College and Research Libraries STELLA FRANK MOSBORG Measuring Circulation Desk Activities Using a Random Alarm Mechanism Librarians frequently use circulation statistics, collection size , and other data to help justify staffing needs, but they lack statistics on the percentage of time their staff actually spends on various activities. Such data have been almost impossible to obtain for complex service units that have a multiplicity of functions, lack of control over frequency and quantity of patron requests, and extended hours of staffing. Recent technological developments combined with work sampling techniques now make studies of such units feasible. This paper reports ·a methodology devised to gather meaningful management data relating to circulation desk activity. INDUSTRY HAS USED "observed work sam- pling" as an evaluative tool. for many years, but Poage, in 1960, first reported using it with library employees . 1 That same year Russell L. Ackoff developed a small, self- powered electronic Random Alarm Mecha- nism (RAM) that made self-observation at random times a practical technique , 2 and he and Martin used RAMs to study chemists' and physicists' use of written scientific material. 3 Between July 1966 and June 1968, Orr and others worked on a project to develop better methods for objectively col- lecting data for planning and improving biomedical libraries and the biomedical in- formation complex. 4 Among the techniques they tested was · ~random time sampling" us- ing a RAM. Orr recommended it as an "ele- gant, powerful technique of great flexibility . . . that . . . makes practical time allocation studies of a qual_ity not attainable by the di- Stella Frank Mosborg is assistant circulation librarian and assistant professor of library ad- ministration, University of Illinois at Urbana- Champaign. The author acknowledges the as- sistance of Arlene johnson , library technical assistant, and of Professor james L . Divilbiss , University of Illinois Graduate School of Library Science, who adapted his RAM for the study, de- signed the tag, and critiqued the study design. ary method. "s In 1971 Spencer used a RAM to study a single activity, 6 and in 1977 Div- ilbiss and Self used one to analyze the activities of an individual librarian. 7 Though it is ideally suited for studying all of the activities of a library unit, no one to date has reported such. a study. In planning postautomation staff utiliza- tion, circulation librarians at the University of Illinois at Urbana-Champaign needed in- formation on the percentage of time spent on the manual circulation functions. A search of library literature revealed no stud- ies of a large staff working extended days and weeks for a cyclical period that might be used as a basis for planning, so a longitu- dinal study was undertaken. It was com- pleted before any aspect of automation was implemented ; this was done so that com- parisons can be made with a postautomation study. STUDY DESIGN AND PROCEDURE The site of the study was the main cir- culation desk, staffed by twenty-three non- professional employees (one full-time and four part-time supervisors, four full-time and eight part-time civil service clerks, and six hourly student clerks). When school was in session, the circulation desk was staffed ninety-two hours seven days per week; dur- I 437 438 I College & Research Libraries • September 1980 ing vacations and between semesters (in- terims), it was staffed forty-eight hours five and a half days a week. All of the em- ployees except the full-time supervisor worked one night and one weekend shift each week school was in session. Since ev- ery member of the staff was trained to per- form the basic circulation functions, de- pending on the time of day, staffing, and patron queues, all employees were asked to participate in the study. The following categories, basic to the ·op- eration of the circulation desk and to most other library circulation units , were selected for analysis: Charging/Renewing Discharging Filing Book Sorting Patron Interaction Processing Overdues Processing Holds Processing Snags Idle Other Miscellaneous activities were recorded with a diary entry under "Other"; similar activi- ties recorded in this category were later grouped for analysis. Sample size was established by estimating the function requiring the most staff time and using a