College and Research Libraries W I L L I A M E . M c G R A T H Measuring Classified Circulation According to Curriculum Circulation statistics can be precise ref ections of library use according to the curriculum. The statistics can help the librarian decide how to allocate the budget to departments. Traditional counts, by department personnel or by broad Dewey or LC classes, are imprecise. An analogy between curriculum and circulation can be constructed by classifying courses in the college catalog (by DC or LC), rearranging the num- bers thus generated by department, and then counting circulation within those groups. The analogy is thus a quantitative measure and a precise reflection of library use according to curriculum. C A N CIRCULATION statistics reflect com- parative use of the library by academic departments? What relationship does use of the library b e a r to the curricu- lum? How does one measure use accord- ing to curriculum instead of by depart- ment personnel? These questions, and their variations, have long concerned the librarian who wants to know which departments are the heavy library users, and which should have the greater number of dol- lars for books. Traditional use counts may not give the answers. For example, some libraries categorize their loans ac- cording to the teaching departments whence the borrower comes. If an indi- vidual borrows ten books, ten loans are credited to his department. Such counts may not accurately reflect use accord- ing to the curriculum. A faculty member or student does not always borrow books which are exclusively related to his de- partment or major. W h e n a physics ma- jor borrows a mathematics book, should the loan reflect use according to the Mr. McGrath is Head Librarian, South Dakota School of Mines and Technology, Rapid City. physics department or according to the mathematics department? Under usual procedure, the loan would be counted for physics. Many such loans would sug- gest that the physics department bor- rows more books than the mathematics department. But if the loans are a re- sult of a mathematics assignment, or in- stigated in some other way by the math- ematics department, or even by a third department, we ought not to give the credit to the physics department. As a parameter to help determine which de- partment should receive the larger fund allotment for books, number of loans to department personnel is interesting, but unreliable. A simple head count of de- partment members and enrollees, with- out a circulation count, might measure the same thing and would be simpler. Other libraries keep count of circula- tion by broad Dewey or L C classifica- tion. These libraries can show that each year so many books were borrowed in the 300's, the 400's, the 500's, the B's, the F s , the Q's, etc. These figures do a fine job of showing how many books were circulated in these broad classifica- tions. But, as with loans to department personnel, the figures may bear little re- / 3 4 7 348 / College b- Research Libraries • September 1968 lationship to the curriculum. The scope of DC or LC classes does not neces- sarily coincide with the scope of a de- partment's curriculum. Even when the broad classifications are broken down in- to more detail—let us say, in the DC classification, by the tens instead of by the hundreds—this still does not give us a true picture of curriculum use. Not all courses offered by the geology depart- ment fall into 550's, and not all books in the 550's may be of interest to the geol- ogy department. Furthermore, a course normally thought of as falling within the scope of one department may be offered by another. At South Dakota School of Mines and Technology, for example, Descriptive Geometry ( D C class 515) is offered by the civil engineering depart- ment, rather than the mathematics de- partment. Nevertheless, we would still like to know how many books borrowed relate to the geology curriculum, the civil en- gineering curriculum, the mathematics curriculum, and so on. Rather than counting heads or volumes in meaning- less broad categories, a precise analogy can be created between circulation and the curriculum. This can be done by classifying a department course much as we do a book: according either to DC or LC. W e then arrange sequentially the classification numbers thus generated, but within the departments. The result- ing numbers, perhaps several hundred, form an analogy; and the circulation pattern therefore can reflect use by de- partment subject alignments rather than by members or enrollment. Any book borrowed within the analogy would thus be counted for that department. In another paper,1 the author used this device to establish an analogy be- tween academic departments and the number of books published annually in the United States. The mechanism is the 1 W i l l i a m McGrath, "Determining and Allocating Book F u n d s for Current D o m e s t i c Buying," College and Research Libraries, XXVII (July 1 9 6 7 ) , 2 6 9 - 7 2 . same. By classifying the courses; by grouping the classification numbers, whenever possible, into spans to em- brace a larger subject scope; and by re- arranging these numbers according to department, we can arrive at a total number of books published or circulated which have a direct relevance to the de- partment concerned. By using the same sequence of numbers for both purposes (or for any other purpose), correlations can be drawn between books published and books circulated; and, of course, comparisons can be drawn between one department and any other. Table 1 shows a selected list of as- signed D C classes. These numbers (based on the 16th edition) codify the departments only at SDSM&T, and nat- urally would vary somewhat from col- lege to college. Note how they lend themselves to short or long spans but are sometimes quite specific and isolated. The long spans help to cut down the list and usually account for most of the titles in a department. But sometimes the spe- cific numbers have considerable influ- ence on a department total, especially if many books fall into that class, as is the case with 510.78. The daily tabulations are fitted into their spans and can be totaled as need- ed. Ideally, all sources of circulation TABLE 1 Departments and Their Inclusive D C (16th) Groups—Partial List D e p a r t m e n t D C Groups Mathematics 510.0 510.77 Computation center . . . . 