WILLIAM J. MAHER and BENJAMIN .F. SHEARER

Undergraduate Use Patterns of Newspapers on Microfilm

This analysis, based on undergraduate use patterns of neu;spapers on mi­ crofilm at the ( ni t.:e rsity of Illinois, presents criteria , Su ch as number of ti­ tles, dates requested , and existence of indexes, that should be considered u;hen purchasing microfilmed newspapers for the highest possible benefit in tenns of money spent. Statistics on use suggest that a user satisfaction let;el of almost 85 percent can be attained u;ith a trcenty-year backfile of ttcenty titles. The analysis presented is offered as one step in det;elopment of an aCl(llisitions policy for nercspapers on microfilm.

IN RECENT YEARS newspapers have become for a newspaper acquisitions policy, and this increasingly important reference and re­ article will attempt to establish criteria for search sources for ·a wide variety of users on the development of a policy for newspapers college and university campuses. They pro­ on microfilm for undergraduate users in vide original documentation for events that medium-size and small four-year colleges are too recent to be covered in monographs and universities. It is based on an analysis and professional journals or that have been of undergraduate use at the newspaper li­ bypassed by the traditional literature of his­ brary of the University of Illinois at tory, political science, economics, and the Urbana-Champaign during the peak periods other social sciences and humanities. News­ of the 1976177 academic year. papers are particularly attractive and useful While the conclusions presented here to undergraduates for course papers and may not hold true for all libraries, the class projects on a variety of subjects. methodology will be valid and should direct They do, however, present librarians with librarians to the formulation of policies for several unique problems. Because of their the selection and purchase of newspapers on physical nature, storage and preservation microfilm. are difficult. While microfilm does represent a solution, newspapers still present obsta­ USER SURVEY cles because of the high cost of microfilm Before presenting use statistics and runs and because of the lack of bibliographic guidelines for acquisition, it is useful to de­ control over their content. Moreover, since scribe the holdings and operations of the newspapers are general in their subject newspaper library at Illinois. With more coverage, the determination of potential than 63,000 reels of microfilmed newspa­ users is difficult. Undoubtedly, these factors pers , the library holds about 324 U.S. and contribute to the lack of professional litera­ 262 foreign titles. While the collection is ture on newspapers as library materials. 1 particularly strong in Illinois titles, there are Few libraries have developed guidelines large microfilm runs for metropolitan news­ papers in at least twenty-five states and in fourteen European and fourteen non­ William ]. Maher is newspaper librarian, Uni­ versity of Illinois at Urbana-Champaign, and European countries. There are strong hold­ Benjamin F. Shearer is government documents/ ings in such areas as labor, religious, liter­ law librarian, East Tennessee State University, ary, and black newspapers. johnson City, Tennessee . Because of an annual use of at least

