College and Research Libraries By J E N N E T T E E. H I T C H C O C K Objective Subjectivity: Four-Year Report on Starred Subject Cards IN T H E W A K E of the Columbia Institute on Subject Analysis in 1952 and the discussion there of catalog use studies and the desirability of devising qualita- tive studies,1 Yale has been collecting data for four years that would show use in terms of specific books. Under certain subject headings, a star was added to the call number on each card. T h e circulation desk saved the call slips with stars each clay and these were photocopied. This routine provided a record of books selected by subject and by academic status of the persons who made the selections. T h e subjects under which all cards were starred were for the most part arbi- trarily chosen by the Subject Experiment Committee, subject catalogers, and ref- erence librarians. Care was taken to in- clude subjects varying in number of cards from under ten to over a thousand and to have representation in all major subject areas. T h e cards starred, how- ever, do not constitute a statistical sam- 1 Carlyle J. Frarey, "Studies of Use of the Subject Catalog: Summary and Evaluation" in Maurice F. Tauber, ed. The Subject Analysis of Library Materials ( N e w York, School of Library Service, Columbia University, 1953), p. 154: " T h a t none [qualitative studies of catalog use] have been made thus far only underlines the difficulties in devising methodologies which might yield reliable results. Perhaps no suitable method can be. found. But until serious attempts have been made, and until the whole problem of qualitative use has been explored more minutely, we cannot rule out such studies as impracticable or unnecessary. Neither can we proceed unquestioningly to propose extensive modifications in the subject catalog based solely upon the evidence of quantity u s e . " Miss Hitchcock, formerly Senior Cat- aloger at Yale University Library, is Head of the Catalog Department, Stan- ford University Libraries. pie. T h e observations cannot with reli- ability and validity be projected through- out the catalog nor to other libraries. There were in the experiment 396 subjects and 32,897 starred call numbers. T h e total number of subject cards in the catalog was estimated at 1,110,000 when the experiment was started. T h e virtue of the star technique is that it does not disturb the patron in his normal course of action. T h e stars pro- vided a robot interview mechanism or automatic transcribing device, so to speak—an objective record of subjective performance. T h e process of netting stars, however, was deadly slow. F O R E I G N L A N G U A G E M A T E R I A L Merritt's study has provided a quanti- tative approach for a qualitative inves- tigation of the use of foreign language material: Although 50 per cent of all titles cur- rently being cataloged are in foreign lan- guages, only 6.2 per cent of all books loaned through the subject catalog were written in foreign languages. . . . Thus the subject-cataloging load could be re- duced by 50 per cent while reducing the efficiency of the subject catalog by only 6.2 per cent, on a purely quantitative basis.2 Yale's figure for foreign language books was 5.3 per cent, 89 out of 1,694 call slips. T h e foreign language percent- ages for total number of call slips within 2 LeRoy C. Merritt, The Use of the Subject Catalog in the University of California Library. (University of California Publications in Librarianship, 1951, N o . 1 ) , p. 15. JANUARY 1959 9 TABLE VII PERCENTAGES OF E N G L I S H L A N G U A G E M A T E R I A L : T H R E E - Y E A R A N A L Y S I S FOR F I F T E E N SUBJECTS SUBJECT TITLES IN CATALOG CALL SLIPS SUBJECT TOTAL NO. % ENGLISH TOTAL NO. % ENGLISH Acculturation 67 89 19 89 Behaviorism 81 92 13 100 Dreams 117 53 25 100 Existentialism 126 26 85 83 Geometry, N o n - E u c l i d e a n 73 25 23 100 Gestalt (Psychology) 48 67 22 100 H y p n o t i s m 75 56 81 100 Impressionism (Art) 64 50 20 80 Isotopes 72 85 32 97 M a n — O r i g i n 101 70 29 100 Q u a n t u m theory 185 54 33 85 Rorschach test 33 82 20 100 Semantics (Philosophy) 23 82 11 100 Short story 87 92 15 100 T i m e 133 44 24 96 each broad category of patrons