College and Research Libraries Results of a Survey of Libraries on Pu-blication Plans for I 7th Decennial Censusl Introduction During November and December 1948, the chairman of the A.L.A. Committee on Public Documents sent questionnaires to 237 selected libraries requesting recom- mendations and suggestions which would be useful to the Bureau of the Census in formulating its publication plans for the 1950 Census. The 237 libraries which were canvassed included the 124 govern- nous census information, and (b) publica- tion policy for the basic volumes, particularly with respect to such items as title page for- mat, method of serialization, and identifica- tion of preliminary and final reports. The libraries canvassed were selected so as to include, insofar as possible, those which are the most extensive users of Bureau of the Census publications. The group to whom questionnaires were sent was made Table I A.L.A. Survey on Publication Preferences for 1950 Census Response Status by Type of Library November-December 1948 Replies T ype of Library Total All College and University Public State Special ment "all" depository libraries and other selected lists consisting of 54 public, 39 col- lege and university, and 20 special libraries. The questionnaire requested the com- ments, opinions and recommendations of li- brarians on : (a) general reference works which were proposed as a convenient means of increasing the usefulness of the volumi- 1 This report is a summary of results of a s urvey of Iibrarie on publication plans for 17th Decennial Cen su s conducted by the 'Committee on Public Docu- ments, American 'Library A ssociation, ovember 1948, with accompanying recommendations. Jerome K . Wil- cox, . t?e chairman of the committee, expresses ap- pr ectahon to th e taff of the Bureau of the 'Census for their cooperation in furni shing necessary informa- tion ami ass \stin g in the deta iled work. OCTOBER~ 1949 Received Not Received 237 130 107 107 65 42 88 47 41 21 10 II 21 8 13 up of 107 large college and state university libraries, 88 public libraries having 100,000 volumes or more, and a small selected group of special libraries. (See Table I.) These libraries included the 124 federal all-deposi- tory libraries. Summary of Results More than one- half (55 per cent) of the libraries canvassed replied to the question- naire ( 73 all-depository, 25 public, 25 col- lege and university, and six special librar- ies) . The replies showed a great deal of interest and a variety of recommendations. 417 Proposed I950 Census Reference Works Comments and recommendations regard- ing one or more proposed reference volumes were made by about 120 libraries, and an (See Table II.) This was especially true in the case of the comprehensive index, the one-volume summary of the results of the census, and the small area data book, as Table II Summary of Comments1 on Proposed Reference Works for 17th Decennial Census Total Defi- Not Very Could No Type of Publication nitely Impor- be Not and Class of Library Com- Favor- Favor- Impor- tant Needed Omit- Needed Com- ments able able tant ted ment ----------------------- I. History 119 100 19 2 2 96 13 6 26 ----------------- Colleges and U ni versi ties .. . . 62 55 7 I 2 52 5 2 14 Public Libraries . .. . . .. . . .. 41 33 8 I 0 3'2 6 2 9 State Libraries . . . .. . . .. ... 8 7 I 0 0 ' 7 I 0 I Special Libraries .. . ... . . . .. 8 5 3 0 0 5 I '2 '2 2. Abstract I'23 113 IO 3 I6 94 6 4 2.'2 ----------------------- Colleges and Universities .... 67 62 5 I IO 51 3 '2 9 Public Libraries ... . ....... 39 35 4 '2 5 28 3 I II State Libraries .... . .. .. .. . 8 8 0 0 I 7 0 0 I Special Libraries 9 8 I 0 0 8 0 I I 3· Index I'26 1'25 I 16 35 74 0 I I9 ----------------------- Colleges and Universities . . .. 68 67 I 5 I6 46 0 I 8 Public Libraries ... . . ... . .. 41 4I 0 7 12 2.'2 0 0 9 State Libraries ... .. ... . .. . 8 8 0 2 5 I 0 0 I Special Libraries ....... .. .. 9 9 0 2 '2 5 0 0 I 4· Small Area Data Book I'2'2 II6 6 I 7 108 2 4 '23 ----------------------- Colleges and Universities . . .. 63 62 I 0 3 59 I 0 IJ Public Libraries .. . .. .... . . 4'2 38 4 I 3 34 I 3 8 State Libraries . ........... 8 8 0 0 I 7 0 0 I Special Libraries .... . ...... 9 8 I 0 0 8 0 I I s. Atlas 118 II2 6 I 5 106 3 3 '27 ------------ -------- ----- Colleges and Universities ... . 59 55 4 0 5 so '2 2 17 Public Libraries . .. .. ...... 4'2 40 '2 I 0 39 I I 8 State Libraries . ...... .. .. . 8 8 0 0 0 8 0 0 I Special Libraries ... . ....... 