ACRL News Issue (B) of College & Research Libraries 274 / C b R L N ews W orkflow a rra n g em en ts a n d th eir e ffect on d isch a rg e a ccu ra cy By P a t W eaver-M eyers a n d K en n eth W. P ea rso n Access Services D epartment Head Circulation Supervisor University of Oklahoma Libraries University of Oklahoma Libraries The distraction factor and discharge record error. E very circulation librarian has dealt repeatedly with the problem of patrons claiming they have al­ ready returned a book that they have been billed for. Arbitrating such a dispute is dependent on the library staff’s confidence in the accuracy of the dis­ charge records. Many things can affect the error rate during the processing of returned materials: worker alertness, terminal wand misreads, work­ load, training, etc. In analyzing a fluctuation in er­ ror rate, Bizzell Memorial Library at the Univer­ sity of Oklahoma discovered the im portant role furniture arrangem ent can play in workflow. Background Bizzell Memorial Library circulation operations are automated on a DataPhase ALIS II system. Discharge is completed by wanding in the OCR number of the book and responding to the screen prom pts. Annual circulation is approxim ately 158,000. About three years ago, the circulation de­ partm ent incorporated monthly monitoring of the discharge error rate into their regular statistical procedures. Originally, this was done as a quality control measure. Based on experiences in billing, we sus­ pected that some problems were occurring in dis­ charge which allowed more than a reasonable num ber of items to get back to the shelf w ithout be­ ing cancelled. At that time, patron’s claims about book return and the number of bills issued was the only resource of information about possible prob­ lems. The following is a description of the proce­ dure which evolved over the course of a year and was designed to provide the additional needed in­ formation. A sampling of approximately two hundred items are isolated from the sort area where discharge itens are routed. Selection of these items is also at randomized times to ensure that the effects of dif­ fe re n t sta ff sch ed u lin g are also governed by chance. These items are then checked against the computer records to determine if discharge of each item has actually occurred. Those items that are found to be incompletely processed make up the percent error. It was initially presumed that 100% accuracy was an admirable goal, but perhaps unrealistic. However, considering the nature of public rela­ tions and the effect discharge errors can have on a p atro n ’s opinion of the library’s competency, a high standard was the goal. Therefore, a less than 1 % error rate was designated as acceptable. It is im portant here to stop and think about the practi­ cal significance of a 1 % error rate. If you cancel 200,000 items per year, for example, a 1 % error rate will allow 2,000 items to be processed incor­ rectly. That could be a potential 2,000 patrons all with a jaded opinion of the library’s ability to pro­ cess materials accurately. Also, it means an associ­ ated time and effort must be spent searching the stacks for items claimed returned. So, in reality, even a 99 % accuracy rate, which sounds very strin- A pril 1986 / 275 Before the change, the terminal operator was clearly visible to patrons waiting at the circulation desk. After the move, the operator is screened from patrons queuing up fo r service. 276 / C & R L N ew s gent, remains costly in terms of staff tim e and li­ rary reputation. Brainstorming Once we began regular m onitoring of the error ate, fluctuations w arned the billing staff to be w a re of a problem m o n th . F or exam ple, one onth the error rate soared to 8 % . It was theorized h a t a com pletely unprocessed booktruck was assed through to the sort room by accident. Since he m onitoring discovered the problem , the billing taff had more confidence in arb itratin g in favor of he patron who claim ed to have returned books uring this period, even if the items were not found n a search of the stacks. As the m onitoring progressed, it became obvious h a t our goal of a consistent error rate of less than % was not being reached. Repeatedly, staff was etrained to stam p and w and one-book-at-a-tim e nstead of processing groups of books together to void mistakes. Also, procedure guidelines were osted on the discharge term inal. The circulation upervisor tested each student worker to ensure heir m ethod was appropriate. However, the fluc­ uations still continued. After repeated review of the problem , a subcom­ ittee of staff members brainstorm ed the prob­ em. The com mittee theorized th a t the location of he discharge term inal could be contributing to the roblem . At th at tim e, the furniture arrangem ent laced the discharge term inal in view of patrons as hey queued up to the desk for service. In th a t posi­ ion, any im patient patron could usually get the a t­ ention of the term inal operator and force them to nterrupt their processing and provide additional ssistance. This usually m eant th a t a small stack of ooks could be left unattended on the desk and ven forgotten if the p atron crush prevented the orker from returning to the discharge term inal efore their shift ended. Originally it was assumed th a t the operator’s bility to provide back-up assistance was a valu­ b r a m t p t s t d i t 1 r i a p s t t m l t p p t t t i a b e w b a able asset—hence the location of the term inal. Af­ e r re p e a te d ex am in atio n of th e e n tire process hough, th a t advantage was deemed less im portant h an discharge accuracy. D ue to the subcom m ittee’s recom m endation, he discharge term inal was shifted to a new loca­ ion. In the new location the p a tro n ’s view of the perator was completely blocked. This prevented atrons from in te rru p tin g the o p erato r and al­ owed the workers to devote their complete atten ­ ion to the discharge patron. Conclusion T able I com pares the error rate for a three- onth period before and after the term inal was re­ ocated. As can be seen, the change in error rate rom 1.8% to 1.0% in November, for example, is ot a large difference. W hen it is m ultiplied by the um ber of books discharged in a m onth, however, he practical significance becomes obvious. Even he small differences detected represent a 357-book iffe re n c e in those m a te ria ls p ro je c te d to go hrough the discharge process w ith some mistake. he consequent staff searching and patron com ­ laints are also proportionately reduced. Although he num ber of items discharged in 1985/86 is down lightly from the previous year, the difference in er­ or rates w ould still account for 332 more mistakes ad the num ber of items processed been the same. The arrangem ent of furniture is just p a rt of an rgonomic approach to productivity. In this situa­ ion, it also appears to be an integral com ponent of ccuracy as well. C om bining a q uality control easure such as discharge error sam pling w ith an nalytic review of the procedures can result in ractically significant improvements. This review roved th a t even basic arrangem ent of w orksta­ ions should be included in analysis, as they may ave a m easurable im pact on overall efficiency. In the current clim ate of perform ance m easure­ ent and library quality measures, this analysis oints to a definite need to include circulation t t t t t o p l t m l f n n t t d t T p t s r h e t a m a p p t h m p TABLE 1 ERROR RATES BEFORE AND AFTER TERMINAL RELOCA TIO N 1984/85 Books Estim ated M onth % Error Rate Discharged Errors Oct 1.6 X 8,408 = 134 Nov 1.8 X 11,008 = 198 Dec 1.9 X 17,726 = 337 total 669 1985/86 Books Estim ated M onth % Error Rate Discharged Errors Oct :6 X 7,823 = 48 Nov 1.0 X 10,688 = 107 Dec 1.0 X 15,650 = 157 total 312 April 1986 / 277 functions. The circulation desk of any library is a focal point of activity and may be the only contact some patrons have with library staff. As a result, their opinion about the services rendered at the cir­ culation desk may dominate their overall assess­ ment of the library’s quality. Therefore, it is essen­ tial that circulation procedures are monitored both r quality and efficiency and that any set of per­ rmance measures which claims to assess library erformance cannot be considered comprehensive ithout some measure of these services, which are tensively patron interactive. ■ ■ fo fo p w in A C R L P u b l i c a t i o n s i n L i b r a r i a n s h i p n e e d s y o u r m a n u s c r i p t s ACRL Publications in Librarianship, a series de­ acknowledged and read by at least two members of voted to scholarly monographs in academic librari­ the editorial board. anship, has issued over 40 volumes since 1952. Manuscripts related to the following topics are Many of these studies represent distinguished con­ particularly welcome: administration of college li­ tributions to the history, theory, and practice of ac­ braries; computer applications; collection assess­ ademic librarianship. ment; management of reference services; self- The editorial board invites authors of book- directed studies; and staff development. length studies to submit manuscripts for review. Relevance, lasting value, and superior writing Proposals for manuscripts in progress