Nomograph Calculation Table8 to determine the total sample needed. By estimating that no activity required more than 20 percent of the staffs time, a sample size of 700 was required for a confidence level of 95 percent ± 3. Because the data were to be further analyzed by employee level, time of day, day of week, and week of semester, a total sample of 4,000 was selected as reasonable and sufficient to produce meaningful results. Circulation in a university library is cy- clical, so the dates selected for the study (January 23 to June 10, 1978) included a semester and an interim. The study was conducted on random days throughout the cycle so the data would be representative. Based on an estimate of 100 tabulations (samples) per day , forty-four days were selected with the aid of a random numbers table. Four days were allotted for a pilot study; the first day's test ran so smoothly, however, that only one day was used. Two additional days, one in the semester and one in the interim, were omitted through supervisory error. A RAM seemed to be the ideal tool for selecting the random time of day. A trained observer tested the possibility of simul- taneously recording every staff member's activity when the RAM beeped. In addition to being unable to see all of the employees at a given time, the observer could not cor- rectly identify everyone's activity without asking and, thereby, intruding. (Some observed activities could be classified under more than one function; e.g., searching in the card file could be for the purpose of filing, discharging, renewing, processing a hold or snag, or providing patron assis- tance.) Because only the employee would know under which category an activity be- longed, it was evident that self-observation was the only technique that would produce useful data. The close proximity of the employees in the desk area made the use of individual RAMs inadvisable. Employees might re- spond to a colleague's beep, while the annoyance of many RAMs beeping at var- ious times on a given day posed a potential staff cooperation problem . It was resolved by adapting a single RAM so that its beep was broadcast through a small speaker over the entire circulation area. It was easily· heard simultaneously by all of the em- ployees, and though some complained about its pitch, volume , and length of beep, it served its purpose well. It was preset elec- tronically so it would average one beep ev- ery forty minutes over the total study; no resetting or manipulation by the staff was required. A lightweight tag was designed as the de- vice on which the data were recorded. Cate- gories were listed on the left half of the tag, and the right half was used for tabulations. Space was provided for the date , time of issuance and return, and initials of super- visor issuing and receiving tags. Lines were provided on the back of the tag for itemiz- ing activities tabulated under " Other. " A screw was placed through a hole at the top of the tag and screwed to a nut that had been ~elded to a small clip ; the tag was clipped to a blouse, shirt, or belt so it was readily accessible when the RAM sounded. Each time the RAM beeped, a tick mark was entered beside the activity; only tick marks by " Other" were explained on the back This fast, simple procedure insured prompt employee response and minimum intrusion on public service. Shift supervisors kept a daily log of the times the RAM sounded. A sheet of lined paper with the day's date was attached to a clipboard on which a digital watch was mounted. When the RAM beeped, the ex- act time was read from the watch and re- corded. This provided a .count of the beeps and served as a validity check: the number of beeps recorded for a given time period was checked daily against the tick marks re- corded for the same time period on each employee's tag. The RAM continued to beep while em- ployees were away from the area for breaks, meals , and special assignments, so a method was devised for keeping their tags accurate. When an employee left the desk area, the tag was given to the shift supervisor, who attached it to the daily log clipboard. If the RAM sounded in the employee's absence, the supervisor recorded a tick mark under " Other. " When the employee retumed , the tag was retrieved and the daily log checked to determine if the RAM had beeped. If it had , the employee completed the