Government Resources

The Five Colleges of Ohio libraries are proposing a retrospective conversion project for their government publications collections. This consortium includes the oldest depository library in Ohio, Ohio Wesleyan University, which became a federal depository in 1845, followed by Oberlin College (1858), Kenyon College (1873), Denison University (1884), and The College of Wooster (1966). In a relatively small geographical area an historical wealth of information is available but not as equally accessible since many government publications are not included in the online catalog.

Traditionally, access to government publications has been through the experience of the government publications librarian, or through various indexes, but not the card catalog. However, with online capability and the ease of access to information regardless of format, it has become desirable to include government publications in the consortium database.

Within the consortium there are several ways in which government publications are handled and organized, although all adhere to the Guidelines for U.S. Depository Libraries. All of the institutions organize their collection by the Superintendent of Documents classification system, with a few selected titles in their library's main classifications systems (whether LC or Dewey). Because of the early dates when four of these institutions became depository libraries there are early American history materials in these document collections. Many administrative departments no longer exist, e.g., the War Department, etc.

Studies in the literature show that including document titles in the online catalog has greatly increased the usage of these valuable resources.

Each library was asked to take a critical look at their collection to identify the historical strengths and possible uniqueness of holdings on their shelves. This, admittedly, is a cursory survey since we were requested to do this December 11, 1997 and have the information available to the Directors of Libraries by January 23, 1998. However, the following information was gathered.

The College of Wooster: The College of Wooster completed an eight year retrospective conversion project in 1994 under the supervision of the librarians and the department associate. The bulk of the work was done by carefully trained students. There is still some clean-up work to do. It has been estimated that there are 2000 titles which need original cataloging and are not in CONSORT. About 900 of these are numbered House and Senate documents from various Congresses beginning with the 33rd Congress, 1854-. Another 900 titles are concentrated in the following departments, most of them published between 1940-1987: Bureau of the Census (115), Interior (145), Labor (95), State (45), Congress (165), commissions (80), misc. including Presidential documents, NASA, Education, Smithsonian Institute, etc. (240) for a total of almost 900. The greatest need for Wooster is to have an experienced cataloger do original cataloging for approximately 2000 titles.

Denison University: Unusual items in Denison's government publications collection which would enhance the CONSORT database holdings have been identified in the following areas: Bureau of the American Republics (3 shelves), Census (26 shelves prior to 1970), Foreign Operations Administration through the Federal Trade Commission (11.5 shelves), Interior (20 shelves), Labor (15 shelves), President through Office of the President (10 shelves), Navy (12 shelves), State (10 shelves), Smithsonian (21 shelves), Treasury (9 shelves), War Dept. (12 shelves not including the War of the Rebellion), Special Commissions and Committees (5 shelves), and Congressional hearings (10 shelves of older materials), "other" (5 shelves). Denison estimates that they have approximately 110,000 uncataloged volumes in their collection. If we assume that 75% can be found in CONSORT or OCLC databases, 82,500 volumes would need to be either linked or downloaded, and 27,500 volumes would require original cataloging. (Note: there may be fewer titles than number of volumes.)

Kenyon College: Unique items which would enhance the CONSORT database holdings are concentrated in the War Department (including documents under "M") (30 shelves), War Relocation (2 shelves), Smithsonian (10 shelves), President's Office (20 shelves from Roosevelt forward), State Department (10 shelves), Justice Department (10 shelves), Naval Department (10 shelves), Labor Bulletins (10 shelves), Civil Rights Commission (7 shelves), GAO microfiche collection (complete series), and Census materials. Because of the varying sizes it is difficult to estimate how many documents there are per shelf. It is much higher than the book average per shelf. A random count of three document shelves yielded 15 bound volumes, 75 titles, and 200 titles. If we picked the average figure of 95, this would be about 10,355 titles (not including GAO and Census) which may be unique titles and of importance in Kenyon's collection. Additional holdings, which do not fall in the "unique" category should be added to the online catalog.

Oberlin College: It is estimated that 33,000 government publication titles are not represented in Oberlin's catalog. This includes titles for which there is a record in the CONSORT catalog (Oberlin is not a participant in CONSORT), titles for which there is a record in the OCLC database and titles for which there is no record available in these databases and original cataloging is needed. An estimate is that over 75% may be found in the OCLC database. If this is true in Oberlin's case, they would have approximately 8,000 government publication titles which need to be originally cataloged. The balance should be included either in CONSORT or OCLC databases and these would need to be either linked or downloaded.