510.78 Mathematics 510.8-514.0 Civil engineering 515.0 Mathematics 516.0-519.0 Geology 526.8 Geology 549.9-551.4 Meteorology 551.5 Geology 551.6-559.0 Geology 622.1 Mining engineering . . . . 622.2-622.9 Civil engineering . . . . 624.0-628.0 Meteorology 629.1324 Measuring Classified Circulation According to Curriculum / 3 4 9 TABLE 2 Number of Circulated Books Relevant to Each Department * Per cent Circulation/ Department No. of Books of Total Enrollment Enrollment Batio Biology 142 2.19 _ _ Chemical engineering 165 2.55 82 2.1 Chemistry 363 5.60 42 8.6 Civil engineering 543 8.38 108 5.0 Computation center . 142 2.19 — — Electrical engineering 450 6.96 146 3.1 Geology 600 9.26 36 16.7 Social science and humanities . 2180 33.65 (Total) 3.1 Mathematics 560 8.64 34 16.5 Mechanical Engineering 149 2.30 158 .9 Metallurgy 421 6.50 37 11.4 Mining 82 1.27 33 2.5 Paleontology (museum) 24 .37 4 6.0 Physical education . . . . 53 .82 — — Physics 604 9.32 26 23.2 TOTAL 6478 100.0 706 9.2 * Because some categories have been left out, the figures cited in this paper do not necessarily represent the complete and true picture of SDSM&T circulation. should b e tapped: d e p a r t m e n t libraries, reserve, and other special collections. Unclassified periodicals cannot contrib- ute to the statistics. Since periodicals are a special problem anyway, we are h e r e dealing only w i t h monographs. T a b l e 2 shows a t y p i c a l set of figures for the number of c i r c u l a t e d books rel- evant to each department. T h e loans, as distributed among t h e departments, re- flect curriculum use of the library. No- t i c e that the figure for social sciences and humanities ( o n e department at South D a k o t a S c h o o l of Mines and T e c h n o l o g y ) is m u c h larger than any of the others. This is n o t surprising, even though a m a j o r is n o t offered in that de- partment. Almost everyone, however, is interested in its books. A professor of mining engineering m a y as likely borrow Uses of the Past as Geochemical Meth- ods of Prospecting. F u r t h e r m o r e , all stu- dents are required t o read in the human- ities. Such interest lends assurance to the feeling that humanities must not b e n e g l e c t e d in a heavily weighted techni- cal and scientific curriculum. Physics and mathematics, two depart- ments which are n o t only self-contained, b u t whose curriculums contribute to oth- er departments, have a relatively small n u m b e r of majors. Yet, as many, if not more, books circulate in these depart- ments as in the high enrollment depart- ments. This results in a high circulation/ enrollment ratio for mathematics and physics and suggests that at least these two departments, like t h e humanities department, n e e d continued heavy sup- port. O f course, other departments with high ratios or high circulation also need heavy support. A note of caution is in order. Unless e a c h d e p a r t m e n t shares in t h e subject analysis and in t h e compilation of D C numbers, t h e faculty will b e skeptical of results. T h e librarian must, therefore, in- terpret and use t h e results with great care. W i t h dangers recognized and proper c a r e taken, this and similar studies can have considerable value. F o r example, data, as t a b u l a t e d above, collected over a given time span, now make possible direct comparisons b e t w e e n t h e number of books published and the n u m b e r of books borrowed. Such a study is now underway at S D S M & T . F r o m such data 350 / College b- Research Libraries • September 1968 and studies we will be able to say that the books falling within the scope of one department are used more than another, and that we should perhaps buy more books in that area than in another. Other functions which might be ana- lyzed by the device are the existing book collection (measurable from the shelf- list); books published throughout the world (measurable, say, through the British National Bibliography); number of periodicals published in the United States and throughout the world (meas- urable from such compilations as New Serial Titles—Classed Subject Arrange- ment); journal articles in abstracting publications which have a detailed clas- sified arrangement, such as some of the Commonwealth Agricultural Bureau journals. Any of these media, especially those involving books, might be com- pared to the collections in large univer- sities by actually measuring shelf lists. The University of Michigan Undergrad- uate Shelflist would be a convenient tool to analyze, since it is generally available. Additional studies in any of these me- dia would have considerable value; and multivariate analyses such as multiple regression, of any or all, may be espe- cially revealing.