diary en- try on the back of the tag. Staff motivation and cooperation were essential if the data were to be valid, and special efforts were made to achieve their support . The purpose and plan were dis- cussed first with the supervisors and then with all of the clerks. These small group meetings , in which everyone discussed , questioned , challenged, and offered sugges- tions , resulted in a detailed itemization of activities , which was typed and posted in the desk area. These meetings clarified for each employee the categories under which activities should be recorded ; they also allowed everyone to help design that part of the study with which they would be most intimately involved. They became quasi- designers of the study-not mere subjects. They saw its value and purpose, were in- terested in its results, were willing to par- ticipate, and were motivated to tolerate the annoying interruptions of the RAM and to conscientiously record the data. To achieve total cooperation, every con- cern raised had to be sa.tisfactorily resolved. Two major concerns were repeatedly vocal- ized: anonymity, and the meaning of "Idle." Circulation Desk Activities I 439 The latter concern was easily resolved by narrowly defining the category to mean "on- the-job idleness, personal telephone calls and conversations." Coffee and meal breaks, to which the employees were entitled, were recorded as diary entries under " Other." Preserving anonymity and obtaining valid data at the same time was a larger problem. The staff was asked to initial their tags for identification of employee level and for validity checks. They feared repercussions for persons recording "negative" data (e.g., excessive idleness) and thought some might be tempted to skew their tabulations to "acceptable" activities if initials were re- quired. It was agreed that the staff need not initial tags so long as validity checks showed no discrepancies; if problems developed, everyone would be asked to initial tags until the problems were resolved. (No problems occurred.) Marks were made on the tags to provide data by employee level: yellow stripe for shift supervisor, green stripe for non-shift supervisor, and red X for stu- dents. Though it was possible to identify the shift supervisors and some students by comparing dates and schedules, no such comparisons were made. A high-level employee who was not par- ticipating in the study but who arrived early each day for work was given the list of preselected random days. On weekday test days , she notified the shift supervisor as soon as he or she arrived for work. Weekend supervisors, who were students in the Graduate School of Library Science, were notified privately on Friday afternoons when a test day was scheduled for that weekend , and as budding professional librarians they lived up to this confidence entrusted to them. Except for the weekend supervisor, the staff was never aware which days were study days until they arrived for ·work. The following list of responsibilities and protocols was typed and posted in tbe desk area adjacent to the activities: Employees 1. Pick up your tag from the supervisor when you arrive for work; check that the date, time of issuance, and supervisor's initials are on the tag. 2. Attach the tag to your clothing and keep it visible at all times . 440 I College & Research Libraries • September 1980 3. Record your activity immediately when the RAM beeps; if you are uncertain where the activity belongs, check the activities list; if still uncertain, check "Other" and diary it on the back of your tag. 4. Leave your tag with the supervisor if you must leave the test area; retrieve your own tag when you return and check to see if the RAM beeped while you were absent. If it did, make a diary entry regarding the na- ture of your absence, e. g.: 10:00-10:15 Coffee break 11:15-11:35 Searching Deck 6 for overdues 5. Hand your tag to the supervisor when you complete your work shift. Shift Supervisors 1. If opening the desk, immediately switch on the RAM. 2. Prepare the daily log sheet and attach it to the clipboard. Keep the clipboard with you at all times. 