254 I Undergraduate Use Patterns I 255

20,000 reels of microfilm, 2 it is possible to search materials. The purpose of under­ analyze the exact nature of undergraduate graduate use is also instructive: During No­ interest in newspapers from statistics main­ vember 1976 and April 1977, 85.5 percent tained on their use. When requesting film, of undergraduate use was for course pa­ all users complete a reference card on pers (the other· types of use recorded were which they record the desired title and " research for publication," "dissertation/ dates as well as information regarding their thesis research," " classroom use, " and purpose and status (i.e., faculty, graduate "other") . student, undergraduate, and public). To identify those newspapers that will These cards are used for all titles except satisfy the most requests at the lowest cost, the New York Times, the Chicago Tribune, the following analysis studies titles re­ and the Wall Street Journal. Because de­ quested, dates requested, and the impor­ mand for these three titles is very high, tance of an index for use. they are stor~d in microfilm cabinets freely accessible to all patrons. All other titles are TITLES REQUESTED shelved in a restricted area, and patrons Since newspapers on microfilm represent complete the reference card to request large investments, libraries must seek some them. The cards provide the basis for the assurance that the money spent yields an current analysis. acceptable cost-benefit ratio when com­ While the three most popular titles-New pared to the level of satisfaction of user York Times, Chicago Tribune, and Wall needs. The case in favor of such basic titles Street Journal-are not includ~d in these as the New York Times , Wall Street jour­ statistics, these titles are of such importance nal, and Chicago Tribune is clear, but the that they should be considered basic items case for purchasing other newspaper titles is for any college and university library. While far more complex. the Chicago Tribune is heavily used at Illi­ The geographical factor is perhaps the nois because of its local appeal, it is of na­ single most obvious element that influences tional importance and should be a priority undergraduate use of newspapers on mi­ for many libraries, especially those in the crofilm. Thus the statistics presented here Midwest. (Another regional ·substitute might on titles requested at Illinois should not be be the Washington Post.) Any library not taken as an inflexible acquisitions policy holding the New York Times and the Wall statement for other libraries. Some general Street Journal should regard these as the principles on undergraduate use patterns first titles to acquire on microfilm. The New do, however, emerge from these data, and York Times, in particular, should be a top these can be applied to any undergraduate priority because of its importance and be­ library in terms of both titles and dates that cause it is indexed for its entire run. Be­ may offer the greatest benefits for money yond these basic titles, acquisitions policy spent. can be determined best by an analysis of Table 1 shows user demand according to undergraduate use. titles requested during November 1976 and At the University of Illinois newspaper li­ April 1977. As this table indicates, under­ brary, two months~November and April­ graduate use of newspapers on microfilm consistently account for the highest use of during the two peak use months of the year film for the period from 1973 to 1977. This was restricted to forty-nine titles (excluding reflects a heavy demand by undergraduates the New York Times , Wall Street Journal, in order to complete their course work as and Chicago Tribune ). Of the titles re­ each semester draws to a close. While quested, thirteen (26.5 percent) were titles undergraduates annually account for 37 per­ no longer published ("dead"), but in terms cent of use, during the months of Novem­ of total use these titles accounted for only ber and April they account for 49 percent. sixteen requests (5. 7 percent). The over­ This fact justifies the isolation of these whelming number of requests were then for months for the current study. The use dur­ currently published ("live") titles. ing this period illustrates the importance The level of user satisfaction increases as undergraduates place on newspapers as re- more titles are added. The data show that 256 I College & Research Libraries • May 1979

TABLE 1 nificant factors. All of the top twenty titles NEWSPAPER TITLES ON MICROFILM requested w~re English-language titles, and REQUESTED BY UNDERGRADUATES, the London Times was the only foreign title NOVEMBER 1976 AND APRIL 1977, in the top twenty. Black newspapers also BY NUMBER OF REQUESTS received relatively high usage-four of the Number of top twenty were black papers (Chicago De­ Title Requests fender , New York Amsterdam News, 1. Dailh Illini 31 2. Was ington Post 31 Pittsburgh Courier, and Los Angeles Sen­ 3. Morning Courier (Champaign) 30 tinel). 4. Los Angeles Times 22 The statistics on titles requested point to 5. News-Gazette (Champaign) 19 some important conclusions. A high level of 6. St. Louis Post-Dispatch 13 7. Chicago Daily News 11 user satisfaction can be attained with rela­ 8. Times (London) 11 tively few newspaper titles on microfilm. 9. San Francisco Chronicle 8 Aside from the school newspaper, six titles 10. Chicago Defender 7 (including the two local papers; two titles 11 . New York Amsterdam News 7 12. Atlanta Constitution 7 from nearby metropolitan areas, such as the 13. Chicago ·Sun-Times 6 St. Louis Post-Dispatch and Chicago Sun­ 14. Christian Science Monitor 6 Times; and two titles of national promi­ 15. Pittsburgh Courier 6 nence, e.g., the Washington Post and the 16. Village Voice 5 17. Los Angeles Sentinel 4 Los Angeles Times) could well satisfy 50 18. Detroit Free Press 4 percent of all requests. It is necessary, 19. Illinois State therefore, that libraries consider geograph­ Journal-Register 4 ical bias, country of origin, language, and 20. Louisville Courier-Journal 4 21. New York Herald special topics taught in the curricula (e.g., Tribune (Dead) 4 black studies) when deciding to purchase 22. Cleveland Plain Dealer 3 microfilmed newspaper titles to meet user 23. Dallas Morning News 3 demands as precisely as possible. 24. National Observer 3 25. Norfolk Journal & Guide 3 DATES REQUESTED 26. Cincinnati Enquirer 2 27. Michigan Chronicle 2 Consideration should also be given to the 28. New Orleans Times-Picafune 2 dates requested. An investment in large 29-49. Twenz-one additiona titles (12 " ead") with 1 backfiles of e~en frequently used titles may request each 21 not prove beneficial if the use of certain Total requests 279 older dates is minimal. Table 2 demonstrates that undergraduate requests for newspapers in microform de­ 47.7 percent of all requests could be crease significantly for the years preceding satisfied with the top five titles, 65.6 per­ 1969. In fact, 19.7 percent of all requests cent with the top ten titles, 77.1 percent could be satisfied by keeping only a one­ with the top fifteen , and 84.6 percent with year backfile in paper, thus avoiding a con­ the top twenty. Thus, only twenty-seven siderable expenditure for microfilm. This requests (9. 7 percent) for "live" titles were saving must, however, be weighed against not satisfied by the twenty most popular ti­ the overall impracticality of storing and ser­ tles. vicing newsprint. Problems of theft, mutila­ The factor of local bias is indeed quite tion, and preservation must be taken into important, as these statistics show. Three of account. the top five titles requested are local news­ Assuming that libraries interested in papers and by themselves constitute 28.7 building a research collection will consider percent of all requests. The Daily Illini is purchasing all but current issues of news­ the student paper of the University of Illi­ papers in microform, ·it is instructive to nois, and its heavy use is quite understand­ analyze use in terms of five-, ten-, fifteen-, able. The Champaign-Urbana Cqurier and and twenty-year backfiles. It should be th.e News-Gazette are the two local papers. noted first, however, that only 22.9 percent Language and country of origin are also sig- of all citations were for dates before 1962, Undergraduate Use Patterns I 257

