraries 49 (1993): 494-500.

Everett, David. “Verification in Interlibrary Loan: A Key to Success?” Library Journal 1 Nov. 1987: 37-40.

Guyonneau, Christine H. “Performance Measurements for ILL: An Evaluation.” Journal of Interlibrary Loan & Information Supply 3.3 (1993): 101-26.

Medina, Sue O’Neal, and Linda Thornton. “Cannot Supply: An Examination of Interlibrary Loan Requests Which Could Not Be Filled by Members of the Network of Alabama Academic Libraries.” Journal of Interlibrary Loan, Document Delivery & Information Supply 6.4 (1996): 11-33.

Seaman, Scott. “An Examination of Unfilled OCLC Lending and Photocopy Requests.” Information Technology and Libraries 11 (1992): 229-35.

---. “Online Catalog Failure as Reflected Through Interlibrary Loan Error Requests.” College and Research Libraries 53 (1992): 113-20.

Article copies available from the Haworth Document Delivery Service: 1-800-HAWORTH.
E-mail address getinfo@haworth.com.

Elsevier	3
Jstore	1
WWW	2
Total Online	6