3. Notify each employee arriving for work that the day is a study day. 4. Issue each employee a tag on which you have recorded the date, time of issuance (read digital watch), and your · initials. Make any identifying marks required: yellow stripe, green stripe, or red X. 5. Record the time of each beep (read digital watch) on the log sheet; record your own activity on your tag; record the beep beside "Other" on any tags left by employees with you. 6. On tags returned by employees at the end· of their shift, record the following informa- tion: time returned (read digital watch), your initials. 7. Remove clips and place returned tags in designated box. 8. Attach clips to clean tags so they will be ready for next employee reporting for work. 9. Switch off RAM at end of closing procedures and immediately before locking door for night. A research assistant was hired to process the data. After each study day, he collected the tags and daily log, checked the validity of each tag, recorded the number and level of staff on duty, noted absences and vacan- cies from the posted schedule, tabulated the data on coding sheets, recorded circulation statistics on coding sheets, and prepared tags for the next study day. He keypunched the data from the coding sheets and also noted any special information on the log sheets that would be helpful in later inter- pretation of the data (e.g., spring break; three-hour day). Using a data-base system (Data Retrieval System [DRS] by Aeronautical Research Associates, Princeton, New Jersey) the data were processed on a General Automation 1830 computer. On-line editing allowed grouping and analyzing data in the "Other" category. The coding instructions were verified by running the data from the pilot study through the computer; no changes were needed. RESULTS A total of 4,304 tabulations was made during the study; the average of 105 per day compared favorably with the projected number of 100. The formula shown as equa- tion 1 was used for computing absolute accuracy. At the 95 percent confidence level, work activities of the total staff had an absolute accuracy between ± 0.4 percent (Snags) and ± 1.1 percent (Patron Interac- tion). The RAM, which had been preset to average one beep every 40 minutes, actual- ly averaged one every 39.4 minutes. Its average daily frequency varied from 27.9 to 66.9 minutes, while its real frequency varied from twice in one minute to once in three hours, twenty-two minutes. Daily validity checks revealed that only rarely was a tick mark missed . a ~ ±1.96 V9- [I] where: a = absolute accuracy expressed as a decimal fraction p = cumulative percentage expressed as a decimal fraction N = total number of tabulations While a few members of the staff expressed mild annoyance with the · RAM as the study progressed, no one thought this affected the reliability of their data. They found the tag easy to use, the categories clear and logical, and the procedure for re- cording the activity so quick it did not in- terfere with good public service. When an activity that belonged in a specific category was occasionally entered under "Other," the research assistant coded it in its proper place. Comments and questions were dis- cussed in the regular weekly staff meetings. After the study was completed, four addi- tional categories were created, three by grouping activities recorded under "Other" and one by recording absenteeism and· staff vacancies from the weekly schedules . The three categories and their definitions were: Supervisory: Opening and closing proce- dures, staff training, and evaluations . Administration: Routine activities essen- tial to unit operations (e.g. , reading corre- spondence, sharpening pencils). Personal: Coffee breaks, meal breaks, visits to washroom. The fourth category, "Absent," was created because of its impact upon the performance of clerical functions by supervisors. Patron Interaction occupied more staff time than any other function (table 1). Twenty-five percent of the staffs total time was spent working with patrons and charg- ing materials, and 40 percent was spent working with the "housekeeping" aspects of circulation (discharging, book sorting , filing, overdues, holds, and snags). Super- visors spent 30 percent of their time on administrative and supervisory functions, compared with 11 percent for the staff as a whole. Supervisors spent less time on most routine clerical functions and more on Patron Interaction and areas were they had verification responsibilities (Overdues, Snags). There is a close correlation between the total staff percentages and the clerks ' percentages, but