is as fol- lows: Undergraduates, 2.7; Graduates, 14.4; Faculty, 15.2; and other, 2.5. This breakdown indicates that the faculty and graduate students would be the patrons penalized by curtailment of subject en- tries for foreign language material—but not to an alarming extent. What would the faculty (alleged in library circles never to use subject head- ings) lose in terms of specific books? Within the starred sample, over the four- year period, nine faculty men used eight subject headings in filling out call slips for twenty-two foreign language books. An assistant professor in Italian se- lected from under E X I S T E N T I A L I S M six out of twelve Italian- titles, one French work of an Italian author, and one Eng- lish translation of an Italian work along with three other English works. An in- structor in Russian took out twelve books on E X I S T E N T I A L I S M : four in French, the others in English. A n assistant professor in French se- lected from S E M A N T I C S two French works on the same day with four in English and from S E M A N T I C S ( P H I L O S O P H Y ) one in Spanish. H I S T O R Y - P H I L S O P H Y was used by an in- structor in German for two titles in Ger- can and one in English. A professor in economics used E C O N O M I C S - P E R I O D I C A L S as the approach for "any recent issue" of two Italian journals. An associate pro- fessor emeritus in engineering drawing used the catalog for a French book on B R I D G E S - P A R I S . A Bonn inaugural dis- sertation in German was found under K O R E A N WAR, 1950- by an associate pro- fessor of economics. A French work was charged out for an associate professor of city planning from under C I T Y P L A N - N I N G - F R A N C E . A n d an associate professor of harpsichord playing took one of seven Spanish titles from the total of eight cards for S P A I N - N O B I L I T Y . Without enumerating the thirty-two books used by graduate students, the twenty-three by undergraduates, and the twelve by other persons, conditional uses are herewith generalized. Foreign language material may be selected, ir- respective of status of the patron: (1) If 10 COLLEGE AND RESEARCH LIBRARIES the language is easily read by the patron (e.g., Japanese books on E X I S T E N T I A L I S M by a Japanese undergraduate); (2) If the language is significant for the subject (e.g., French for I M P R E S S I O N I S M ( A R T ) or German for Q U A N T U M T H E O R Y ) ; (3) If there is little material in English avail- able (e.g., all three books under THE- O P H I L A N T R O P Y used by one patron, two in French and one in English). On the other hand, by comparing se- lections as shown by the call slips with cards in catalog (a study made for sub- jects that had more than fifty cards and more than ten call slip requests), it was very apparent that foreign language titles usually were bypassed before the patron made his selections. This is in- dicated in Table I, but the card-by-card comparison was even more convincing. L A R G E V S . S M A L L S U B J E C T S Yale's special interest in a qualitative technique was to derive criteria for as- signing material to compact storage. There is fairly general satisfaction now that books in foreign languages, when T A B L E I I M A I N S A M P L E , A L L S U B J E C T S : P E R C E N T A G E S O F M A T E R I A L U S E D SIZE GROUP YEAR I YEAR II YEAR III O v e r 4 0 0 0 . 9 1.8 0 . 6 1 0 1 - 4 0 0 3 . 6 4 . 5 1.7 5 1 - 1 0 0 9 . 1 4 . 7 3 . 1 1 1 - 5 0 5 . 5 3 . 4 1 . 9 1 - 1 0 4 . 3 3 . 0 2 . 3 T A B L E I I I P S Y C H O L O G Y S A M P L E , A L L S U B J E C T S : P E R C E N T A G E S O F M A T E R I A L U S E D SIZE GROUP YEAR I YEAR II YEAR III O v e r 4 0 0 0 . 9 0 . 6 0 . 4 1 0 1 - 4 0 0 2 . 5 2 . 2 2 . 0 5 1 - 1 0 0 8 . 8 5 . 6 6 . 7 1 1 - 5 0 9 . 6 1 1 . 1 5 . 