9 9 0 0 0 9 0 0 I 1 Includes 15 comments received as a result of a nnouncements of proposed reference works in the December 1948 W ilson Library Bulletin and Dec. r, 194 8 Library Journal. additional 15 libraries made suggestions as a result of the announcements regarding the proposed general reference works in the December .1948 Wilson Library Bulletin and the Dec. I, I 948, Library Journal. Practically all libraries which replied ex- pressed a high degree of interest in the five proposed reference volumes mentioned and pointed out the need for these library tools. indicated by the librarians who stated their order of preference. Librarians were unanimous in assigning highest priority to a comprehensive index to all publications of the 1950 Census. The comments expressed a strong need for a comprehensive detailed index which would analyze the tabular contents of all the 1950 Census volumes rather than merely list the 418 COLLEGE AND RESEARCH LIBRARIES general subject contents of each volume. Such an index should have the same degree of detail as a smaller subject index, for each volume but would consist of a single volume covering the entire census. Many librarians pointed out that the large number of census reports and the wide variety of subjects covered make it very difficult to use intel- ligently the census volumes without an ade- quate analytical index, and that a detailed index of this kind would immeasurably in- crease the accessibility and usefulness of the census information. Perhaps a special edi- tion of the Census Bureau's Catalog and Subject Guide could be prepared. Expressions of opinion from librarians who indicated the order of preference among the reference works ranked the pro- posed one-volume summary of the results of the census second in order of importance. Approximately four-fifths of the librarians who indicated their preferences assigned first or second place to a one-volume ab- stract of the census which would serve as a convenient and useful reference tool. Simi- lar one-volume summaries, called either a "compendium" or "abstract," were prepared for each decennial census during the period I85o to I930. However, no such summary was prepared for the I 940 Census because of limitations imposed by the War. Almost all (95 per cent) of the respond- ents also agreed that a small area data book summarizing the most important data col- lected for cities and cou'nties in a single vol- ume would constitute an important ref- erence tool. Many reference librarians pointed out that an increasing proportion of requests are for information on specific local areas. A similar compilation, called the County Data BookJ was published by the Bureau of the Census in I947· Other general publications proposed in- cluded a statistical atlas, a history of the OCTOBERJ 1949 census, a guide to the use of the census, state compendiums, and similar publications. While the need for these publications is not as great as is that for the three recom- mended above, it is strongly urged , that these additional publications be prepared, as funds permit. The three publications recommended above, the comprehensive index, the sum- mary volume, and the small area data book, however, are essential library tools and should be included as basic publications of the I 7th Decennial Census. Publication Style The following inquiries on preferred pub- lication style for the I950 Census were sub- mitted to librarians with statements on the previous experience of the Bureau of the Census: 1. "What information should be included on the title pages o{ the I 7th Decennial Census ( I950) publications to make them more useful and convenient than those of the I 6th Census ( I 940) ?" 2. "What improvement is suggested as to serialization of the Bureau's decennial cen- sus publications?" 3· "What improvement is suggested in the indication as to the preliminary or final nature of the data in the various publica- tions?" The Bureau's statement outlining the essential features of the I 940 Census in re- gard to the above problems noted that pre- vious publication practices had apparently not fully met the needs of catalogers and reference librarians. The statements men- tioned possible alternatives for improving publication style, but called attention to the fact that certain practical requirements in the preparation of census volumes made a simple solution of these problems very diffi- cult. 419 Title Page Format Replies to the query on title page format were received from I29 libraries. (See Table III.) The purpose of this query was to determine whether librarians pre- ferred a rather detailed title page, as in the I940 Census, or a much simplified title page. The preferences expressed by li- brarians on the choice of detailed versus of the title on each title page (e.g., United States Seventeenth Decennial Census: I950 Popul~tion Volume I, Part I). This would then be followed by the volume and part titles and other necessary information. It will be necessary, of course, to so plan the page that it will be clear to the user as to which is the volume and which is the part title. Table III Summary of Replies to Query on Title Page Format Preference for 1950 Census Type of Title Page Preferred Number Replies Percent of Total All . ... . .. . ... ........ . . ....... . . . . .... . .. . .. .. . . . . . . . . . 100 Simplified title page, _ total_. . ;, . . . . . .... ·,; . . ... . ... . .... .. . . . .. .. . T~tle page a~ ou_tlmed ·m "Comments, ....... . ........ .. .. . . . . 47 29 18 T1tle page d1ffenng from Comments . . ..... .. . ... ... . .. .. . . . Detailed title page, total . . ...... . .. .. ..... . . ... ..... . .... ... ... . 68 57 53 44 9 Title page as used in 1940 Census . .. . .. . ...... . . . .. ...... . . . . Title page differing from 1940 Census . . ..... . . . . .. ... .. . . .. .. . I I simplified title f)age were almost evenly di- vided. However the comments and sugges- tions regarding this question varied con- siderably with respect to the exact amount and kind of information which should be included on the title page. The nature of the comments indicated that reference li- brarians apparently prefer detailed title pages, while those concerned with acquisi- tions and cataloging leaned toward limit- ing the information on the title page to the minimum required for identification and convenience in shelving. In view of these returns, it appears. that a title page format having the following characteristics might fulfil the requirements of most librarians. The title page should show detail similar to that shown on the I940 Population Census title pages but would group the series name, subject group, and volume and part notation at the head Serialization of Census Reports Approximately I 20 librarians made sug- gestions and recommendations on the serial- ization of the decennial census reports. These recommendations almost unanimously favored more complete serialization· of the census reports, but differed considerably with respect to preferred methods of or- ganization. As shown in Table IV, slightly more than one-third (36 per cent) of the librarians recommended some form of complete serializatio'n; almost the same pro- portion ( 3 I per cent) favored partial serial- ization, while a fourth of the respondents preferred serialization of all reports or as many reports as possible. Some of the re- spondents favoring complete serialization recommended the numbering of all census reports in one consecutive series, while others suggested that all reports on a given subject be numbered in the same series. 420 COLLEGE AND RESEARCH LIBRARIES Table IV Summary of Replies on Preferred Method of Serializing 1950 Census Reports Replies Method of Serialization Preferred Number Percent of Total All ... . ....................................... . ....... . . I I9 IOO Complete Serialization, total . ............ . . .. .. .. .. .. ...... . .... . . 43 36 Serialize all census reports in one consecutive series .............. . 26 22 Serialize all reports on same subject in same series . . . . .. . . . . . .... . 17 I4 37 31 I I 9 Partial S erialization, total . ..... . .......... .. .. . .. . .. . .. . ...... . . . Serialize all reports as in 1940 Census ....... . . . ... . ...... . ... . . . 8 7 I8 IS Serialize all special reports consecutively in one series ... . . . ...... . Serialize special reports separately, within subject . . . ..... . .. . .. . . Serialize All or as many Reports as Possible, total .... . ... . . .. . .. .... . 28 24 Other Methods, total . ... .. .. . . . ... . . .. .. . .... . . . ... .... ...... . .. . II 9 Preferences of librarians recommending partial serialization suggest · three general methods. One group recommended that all special reports be serialized separately from the basic reports by numbering all special reports, irrespective of subject, in one con- secutive series. Another group preferred the serialization of special reports within broad subject groups. The third group suggested the same method of serialization used for the I 940 Census reports. Considering the divergent viewpoints of librarians and the practical difficulties of consecutively numbering all reports of the 1950 Census, it appears that numbering or serializing all 1950 Census reports within broad major subject groups would best meet the needs of most librarians. Classification Table V Summary of Recommendations for Improving Identification of Preliminary or Final Nature of I950 Census Data and Reports Type of Recommendation All . . . .. .. . .................... . ... . ..................... . ........ . .. . . . Identify all reports of 1950 census including preliminary and advance reports with month and day of issue and indicate which earlier reports are superseded and, if possible, which subsequent reports will serve as replacements ......... . ....... .. .. .. ..... . .. . Use of different color paper to di s tinguish preliminary from final reports . .. .. ... .. . .. . . Prepare s tandard definitions of "preliminary," "advance," "final," etc. and uniformly identify all census reports .. . ... .. . . . . ... .... . . .... . .. . .. .. . .. ..... . . .... .. . ... . Publish list of final reports ..... . ... ... ..... . . .. .. . ... . . .. .... .... .. .... ...... ... . Publish at intervals, li s tings of superseded and final reports ............ . ... . ..... . . . . Notify libraries when preliminary reports become final . . .. .... . ......... . . .. ... . .. . . Publish only one preliminary and one final series of reports . ... .. . .. . ... . . .. . ... .... . Identify only final series of reports ......... .. . ... .... . ... .. ...... .. . .. ... .. ... .. .. . Eliminate all preliminary reports . .. .. ... ... .. .. .. . . . . .... .. .... .... .. ...... .. ... . Place notification in Monthly Catalog of U. S. Government Publications when preliminary reports become final .... ............... . ....... . ... . . . .. . ... .. ... .. .. . . .... .. . Distinguish "preliminary data" from "preliminary format" ..... . .. . .. . ....... .. .... . Use 1'partial" for incomplete data, "complete" for final data ... . . . . . ... .. .. . . . .... . . Other .. ... .... . . . . . . . .. .... ...... . .... . . . .... . ... .... .. . . .. .. ........ .. . .. ... . OCTOBERJ 1949 Number 5 3 4 2 2 2 2 I I6 421 of reports by major subject group would in- crease their usefulness for reference pur-· poses, while the numbering of both basic and special reports would facilitate their cataloging. Identification of Preliminary or Final N a- ture of 1950 Census Data and Reports As shown in Table V, the majority of the 9I respondents to this query recom- mended that all reports of the I950 Census, especially the preliminary and advance re- ports, be identified by month and date of issue and that each report show which re- port or reports it supersedes or which sub- sequent reports will replace it. Numerous other suggestions for distin- guishing between preliminary and final re- ports were also received. Some comments suggested the use of a different color paper for preliminary and final reports. Others recommended the publication of a list of final reports; publication at intervals of lists of supe~seded and final reports; or in- forming libraries when preliminary reports become final, etc. It is apparent from the nature and variety of replies to this query that the lack of clear- cut identification of the preliminary or final nature of previous census publications has handicapped librarians. Adoption of an identification system for the I 9 50 Census publications which would show the month and date of issue of each report as well as its relation to prior and subsequent reports would not only meet the requirements of most librarians but would enable the Cen- sus Bureau to satisfy the demands for pub- lication of data at the earliest possible mo- ment. Recommendations on Proposed Reference Works and Publication Policy for the 17th Decennial Census RECOMMENDATION I : A comprehensive index to all publications of the I 7th Decen- nial Census should be prepared. RECOMMENDATION 2: A one-volume s,ummary of the results of the census should be prepared. RECOMMENDATION 3 : A small area data book should be prepared. RECOMMENDATION 4: The title page used in the reports of the I 7th census should be standardized and show detail similar to that of I940 Population Census title pages. However, at the head of the _title on each title page should appear the serialized sub- ject title with volume and part notation (e.g. United States Seventeenth Decennial Census: I950 Population Volume I, Part I). RECOMMENDATION 5 : All reports of the I 7th Decennial Census should be numbered or serialized within broad subject groups, such as population, agriculture, etc. RECOMMENDATION 6: All reports of the I 7th census should show the date of issue. Preliminary and advance reports in particu- lar should be identified as to month and day of issue. RECOMMENDATION 7: Each report should show which earlier reports are super- seded by it and, to the extent possible, which subsequent reports will replace this report. 422 COLLEGE AND RESEARCH LIBRARIES