this is to be expected since Circulation Desk Activities I 441 clerks recorded 68 percent of the data. Stu- dents, who constituted 13 percent of the staff, contributed only 7 percent of the data. No· doubt their high rate of absenteeism (26.4 percent) contributed to their low in- put. When the functions are grouped by type, circulation activities occupied 68.5 percent of the staff's time , administration and supervision, 11 percent, and personal activities (Idle, Personal, Absent), 20.5 per- cent. Table 2 shows that activity by week tends to follow the semester cycle. Patron Interac- tion, Charging/Renewing, and Filing were higher in the middle of the semester, when research papers were being written and cir- culation was high. Book Sorting, Discharg- ing, and Overdues Processing were higher at the end of the semester, when materials were being returned. Analysis by day of week (table 3) also re- vealed a cyclic pattern created by both academic patterns and desk work flow poli- cies. Patron Interaction was higher on Thursdays, when students often worked on assignments due the next day or began assignments before the weekend, and on Saturdays, when the general public and visiting scholars came to the library . Dis- charging increased on Wednesdays and Fridays . Filing, accumulated over the weekend, was heaviest on Mondays, and Charging increased on Mondays and Thurs- days, reflecting patron activity before or af- TABLE 1 PERCENTAGE TIME STUDY OF CIRCULATION FUNCTIONS BY EMPLOYEE LEVEL Supervisors Cle rks Students Total Staff Function Count Perce nt Count Perce nt Count Perce nt Count Patron interaction 183 16.9 426 14.6 52 16.5 661 Discharging 100 9.2 493 16.9 43 13.6 636 Filing 55 5.0 394 13.5 57 18.1 506 Charging/renewing 69 6.3 309 10.6 26 8.2 404 Administration 268 24.7 127 4.3 1 0.3 396 Personal 51 4.7 255 8.7 8 2.5 314 Idle 87 8.0 188 6.4 23 7.3 298 Absent 0* 0.0 190 6.5 83 26.4 273 Book sorting 73 6.7 173 5.9 14 4.4 260 Overdues processing 53t 4.8 ll6t 3.9 It 0.3 170t Holds processing 19 1.7 ll5 3.9 1 0.3 135 Other 15 1.3 82 2.8 0 0.0 97 Supervisory 56 5.1 20 0.6 2 0.6 78 Snags 53 4.8 20 0.6 3 0.9 76 -- Total 1,082 2,908 314 4,304 *A supervisor was always on duty to perform the supervisory functions because supervisors cover each other's absences. tOverdues processing is divided between the desk and the circulation office . Pe rce nt 15.4 14.8 ll.8 9.4 9.2 7.3 6.9 6.3 6.0 3.9 3.1 2.3 1.8 1.8 TABLE 2 CIRCULATION FUNCTIONS ANALYZED BY WEEK OF STUDY Study Circulation Functions in Percentage t Week Days/ Tabu- Patron Dis- Chg./ Book Over- t-0 No. Week lation s Inte ract . chg. File Renew Adm . Pe rs. Idle Abs . Sort . du es Holds Other Supvr. Snags 1 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 ~ 2 1 122 14.8 6.6 22 .1 9.6 ll.5 7.4 8.2 3.3 7.4 0.0 4.9 0.8 3.3 0.8 c 3 2 174 15.5 10.3 15.5 14.9 10.9 5.7 7.5 8.0 1.1 2.3 1.1 2.3 1.7 2.9 f 4 1 107 13. 1 4.6 3.7 10.3 7.5 5.6 20.6 4.6 0.9 12. 1 9.3 2.8 3 .7 0.9 ~ 5 1 130 16.9 21.5 14.6 10.0 5.4 13.8 9.2 0.0 0.8 0.8 2.3 0.8 3.1 0.8 c:- 6 1 128 24.2 5.5 ll.7 11.7 10.1 0.8 7.0 12.5 5.5 1.6 0.8 3.1 3.9 1.6 !l:l 7 2 142 18.3 10.6 8.5 14.1 8.5 6.3 9.9 3.5 2.8 4.2 6.3 4.9 1.4 0.7 ~ 8 1 9 22.2 0.0 0.0 0.0 22.2 0.0 0.0 0.0 11.1 11.1 0.0 11.1 11.1 11.1 ... ~ 9 5 557 16.7 14.5 14.0 ll.O 8.1 8.8 9.9 2.5 4. 1 1.3 2.3 . 3.1 1.6 2.2 ~ ~ 10 4 439 19.4 12.5 10.0 12.3 8.9 5.7 5.9 6.4 5.5 3.0 3.6 2.8 1.8 2.3 ~ ll 1 97 24.8 17.5 5.2 5.2 2.1 13.4 5.2 4.1 4.1 8.2 4.1 0.0 2.1 4.1 ~ 12 5 530 17.5 9.1 14.3 10.4 14.7 9.1 3.2 6.2 5.3 3.4 3.6 0.6 1.7 0.9 5-:. 13 3 . 383 14.9 13.3 14.9 12.0 9.7 5.7 6.5 5.7 7.0 3.4 5.0 0.5 1.0 0.3 ~ 14 2 199 12.6 16.6 11.0 11.6 9.0 6.5 5.0 ll.6 7.5 2.0 3.0 0.5 1.0 2.0 ~- 15 4 405 12.3 17.5 13.1 8.6 4.7 6.4 6.2 9.9 10.9 3.7 1.7 0.7 0.7 3.4 ... 16 2 227 10.6 24.7 5.3 4.0 8.4 3 .5 7.0 18.5 9.7 1.8 1.3 2.6 1.3 1.3 17 2 141 8.5 24.1 8.5 2.8 7.1 6.4 9.2 0.0 6.4 12.8 1.4 7.8 0.7 4.3 en ~ 18 3 367 ll.2 21.5 7.6 3.5 12.3 9.0 4.1 3.5 8.4 9.3 1.4 6.5 0.3 1.4 ~ 19 1 147 8.2 19.0 8.2 2.0 13.6 8.8 6.8 8.2 6.8 7.5 8.2 1.4 0.0 1.4 ~ 4,304 ~ ~ Total percentage 15.4 14.8 ll .8 9.4 9.2 7.3 6.9 6.3 6.0 3.9 3.1 2.3 1.8 1.8 .., ....... ~ TABLE 3 c CIRC ULATON F UNCTIONS ANALYZED BY DAY OF WEEK Day Study Circulation Functions in Percen tage of Days/ Tabu- Patron Di s- Chg./ Book Ove r- Week Week lations Inte ract. chg. File Renew Adm . Pers . Idle Abs. Sort. du es Holds Other Supvr. Snags s 4 298 15.4 13.0 9.3 7.7 1.3 8.7 15.4 11.4 4.6 1.6 0.6 3.6 1.6 5.0 M 5 609 13.7 10.6 17.2 14.1 ll .O 8.7 6.2' 2.6 5.0 3.6 2.9 1.6 1.6 0.6 T 10 1,256 15.4 15.1 12.3 8.6 10.2 7.6 6 .7 5.8 6.2 2.8 4.2 2.3 0.9 1.1 w 5 516 14.5 21.1 8. 