are welcome describe the manuscripts that we are seeking. Ad­ and should be accompanied by an outline of the dress inquiries and proposals for publication to: Ar­ contents. The submission of dissertations is encour­ thur P. Young, Dean of University Libraries, Uni­ aged; however, many may require substantial revi­ versity of Rhode Island, Kingston, RI 02881-0803; sion by the author (see the guidelines for converting (401) 792-2666. ■ ■ a dissertation to a book below). Manuscripts will be H o w t o c o n v e r t a d i s s e r t a t i o n o r t h e s i s t o a b o o k Editor’s note: These are guidelines adopted by the ACRL Publications in Librarianship Series Edito­ rial Board. A manuscript designed to communicate mastery of the research process to an examining committee frequently fails to satisfy the requirements of a publisher. Authors submitting theses and disserta­ tions to a publisher may anticipate requests for ex­ tensive modifications of their manuscript if it is ac­ cepted for publication. Indeed, some effort at revision prior to submission will enhance the likeli­ hood of acceptance. A book is addressed to an audience that is very different from the audience to whom a thesis, dis­ sertation, or other research report is addressed. These differences include level of interest, prior knowledge of the subject, and objectives in reading the work. Major revisions are usually necessary, even to the most effective works. Revisions which are often required include dele­ tions, reorganization, and the writing of additional material. Some examples: •T he style of a dissertation frequently requires the repetition of material from section to section. In many cases this redundancy can be eliminated. Tables often should be deleted or converted into an explanatory narrative. •M any of the fine points concerning prior re­ search or methodology on the subject should be placed in appendices or footnotes. •Abstractions must be carefully related to the concrete world through more extensive interpreta­ tion than would be necessary in a dissertation or thesis. The editorial board of ACRL Publications in Li­ brarianship encourages authors wishing to submit theses and dissertations for publication to read the following items prior to submission. Olive Holmes, “Thesis to Rook: What to Get Rid of,” Scholarly Publishing 5 (July 1974):339-49; 6 (October 1974): 40-50. Olive Holmes, “Thesis to Book: What to Do with What Is Left,” Scholarly Publishing 6 (January 1975): 165-76. Constance Greaser, “Improving the Effective­ ness of Research W riting,” Scholarly Publishing 11 (October 1979):61-71. ElsiM. Stainton, “A Bag for Aut Publishing 8 (July 1977):335-45. hors,” Scholarly ■ ■ LIMITED OFFER: $2,335 WORTH OF BOOKS FOR $ 2 5 0 -A C T NOW! Build A Complete Collection Of Indexes Covering The M ost Significant Public Policy Literature From Around The World! PAIS FOREIGN LANGUAGE INDEX You Already Know The PAIS BULLETIN— Now Discover Its Companion Resource, The PAIS FOREIGN LANGUAGE INDEX First 200 New 1986 PFLI Subscriptions For years, thousands of leading libraries have made Come With One-Time-Only Offer: A PAIS BULLETIN the “ standard” indexing resource Complete Set Of Back Volumes For Just $250 covering public policy/business/social science To introduce you to PAIS FOREIGN LANGUAGE literature published in English. Now, discover another INDEX, and help you acquire indexes to public “ standard” from PAIS, and a superb companion policy literature spanning nearly two decades, PAIS resource to the PAIS BULLETIN: PAIS FOREIGN announces a special, limited-time offer. If you’re LANGUAGE INDEX. among the first 200 respondents to order new subscriptions to 1986 PAIS FOREIGN LANGUAGE PAIS FOREIGN LANGUAGE INDEX INDEX (Vol. 16), at the regular annual price, you Gives You International Coverage qualify to also order a complete set of back volumes Of Non-English-Language Materials (Vols. 1-15) for just $250! A $2,335 value at list price, PAIS FOREIGN LANGUAGE INDEX provides access to these volumes are yours at a savings of nearly a stringently edited range of public policy literature in 9O°/o— if you act now\French, German, Italian, Portuguese and Spanish. And its English-language headings and abstract-like Subscribe to PAIS FOREIGN LANGUAGE INDEX, and notes help you locate material quickly and easily. Your order your complete set of value-priced back volumes annual subscription brings you three quarterly issues today! Call toll-free 800-841-1416 (in NY and outside plus an annual bound volume, all with authors index, the continental U.S., call collect 212-869-6186). Or fill for $495 inclusive. in and mail the coupon below.