TABLE 2 which are seldom used, are available DATES OF UNDERGRADUATE REC)UESTS through cooperative lending. In short, there FOR NEWSPAPERS ON MICROFILM, is little reason for a four-year college library NOVEMBER 1976 AND APRIL 1977 to invest in long backfiles of newspapers on umber of microfilm, especially if budgets are tight. Year Requests If a library is experiencing an especially 1976 117 severe budget crunch, it may choose to let 1975 71 1974 24 its collection of newspapers in microform 1973 40 grow organically by ordering only current 1972 33 subscriptions. The cost of current subscrip­ Total 1972-76 285 (47.9%) tions of filmed newspape rs usually is sig­ 1971 34 nificantly less than the cost of backfiles. For 1970 28 instance, based on the 1978 catalog of the 1969 27 Microfilming Corporation of America, a cur­ 1968 13 1967 13 rent subscription to the Atlanta Constitution Total 1967- 71 115 (19.3%) costs $362, whereas recent backfiles cost Total 1967-76 400 (67.2%) $540 per year. In general, recent backfiles 1966 15 (within the past fi\'e years) tend to be about 1965 8 40 percent more expensi\'e than the current 1964 8 subscriptions themseh-es. 1963 19 Although current subscriptions to approx­ 1962 9 imately six titles may satisfy as much as 20 Total 1962-66 59 (10.0%) percent of user demands, more user satis­ Total 1962-76 459 (77.1%) hlction is attained as the size of the backfile 1961 7 increases. Thus, with a fi\'e-year backfile 1960 4 1959 5 (1972-76), 47.9 percent of all requests could 1958 21 be fill ed. With the purchase of a ten-year 1957 7 -- backfile (1967-76), 67.2 percent of all user Total 1957-61 44 (7.4%) demands could be satisfied; with a fifteen­ Total 1957-76 503 (84.5%) year backfile (1962-76), 77.1 percent of de­ 1947-56 40 mands; and with a twenty-year backfile 1937-46 16 (1957-76), 84.5 percent of all demands. 1927--36 1 These figures illustrate the types of ques­ 1917-26 6 1907-16 4 tions librarians must ask themseh es when 1897-1906 11 deciding to purchase backfiles of mi­ 1887-96 5 crofilmed newspapers. For example, if 77 .1 1877-86 1 1867-76 2 percent of all requests can be satisfied with 1857-66 6 a fifteen-year backfile, is the purchase of Total 1857-1956 92 (15.5%) fi\'e additional years of a certain number of Total 1857-1976 595 (100%) titles worth satisfying an additional 7.4 per­ cent of user demands? These statistics indi­ cate that a fifteen-year backfile in microform and 8. 7 percent for dates before 1946. of the most fi·equently used newspapers (as These are significant percentages of all cita­ determined by the factors described abo, e) tions, but the cost of filling user demands oflers a ,·ery high benefit f(.>r its cost. The for citations prior to 1962 must be deter­ money that could be spent on fi,·e more mined not only in re lation to b e nefits but years of backfiles might be better used to also in regard to the availability of older purchase new titles. papers on some kind of cooperative loan When the data on titles requested and arrangement. dates requested are brought together, a sig­ Interlibrary loan and membership in sys­ nificant parallel emerges in the le\'el of user tems, such as the Center for Research Li­ satisfaction that can be attained with num­ braries, are good examples of cooperative bers of titles and the extent of backfiles. arrangements. Certainly most "dead" titles, Summaries of user satisfaction by number of