1 7.7 7.5 6.3 4.6 5.0 8.7 6.3 2.1 5 .6 0.7 1.1 T 6 657 16.4 12.4 12.0 11.7 10.9 4.8 4.4 8.2 8.0 3.5 2.1 1.2 1.8 2.1 F 5 622 14.7 16.8 12.2 6.9 10.7 8.0 6.2 7.8 4.5 3.5 3.0 1.4 1.9 1.7 s 6 346 17.9 13.2 6.0 7.5 5.2 6.9 10.6 5.7 3.1 8.3 5.2 3.7 2.8 3.1 4,304 Total percentage 15.4 14.8 ll .8 9.4 9.2 7.3 6.9 6.3 6.0 3.9 3.1 2.3 1.8 1.8 ·ter the weekend. Idleness was highest on weekends, when fewer patrons were in the library . . Absenteeism , primarily a student problem, was greatest on Sundays and Thursdays , the days before many assign- ments were due. Faculty Overdues were processed only on weekends, and that was reflected in the data. Special assignments were usually a part of weekend activity when routine functions required less time for both clerks (Holds , Other) and super- visors (Snags). Early in the study it became obvious that it was not possible to identify the time of day an activity was being performed by per- sons who worked full time or during a meal hour. Beginning with the eighth day of the study, the staff drew vertical lines on their tags representing 12 noon and 5 p.m. and placed tick marks in the appropriate sec- tions. Table 4 reflects activity by time of day for the last thirty-three days of the study. Analysis by employee level was ap- plied only to supervisors and clerks because there were only 257 student tabulations dur- ing that period; total percentages, however, include student data. Almost 50 percent of the days' total activity occurred in the after- noons; mornings were slightly busier than evenings, but the type of work done in the mornings differed from that in the evenings. Patron Interaction was negligible in the mornings and quite heavy in the after- Circulation Desk Activities I 443 .noons-60 percent more than in the eve- nings. Discharging was the primary morning activity and second most frequent activity in the afternoons. Though all of the previous day's circulation was filed each morning, Monday through Friday, Filing, was not the primary morning function. This was a sur- prise discovery. CONCLUSIONS The technique of random time sampling, using a RAM, is practical for evaluating all of the activities of a large staff working ex- tended days and weeks. The RAM is an ideal instrument for selecting random times; one that is electronically preset, so it need not be touched by the study participants once it has been activated, has the added advantages of high reliability and minimum intrusion. A self-observation study will fail unless the staff willingly participates. Total honesty in explaining the purpose and plan, scrupu- lous care in following agreed-upon proce- dures, prompt resolution of participants' concerns, and complete trust between administrator and participants are required . When information is being sought that might be interpreted negatively, few people will honestly record "negative" data unless they can be guaranteed anonymity. Study procedures may need to be modified in order to provide this guarantee, but it is TABLE 4 CIRCU LATO N F UNCTIONS A NALYZED BY TIME OF DAY Function Patron interaction Discharging Filing Charging/renewing Administration Personal Idle Absent Book sorting Overdues processing ·Holds processing Other Supervisory Snags Total *Based on 913 tabulations. tBased on 2,508 tabulations. Mom . 2.8 2.8 0.4 1.6 9.5 1.3 2.5 0.0 2. 1 2.0 0.3 1.2 1.2 1.5 29.2 Supervisors'* Percentages Aft . Eve . 7.2 6.2 4.7 2.6 0.4 4.2 1.8 2.6 13.9 2.3 1.9 0.2 2.8 1.8 0.0 0.0 3.6 1.6 2.7 0.7 1.4 0.2 1.6 0.0 1.5 0.7 2.0 1.8 - 45.5 24.9 :j:Based on 3,678 tabulations (includes 257 student tabulations). Clerks 't Mom . Perce ntages Aft. Eve . 3.4 7.2 3.6 6.2 7.6 3.8 5. 1 3.6 4.3 2.0 5.5 2.6 2.0 1.5 0.6 2.6 4.2 2.4 1.4 2.6 2. 1 2.3 3.1 2.1 2.5 2.7 1.2 2.5 1.4 0.2 1.0 2.2 0.6 0.6 2.0 0.2 0.2 0.2 0.1 0.2 0.