Ohio Wesleyan University: Unique items include Water Supply Papers, North American Fauna (1/2 shelf), Ohio County Soil Surveys, early Census materials (23 shelves), Coast and Geodetic Survey (48 shelves), Commerce reports and related materials (26 shelves), Children's Bureau (6 shelves), Women's Bureau, Correspondence with Continental Congress, and Journals of the Continental Congress (1774-1789), Navy Register (1908- ), Official Records of the Union and Confederate Navies in the War of the Rebellion, Register of the Dept. of State (1865-1866, 1920- ), President's Office (sizeable collection), Documents on German Foreign Policy (1 shelf), Papers Relating to the Treaty of Washington, International Railway Commission, U.S. Prohibition Bureau (1927-1929), American Archives (1775- ), and medical records in the Civil War. Since shelves were not counted in some instances, there is no title count available at this time. The entire collection would need to be processed in one of three ways listed below under considerations.

For four of the five collections, it is necessary to 1) link titles to records in their local catalog, 2) find OCLC records for those unique titles not in their local catalog, and 3) then one gets to the original cataloging which is needed. Only after the first two steps are completed, can the third step be done. For one collection, it would be necessary to search OCLC for the outstanding titles not in CONSORT, and then proceed to original cataloging.

Oberlin College is not part of the CONSORT database. What considerations need to be made by Oberlin and CONSORT?

What is the degree of overlap and duplication in the institutions? We could do a small study of two or three classes in two of the collections to get an idea of overlap. The actual number of titles to be processed will be fewer than the total titles in the collections.

It is felt that because of the fragile nature of many of the government publications which would be included in this project, libraries do not want to box these titles up and send them to OCLC for original cataloging. We would be without the sources (some of which are also quite valuable) for a length of time, and the documents are in a high-risk category for damage.

Because of the nature of government publications, a cataloger would need to be an experienced cataloger, preferably having documents understanding.

If a cataloger could do 5 original titles a day minimum, a cataloging standard used at the University of Idaho, this would be a minimum of 1250 titles per year (250 work days per year). Since this cataloger(s) would be dealing with like material, the titles done per day could be higher; however, the searching and author and subject authority work will continue take a certain amount of time for original cataloging. We should allow for start-up organizational time, training, etc.

There is an OCLC rate for retrospective conversion searching and cataloging. The rate is in effect before 9:00 am and after 5:00 pm on weekdays and all day on the weekends until 8:00 pm on Sunday. We should consider taking advantage of this substantial savings over normal cataloging costs; however, realistically, to employ experienced qualified catalogers, it may be necessary to have them work normal hours.

The studies of uncataloged materials mentioned above is not based on when the government publication title was published. It seems that the title should be included in our database(s) regardless of date.

It is the recommendation of the Five Colleges Government Publications Committee that the Directors apply for a grant to catalog all the government publications' titles in the five colleges government publications departments which are not cataloged. This includes the three levels described above: linking to records in the local catalog, locating OCLC records for unique titles not in local catalog, and cataloging those titles not in the OCLC database.

Our recommendation is that the consortium employ at least three to five catalogers for this project, and that each be assigned process specific Superintendent of Documents classes. This will avoid duplication of work. Assignments can be adjusted and coordinated as necessary. It is projected that the duration of the project would be a three year minimum.

Since it will be necessary to handle individual titles for the cataloging project, a preservation component for the Five Colleges should be considered to protect the more valuable documents in our collections from further deterioration. A preservation decision can be made as the documents are handled. Supplies for preservation needs, such as preservation boxes, should also be a part of the grant.

Supplies for the cataloging of government publications, such as barcodes, binding costs, OCLC search costs, downloading costs from OCLC per record, and travel expenses between libraries for necessary meetings of the catalogers should be built into the grant. An available terminal to the cataloger, and access to the OCLC database are considerations to be worked out at each institution. If the lower OCLC retrospective conversion rates are to be figured into the project, the catalogers may be working during some hours (but probably not all) when library terminals are not occupied by regular staff members.

It is recommended that each cataloger employed for this grant also be given a student assistant who can assist the cataloger in ways helpful in moving the project forward.

Discussed by the Five Colleges Government Documents Committee on January 15, 1998: compiled by Barbara L. Bell, convener, with input and assistance from Beverly Gage, Mary Prophet (Denison University), Andrea Peakovic, Priscilla McIntosh, Donna Wilson (Kenyon College), Tom Hinders, Cecilia Robinson (Oberlin College), Terry Maloney-Rose, Judy Orahood (Ohio Wesleyan University), and Jennifer McMullen (College of Wooster).