~------258 I College & Research Libraries • May 1979

titles and size of backfile are shown in tables not likely to be of long-term interest to 3 and 4. serious researchers. Moreover, their cost The close relationship shown in tables 3 (about $1,000 per year for Newsbank) may and 4 between titles and years requested for seem high for the benefit they bring. 3 user satisfaction demonstrates the necessity Thus adequate access to newspaper ac­ of considering these two factors together counts on a given subject can be obtained when deciding to purchase newspapers in only by the use of the available indexes for microform. For example, a 77 .1-percent individual titles. Even then, the researcher level of user satisfaction could not be at­ is still confronted with a serious problem tained merely by acquiring fifteen titles because of the lack of indexes for all but the without also acquiring a fifteen-year backfile most recent years of most newspapers. of these titles. The two criteria of number of While the New York Times and the London titles and years of backfiles are inextricably Times are indexed for their entire runs, related and must be considered jointly if a most metropolitan U.S. newspapers were library wishes to obtain the highest level of not indexed until about 1970. user satisfaction for the money it spends on It is reasonable to argue that the exis­ ' newspapers in microform. tence of an index will contribute to more requests for a given title, and statistics at TABLE 3 the UniYersity of Illinois support this con­ USE-H SATISFACTJO!\ BY TITLES HELD clusion. Table 5 describes the relation of use of indexed papers to use of papers not l'erct•nt nf · umhPr Ht'((llt'SI' nr indexed. An"''t"red Title, These statistics lead to seYeral important 47.7o/c 5 conclusions. First, for dates requested 65.6o/c 10 77.1% 15 within the past ten years, indexed titles ac­ 84.6o/c 20 count for more than one-half of all newspa­ pers used. The average beginning date of TABLE 4 most newspaper indexes is 1970 and is cal­ culated on the following titles: Atlanta Con­ USEH SATISFACTJO!\ BY YEAHS R E<)l'ESTED stitution (index begins 1971), Index tv Black J'erl"Pill nf Yt·ar' Newspapers (1977), Christian Science Ht·qnt"'h or Answt•red Baddilt• 47.9o/c 5 (1972-76) TABLE 5 67.2% 10 (1967-76) ll\DEXED N E\\'SI'AI'EHS' s II AHE 77.1% 15 (1962-76) OF TOTAL U~E 84.5o/c 20 (1957-76) J't'l't'('lll nf U"· ROLE OF INDEXES 1976 78.6% Date and number of titles are important 1975 59.3% 1974 62.5% considerations in acquisitions policy, but ac­ 1973 60.9% cessibility is equally significant. Newspapers 1972 82.6% are a major source of information on 1972-76 average 68. 1o/c numerous subjects of great interest to 1971 39.1% undergraduates. They do, however, present 1970 52.4o/c problems for bibliographical control because 1969 36.0o/c 1968 50.0% they are comprehensive in their subject 1967 57.1 o/c coverage. 1967- 71 average 46.9% There are no indexes covering all news­ 1967-76 average 51.8o/c papers, with the exception of microform se­ 1966 50.0o/c ries, such as Newsbank, which contain top­ 1965 42.9% ically arranged clippings fi·om several news­ 1964 50.0% 1963 5.3% papers. While such collections do provide 1962 1.0% subject access to newspapers, their coverage 1962-66 average 29.8% is not deep and the subject categories are 1962-76 average 48.5% Undergraduate Use Patterns I 259