1 1 - 1 0 not significant as primary source ma- terial, are very good candidates for stor- age; but that subject heading assignment should be on the same basis as that for material in the stacks. This still leaves the million-plus sub- ject cards in the catalog, occupying a thousand increasingly expensive catalog trays. Proceeding on the assumption that large subjects occupying a tray or more take up more space than warranted— and constitute through bulk a difficult hurdle for the patron—particular atten- tion was given to large subjects vs. small subjects. At the beginning of the experiment, a quota of 15,000 starred cards was allot- ted for a "main sample," set up on the basis of an estimated 9:13 ratio for cards under such large subjects to cards under other subjects (450,000: 660,000). This main sample contained 288 of the 396 subjects in the experiment. It was or- ganized for tallying results by listing the subjects according to the number of cards under each subject. Table II shows the percentages for the starred JANUARY 1959 11 TABLE VII M A I N S A M P L E , S U B J E C T S U S E D : P E R C E N T A G E S O F M A T E R I A L U S E D SIZE GROUP YEAR I YEAR II YEAR III Over 400 1.1 1.2 0.7 101-400 6.2 6.6 2.9 51-100 13.1 6.7 5.7 11- 50 11.7 11.2 9.4 1- 10 41.4 37.9 33.3 call slips received with respect to the total number of starred cards. A sample confined to headings in a single subject field, psychology, (4,120 cards) gave comparable results, as shown in T a b l e III. These two tables show that the greatest amount of material used (small though the amount seems to be) was in the subjects with fifty or a hun- dred cards. T h e percentages when restricted to those subjects which actually produced call slips (Table IV) show that the most material used per card in the catalog is in the smallest subjects. As assumed, there is demonstrated in Table IV a dead- weight of cards under the largest sub- jects—and a most discouraging dead- weight of subject cards in general. If no more use was made of the million cards than the 32,000 in the experiment, then the expense of housing these cards is largely for purposes other than consult- ing them at the catalog for charging out books—or for no justifiable reason at all. By way of evaluating the low figures, in the summer of 1955, date-due slips in the books were checked for total charges for some subjects of the sample. Table V not only shows variation in sub- ject use among the subject headings, but also high percentages of no use at all. When the percentages of Tables II-IV are uplifted by subtracting books with no charges at all, they reach the more impressive levels for subject-derived charges shown in Table V. In view of Merritt's figure of 26.2 per cent for sub- ject-derived charges,3 the subjects in the total sample for the Yale experiment may be considered below average in use. Attempts to project the call slips received from the sample with the total circulation and total number of cards in the catalog 3 Ibid., p. 4. T A B L E V D A T E D U E S L I P C H E C K I N G SUBJECT % WITHOUT CHARGES TOTAL CHARGES SUBJECT CHARGES % SUBJECT CHARGES Office management 80 7 2 29 Geometry, Non-Euclidean 75 35 14 40 Korean War, 1950- 41 50 6 12 Geopolitics 76 47 2 4.3 88 new subjects (ten cards or less) 53 241 15 6.2 Television 58 31 13 42 Hypnotism 69 49 30 61 Political parties—U.S. 48 138 26 19 12 COLLEGE AND RESEARCH LIBRARIES not only were statistically unsound but did not get far beyond a 10 per cent level. Yale's experiment included serials; Merritt's did not. All this is grist for the mills of pro- posals for publishing union subject catalogs and unwieldy sections of large card catalogs. But it does not give any good guide for determining which spe- cific books will not be used. S E L E C T I O N O F S U B J E C T S A N D B O O K S There seems to be but little basis for deciding which particular subject head- ings may be of interest. It cannot be said that the largest subjects are not used, but only that the least amount of material under them is used, relatively speaking. It had been speculated that nobody uses a large general subject like E C O N O M I C S . It was almost a year before a call slip appeared in E C O N O M I C S but it eventually piled up eighteen readers. T h e interest among the five large general subjects of the experiment varied as shown in Table VI, which gives use in terms of readers rather than number of call slips. It is not a large number of cards per se that acts as a deterring force; nor on the other hand, is a greater number of readers to be expected because there is a large amount