~ 0.0 - 32.0 44.2 23.8 Total StalT:j: Percentages Morn . Aft . Eve. 3.1 7.3 4.5 5.1 6.8 3.6 3.8 3.1 4.3 1.8 4.5 2.6 3.7 4.5 1.0 2.1 3.4 1.8 1.6 2.6 2.0 1.8 3.2 2.4 2.4 2.8 1.3 2.2 1.6 0.3 0.8 1.8 0.5 0.7 1.8 0.2 0.5 0.5 0.2 0.5 0.8 0.5 30.1 44.7 25.2 444 I College & Research Libraries • September 1980 worth the effort when highly reliable data result. Self-observation is the best tech- nique for complex staffing studies where · an activity can be classified in more than one category. Library administrators now have available the tools and technique to obtain valuable information that was almost impossible to collect a few years ago. Staffing studies of this type isolate problems, confi~ or refute staffing theories, and provide conclusive evidence to support staffing needs. Admin- istrators need not be restricted to staffing studies, for with the aid of computers, RAMs, and work sampling techniques, they can evaluate a variety of library operations to make them more cost effective. REFERENCES 1. Scott Tabor Poage, "Work Sampling in Library Administration," Library Quarterly 30:213-18 (July 1960). 2. Richard H. Orr, "The Scientist as an Informa- tion Processor: A Conceptual Model Illus- trated with Data on Variables Related to Li- brary Utilization," in Communication among Scientists and Engineers (Lexington, Mass. : Heath, 1970), p.181. 3. Miles W. Martin and Russell L. Ackoff, "The Dissemination and Use of Recorded Scientific Information," Management Science 9:322-36 (Jan. 1963). 4. Richard H. Orr and others , "Development of Methodologic Tools for Planning and Manag- ing Library Services : 2. Measuring a Library's Capability for Providing Documents," Bulletin of the Medical Library Association 56:241--67 (July 1968). 5. RiChard H. Orr, "The Scientist as an Informa- tion Processor," p.185. 6. Carol C. Spencer, "Random Time Sampling with Self-Observation for Library Cost Stud- ies: Unit Costs of Interlibrary Loans and Photocopies at a Regional Medical Library," Journal of the American Society for Informa- tion Science 22:153-60 (May-June 1971). 7. J. L. Divilbiss and Phyllis C. Self, "Work Analysis by Random Sampling," Bulletin of the Medical Library Association 66:19-23 (Jan. 1978). 8. Bertrand L. Hansen, Work Sampling for Mod- ern Management (Englewood Cliffs , N . J.: Prentice-Hall, 1960), p.50. ADDITIONAL SOURCES ON THE TOPIC OF WORK ANALYSIS Divilbiss , J. L. "Time Analysis," in his The Tools of Systems Analysis, forthcoming. 22p. Drott, M. Carl. "Random Sampling: A Tool for Library Research," College & Research Librar- ies 30:119-25 (March 1969). Evans, G. Edward. Management Techniques for Librarians. New York: Academic Press , 1976. 276p. Goodell, John S. Libraries and Work Sampling. Littleton , Colo. : Libraries Unlimited , 1975. 60p . Hines, Theodore C . "Circulation Systems," Li- brary Journal 86:4240-43 (Dec. 15, 1961). Holmstrom , Bengt. "Work Simplification in Swedish Public Libraries ," Library Journal 88:4312-15 (Nov. 15, 1963). Pings, Vern M. " Development of Quantitative Assessment of Medical Libraries ," College & Research Libraries 29:373-80 (Sept. 1968). Pizer, Irwin H., and Cain, Alexander M. "Objec- tive Tests of Library Performance," Special Libraries 59:704-11 (Nov. 1968). Classics of the Past, seldom available until now, in Beautifully Designed New Editions THE MIND OF MAN SERIES: EXPLORATIONS THROUGH THE PAST AS AN AID TO REDEFINING PRESENT VALVES~ JONATHAN SWIFT: A Tale of a Tub & The Battle of the Books. From the First Edition, with Foreword and Notes by Robert Folkenflik, University of California, Irvine. With 12 double-spread plates, William Hogarth's series, Industry and Idleness. 0-934710-00-7 $25 DESIDERIUS ERASMUS: The Praise ~ ~ ofFolly. The John Wilson translation, fAA ~t . with Foreword and Notes by Harvey L""'.,,~,, .,,._,,.,.... Gross, State' University of New York at Stony Brook. With 22 sketches by Hans Holbein from the margins of a Froben edition copy of 1515. 0-934710-01-5 $20 RALPH WALDO EMERSON: Represen- tative Men. Emerson's insights into Plato, Sweden borg, Montaigne, Shakespeare, Napoleon, and Goethe. With Foreword and Notes by Myron Simon, University of California, Irvine. Including 7 full page portrait engravings. 