Monitor (1950), Daily Illini (1961), journal possible to show an overall rise in use be­ of Commerce (1976), Los Angeles Times cause of indexes, there is a demonstrable (1970), Minneapolis Star and Tribune increase in the preference for indexed (1971), National Obsen;er (1969), St. Louis newspapers. This would seem to be a natu­ ·newspapers (1975), San Francisco Chronicle ral reflection. of, the greater availability of (1970), New Orleans Times-Picayune (1972), indexes since 1970. and Washington Post (1971). The New York Considering the importance of indexes for Times, Times of London, and Wall Street the use of newspaper titles, one would be journal were excluded from the average. inclined to purchase only indexed titles. In­ Thus it would be more appropriate to use dexing should not, however, be the sole cri­ the 1972-76 average (68. 7 percent) as indic­ terion because certain titles (e. g., those ative of undergraduate preference for in­ fi·om many major U.S. cities) will be heavily dexed titles. used despite the lack of an index. In view of this, formulators of acquisitions Of the twenty most requested titles listed policy should be very cautious about the in table 1, only eight (40 percent) were in­ purchase of newspapers on microfilm unless dexed at the time of use (four of the titles an index is available and also purchased. were black newspapers that have since been While the cost of indexes is high (averaging covered by the Index to Black Newspapers ). about $185, or one-half of the cost of many The fact that 60 percent of these twenty film subscriptions), they clearly contribute most-requested titles were not indexed to a large part of undergraduate use of should cause librarians to think carefi.tlly he­ newspapers in microform. 4 fore refusing to purchase any nonindexed ti­ To a large extent, the purchase of an tles. Of the remaining twelve titles, five index for a title already held is preferable to (two from Champaign-Urbana, two from acquiring new titles. For example: If a li­ Chicago, and one fi·om Springfield, Illinois) brary were receiving the Washington Post are "locally" focused papers that have no and the St. Louis Post-Dispatch at a total index. If these five "local" newspapers were cost of $876 and were considering the pur­ excluded fi·om the analysis of the relation of chase of another title such as the Times­ indexes to most-used titles, indexed papers Picayune at a cost of $390, it would be bet­ would account for 53.5 percent of the fifteen ter advised to use that money for the in­ most-requested papers. dexes to the Post and the Post-Dispatch. It should also he emphasized that 68.7 Thus the selection of titles would be more percent of all use of newspapers covering limited, but access would be simplified and the years 1972-76 is in indexed titles, even user satishlCtion could be increased. If a li­ if indexed papers account for only 40 per­ brary cannot bear the cost of monthly cent of the twenty most-requested titles. and/or quarterly indexes, it may consider Thus if a library were contemplating the purchasing only annual cumulative updates purchase of two papers of equal stature, one when these are available separately. indexed and the other nonindexed, the in­ It is logical to assume that overall use of dexed title should be preferred to achieve newspapers will increase with the presence greater patron satishlCtion. More

REFERENCES

1. A search of Library Literature from 1970 to 1977) is an excellent survey of newspaper in­ 1978 renders only a handful of citations under dexes and includes both published indexes and the category of "Newspapers," and none of unpublished ones (largely clipping files). A these address the question of use patterns. reading of her list of titles with published in­ Mark Yerburgh's recent article "Academic Li­ dexes would illustrate the overall lack of braries and the Evaluation of Microform Col­ adequate bibliographic control of newspapers. lections," Microform Redew 7:14-19 (Jan.­ 4. The indexes included in this average price are Feb. 1978), is a first step in recognizing the Atlanta Constitution, Index to Black Newspa­ necessity of evaluating microfilm collections in pers, Christian Science Monitor, Los Angeles terms of "cost/benefit." However, the article Times, Minneapolis Tribune, New York Times, concerns only microfilm collections in general Washington Post, Chicago Tribune, San Fran­ and does not address specific problems of cisco Chronicle, Times-Picayune, and St. Louis newspapers. Post-Dispatch. 2. "Newspaper Library Annual Report, 1976--77," 5. This average of $263 for twenty titles includes compiled by Charles B. Elston, newspaper li­ four black newspapers at a relatively low cost brarian at Illinois from 1971 to 1977. By estab­ . ranging from $17 to $55 each. If these were lishing various procedures for keeping statis­ excluded, the average cost would rise to $321 tics, Elston has laid the basis for the data used per title per year, based on current subscrip­ in the present analysis. tion prices. This higher figure is probably 3. Anita C. Milner's recently published News­ more realistic. paper Indexes (Metuchen, N.J.: Scarecrow,