of published material. N o call slips were received in any of the four years for E U R O P E A N W A R - C A U S E S (483 cards). H Y P N O T I S M (79 cards and 39 read- ers) and E X I S T E N T I A L I S M (132 cards and 38 readers) together with P H I L O S O P H Y pulled the greatest number of readers. Nor is large size a disastrous hurdle for anyone who really goes after what he wants. More readers used subdivisions than general works in P H I L O S O P H Y (seventeen readers for the single sub- division H I S T O R Y against fourteen for general works) and also in P S Y C H O L O G Y and S O C I O L O G Y . These readers were not intimidated by more than one tray nor guided filing arrangements. TABLE VI NUMBER OF READERS IN LARGE GENERAL SUBJECTS SUBJECT NUMBER OF CARDS NUMBER OF READERS ( 4 YRS) C h e m i s t r y 1 , 3 4 7 2 2 E c o n o m i c s 2 , 3 8 8 18 P h i l o s o p h y 1,771 4 8 P s y c h o l o g y 1 , 2 9 2 2 3 S o c i o l o g y c a . 9 8 0 16 A small number of cards does ap- parently mean less readers interested in the subject. Call slips were received for 112 out of 172 subjects which had more than ten cards; that is, for two out of three subjects. But for subjects which had ten cards or less, call slips were re- ceived for only 36 out of 221, one out of six subjects. T h e particular subjects that may be used are still considered unpredictable after four years. It was with some dismay that we watched the experiment open with stars for V A U D E V I L L E (10 cards) and C O N J U R I N G (39 cards). Yet a whole year went by before a star came in for W O R L D P O L I T I C S with its handsome array of 797 cards in the catalog. It seemed highly unlikely that anyone would use A B N A K I L A N G U A G E o r O R G A N I S T S O r L U X E M B U R G ( G R A N D D U C H Y ) D E S C R I P T I O N A N D T R A V E L , but someone and somebody did. But why in four years did no one request any book by subject under I N T E R N A - T I O N A L E D U C A T I O N , M E D I C A L S O C I A L W O R K , or R A D I O C H E M I S T R Y ? Quite possi- bly, because of loopholes in the star technique. T h e data collected really can be used only in a positive way. Thus it may be said that each year more sub- jects appeared on the list of those that had been used. It is not impossible to believe that when the life of the sur- veyors reaches infinity, all subjects in the catalog will have been used. It is less easy to believe that all cards JANUARY 1959 13 TABLE VII P O S I T I O N IN T H E A L P H A B E T : T H R E E - Y E A R A N A L Y S I S * NUMBER OF CALL SLIPS BY TOTAL NO. QUARTER OF THE ALPHABET NUMBER OF SUBJECT CALL SLIPS READERS 1ST 2ND 3RD 4 T H Acculturation 19 6 4 7 2 4 Behaviorism 13 4 1 1 7 6 Dreams 25 11 9 5 0 6 Existentialism 85 31 24 8 22 30 Geometry, Non-Euclidean 23 9 0 6 8 10 Gestalt (Psychology) 22 7 7 2 6 10 Hypnotism 81 53 12 6 10 32 Impressionism (Art) 20 6 3 8 3 11 Isotopes 32 9 10 8 5 12 Man—Origin 29 15 7 5 2 5 Quantum theory 33 5 10 7 11 16 Rorschach test 20 7 5 5 3 7 Semantics (Philosophy) 11 4 4 0 3 7 Short story 15 5 1 7 2 6 T i m e 24 9 5 6 4 11 * Based on call slips and cards in "general works" only. Alphabet quartered by dividing total of stack and non-stack titles; call slips for stack titles only. under all subjects will ever be used within one large library. So again comes the question of what specific books could be omitted from subject display in the catalog. T h e scrutiny of selections made in fifteen subjects, as mentioned in Table I, showed that foreign language material was usually bypassed. There was a pattern of selection taking place in the first portion of the alphabet; but this pattern was not clear cut in all sub- jects. It was marred by selections at the X-Y-Z end, selections sometimes made on the basis of recent publication date or a well-known author—but sometimes for no perceivable reason. T h e distribution of call slips by the quarter of the alpha- bet in which the main entry is filed under the subject heading is shown in Table VII. T h e analyses were far from conclusive, and the Catalog Use Study4 now has data 4 A m e r i c a n L i b r a r y A s s o c i a t i o n , R e s o u r c e s and Technical Services D