0-934710-02-3 $25 1YPOGRAPHICALLY DESIGNED FOR EASE OF READING, PRINTED ON ACID-FREE WARM WHITE PERMALIFE TEXT, CASE BOUND IN FULL, QUALI1Y CLOTH AND SLIPCASED. Order Direct Only 10% discount after first copy on Standing Orders. Joseph Simon/publisher BOX 4071, MAliBU, CA 90265 THE APPROVAL PROGRAM THAT'S RIGHT FOR ANY SIZE ACADEMIC LIBRARY. Baker & Taylor offers college and university libraries an outstanding acquisitions service that provides for the automatic shipment of new books according to each library's individual selection criteria. Baker & Taylor's Approval Program is right for university libraries, small and medium sized college libraries, research oriented libraries-in fact. it is right for libraries of all types. Over the years it has developed to be able to serve many needs-even where selectivity on a small scale is important. If desired, it can be shared between libraries coop- erating in their acquisitions efforts, and it can accommodate publisher as well as subject profiles. Today's Approval Program is a highly evolved version of the familiar standing order plan of the past. It has emerged as a very effective tool for devel- oping and maintaining collections in all types of college and university libraries. The Approval Program is an individualized service joining the library and the bookseller in a partnership to better serve the needs of collection development. Read how it works-and how the ser- vice can make a significant contribution to your ' library. CUSTOMIZED PROFILES Each library establishes its own profile by selecting !rom a com- prehensive list ol subjects in those areas in which coverage is desired. This list ol subjects contains over 4.400 individual terms arranged in 126 subject lields and is regularly updated to adapt to new areas ol study and research. The profile is fur- ther relined to reflect each library's collection needs by choosing appropriate non-sub- ject mod!1iers- for example. academic level. publisher. physical format. This profile becm:nes a record ol part ol the library's collection develop- ment policy. Of course. the library can change its prolile at any time to reflect changing needs. GREATER ECONOMY Baker & Taylor can help libraries save money by cutting down on clerical costs such as order typing. The Approval Program frees library staff members from routine func- tions, and allows you to devote your valuable time and talents to other important projects. GREATER CONTROL We maintain close contact with thousands of publishers, moni- toring each one to provide comprehensive coverage o f new titles. The result: many par- ticipants in the Baker & Taylor Approval Program report hav- ing received books which they would not otherwise have known about. The Approval Program is recognized as an effective vehicle for learning about numerous scholarly titles which never appear in review media. BETTER SELECTION Libraries receive automatic book shipments and can make selection decisions from the books themselves, with full return privileges. of course. GENEROUS DISCOUNTS The streamlined efficiency of our Approval Program. plus sound business management based on years of experience serving the academic commu- nity, enable us to provide dis- count plans that are the best in the field. Each customer is offered a choice between a flat 13% discount or a sliding 30- 10- net discount scale. Plus. books are delivered to you postage- paid and there are no shipping or handling fees. ONGOING SERVICE The same academic library professionals who initially present and explain Baker & Taylor's Approval Program continue to service your library. They will, for example, monitor returns to alert you to possible profile revisions and provide customized management reports to aid in your budget planning. Baker & Taylor's Approval Pro- gram serves more college and university libraries than any other. It is also the oldest contin- uing Approval Program. Our staff of 15 professional librari- ans have years of Approval Program as well ds academic library experience. We otter an illustrated brochure which describes Baker Be Taylor's Approval Program in complete detail. To acquire a copy, please use the coupon below. Or call this toll-tree number: 800-526-3844 . r--------, YES! Tell me more about Baker & Taylor's Approval Program D Send me your brochure which describes in detail the Approval Program you offer. D Have a representative contact me. Name and Tille Ubrary street City State Zip a The Baker &TaylorCoutpany • The Librarian's Library Eastern Division 50 Kirb y Avenue. Somerville. New Jersey 08876 Tel. 20 1-722-8000 Southern Division Commerce. Georgia 30599 Tel. 404 -335-5000 • U:idwestern Division Gladiola Avenue. M om en ce. lllinois 60954 Tel. 815-4 72-2444 • Western Division 380 Edison Way. Reno. Nevada 89564 Tel. 702-786-6700 Gflistorical ~b