i v i s i o n , Cataloging and Classifica- tion Section, Catalog Use Study, director's report by Sidney L . Jackson, ed. by V a c l a v M o s t e c k y ( C h i c a g o : A L A , 1 9 5 8 ) , pp. 33-38. which far eclipse those gathered by the star technique. W e can say that some readers found their Watson at the end of eighty-three cards under B E H A V I O R I S M . W e speculate that the readers knew his surname only and could not find him by author card. W e speculate whether other readers missed this good author because they stopped before the middle of the file even and wonder if it would be a kind- ness to separate recent works in English and those by prominent authors behind a separate guide card. But our deduc- tions are subject to variables of the nature of the subject, the status of the reader, the available material, and so on. T h e variables seem, in the case of selec- tion by imprint date, to have carried great weight. R A N D O M O B S E R V A T I O N S T h e average number of call slips per reader was two and one-half, as calcu- lated at the end of three years. Five or (Continued on page 62) 14 COLLEGE AND RESEARCH LIBRARIES Objective Subjectivity (Continued from page 14) more charges at one time were not un- common. T h e interesting observation here is that for these mass charges, under- graduates were in the minority whereas they were in the majority for the total charges. T h e highest number of call slips at one time was thirty-one, for a varsity debate. A Law School student took out twenty-two books in one day for R E G I O N A L P L A N N I N G . From the fac- ulty, who apocryphally never use sub- jects unless possibly outside of their sub- ject area, a professor of political science called for seven books at once in P O L I T - I C A L PARTIES-U.S. There was frequent use of the subdi- vision PERIODICALS (e.g., S O C I O L O G Y - P E R I - O D I C A L S ) to request "any recent issue" of a periodical and to call for known vol- umes and years instead of the main entry approach. This may reflect on the filing arrangement and the difficulty of finding a title that has the same fding word as a long file of subject cards. Readers work diligently to find an "author" for the author line on the call slip form. This was noticed in the pro- cedure for checking against the master file of the experiment to determine the subject headings consulted by the read- ers. T h e number of instances when the catalog entry was a corporate body or title and the reader had supplied a per- sonal author was not slight. This could be taken as reader's preference, by those occupied with revision of descriptive cat- aloging rules. T H E S T A R T E C H N I Q U E Scrutiny of the call slips as they came in provided many conversation pieces during the long months of the experi- ment—too many and too speculative to be recorded herein. Such individual ob- servations and speculations were not much different from the reminiscences of reference librarians or a diary tech- nique of catalog use study. But sifting and sorting of the accumulated call slips produced objectively based inductions, as have been included above, not always preenvisaged and not constrained by the artificiality of a questionnaire or bias of an interview. T h e experiment may be considered an experiment in qualitative methodol- ogy rather than an experiment that has produced catalog-shaking results. T h e technique was objective; but the arduous labor of sifting and sorting slips, plus in- sufficient control of variables that rein- troduced subjectivity in the interpreta- tion of some data, has nullified efforts that could have added even more results to the increments of catalog use studies. T H E Y ARE WISE P A R E N T S — w h o , shopping around for a college to try to get son or daughter into, make their first campus call at the college or university library. A campus library is by sheer necessity the hub of the educational wheel in any American higher learning institution. A well stocked library, directed by an accomplished librarian and served by a competent staff, sets the stage for further investigation at least. —Holyoke (Mass.) Transcript-Telegram, October 25, 1958 62 COLLEGE AND RESEARCH LIBRARIES