The feasibility of developing an Australian thesaurus was discussed at the Towards Federation 2001 Review Meeting in December 1993, and it was resolved (resolution RC18) that the Australian Committee on Cataloguing (ACOC) should conduct a study of the feasibility of developing a single Australian thesaurus. ACOC commenced the feasibility study with a survey of thesaurus usage in Australian libraries. The survey was conducted by AIMA Training and Consultancy Services, and funded by the Working Group on High Priority Cross Sectoral Projects.
The report on the survey was considered by ACOC at its August 1996 meeting. ACOC reaffirmed its joint decision of 1995 with the ALIA Cataloguers' Section that the List of Australian subject headings (LASH) be incorporated into the Australian additions and changes to Library of Congress subject headings, using the ABN Subject Headings Review Panel as the mechanism for consultation.
ACOC noted the recommendations in the AIMA report of the survey of thesaurus use in Australian libraries. While ACOC considered that there is no demonstrated case for the development of a new Australian thesaurus, it would like to encourage developers of thesauri to facilitate cross database searching by achieving consistency in the selection of terms where possible.
With regard to the names for Australian languages, ACOC noted the convergence between the Aboriginal and Torres Strait Islander thesaurus and ABN practice, and encouraged the compilers of both lists to continue this convergence.
ACOC agreed to consult the Local Studies Section of ALIA about the development of suitable subject terms for access to local studies collections.
In September 1996, the ABN Standards Committee endorsed a detailed plan of action for the incorporation of LASH in the Library of Congress Subject Headings on ABN, and work commenced in the National Library in September 1996. The work will result in a useful body of Australian subject headings for implementation in WORLD 1.
The report from AIMA Training and Consultancy Services is presented below:
AIMA Training and Consultancy Services, in conjunction with Dr Peter Clayton from the University of Canberra, is delighted to present this report on thesaurus use in Australian libraries.
This report forms part of a project which has the following overall aim:
Specifically AIMA undertook to:
During the course of this study AIMA took into account previous work in this area by Margaret Henty. The Henty study, Indexes in Australian Libraries: a towards Federation 2001 survey, provided a listing of the indexes used for Australian newspapers and journals in Australian university libraries, state/territory libraries and parliamentary libraries as well as the National Library of Australia.
The results of this survey indicate there could be a need for a single Australian thesaurus which draws upon terms appearing in APAIS, LASH and ABN and that the content include terms relating to Aboriginals and Torres Strait Islanders and local history.
The recommendations are:
That a single thesaurus of Australian terms be developed, which draws in part upon the terms appearing in the present APAIS, LASH and ABN Authorities listings.
That this Australian thesaurus must include, as an integral part, terms which cover Aboriginal communities, language groups, clans etc.
That the questions these respondents raised in relation to local history be explored with those working in this area, including respondents to our survey. It would also seem appropriate to contact the Local Studies Section of the Australian Library and Information Association.
The survey methodology submitted by AIMA in its project proposal was modified slightly after consultation with Pam Dunlop and Julia Trainor from the Australian Committee On Cataloguing (ACOC).
The survey was also developed during this consultation and the content of the letter to accompany the survey was agreed.
In fact two covering letters were prepared, one to go to university and state libraries identifying the difference between the data collected in this survey and that previously collected by Margaret Henty. The other letter was sent to public and special libraries and simply explained the requirements of this survey.
AIMA used a standard survey process with a stratified sample of Australian libraries. The details of the survey and its design are discussed in Section 3.
One difficulty not anticipated in the implementation of the survey was the slowness of the libraries in responding. This meant that the original draft report date of 23 March could not be met because most of the survey responses had not been received by this date. A brief progress report was provided to the Committee. The lateness of responses also caused a delay in the dispatch time of follow up letters.
However, as is noted in Section 3, the final response rate was considered to be excellent and the delay in early reporting was balanced by having better data to work with in the final report. It is worth noting that one response was received from a large university library on 22 May. This response was included in the final calculations and data analysis.
This project involved the design and execution of a survey of Australian libraries. A questionnaire was designed in consultation with the ACOC. Design considerations were:
Draft questions were discussed with ACOC until agreement was reached on format and wording. A copy of the final questionnaire is included as Appendix A to this report. It should be noted that the first question was a lead to further action. If the response to this question was 'no' the respondent was not required to complete the return any further other than providing contact details.
In our initial discussions with ACOC regarding this project, it had been decided to survey a representative group of Australian libraries rather than all libraries. In our judgment, little additional information of value would have been obtained by attempting to survey all libraries whilst the increased cost of such an approach would have been considerable. A decision was made to draw a stratified random sample of Australian libraries comprised of:
Despite their large number, school libraries were deliberately omitted as unlikely to be involved in individual thesaurus activity. This gave a total sample of 280, with relative coverage of public and special libraries which took account of both their likely involvement in thesaurus activity and the numbers of libraries within each category. A list of the libraries surveyed is provided in Appendix B.
In order to test that the questionnaire which had been developed was straightforward for respondents to complete, devoid of ambiguity and readily permitted data entry, a pilot survey was undertaken. Ten pilot study recipients were chosen, including a state library, two universities, three public libraries and four special libraries. The early return of the pilot surveys from some of these libraries established that there were no apparent problems with the questionnaire or the overall approach adopted. No amendments were made to the questionnaire.
A covering letter from both the Chair of ACOC and the Executive Director of AIMA was prepared, and mailed with the questionnaire to the libraries selected. Because the questionnaire was unaltered from the pilot, returns from the pilot participants have been included with those of other respondents in the following sections of this report. A copy of both letters are provided in Appendix C.
In order to maximise response, a follow up was conducted in the week following the original due date suggested on the questionnaire form. This follow up resulted in the receipt of a substantial number of additional returns. No further follow ups were possible, due to the need to meet ACOC's time table for the completion of this project.
Responses were received from 139 institutions out of the 280 selected for this project. This is a 50% response rate, calculated on an institutional basis, which is excellent for studies of this kind. In fact, those 139 institutions provided 156 responses, as 13 institutions (all large libraries) provided an additional 17 responses. These reflect different areas of activity within those libraries.
The response rate varied greatly according to the category from which respondents came. A 100% response rate was obtained from the National Library, state libraries, and the ACT Library Service. A 61% response was obtained from the university libraries. In both these cases, all eligible libraries had been surveyed. However, only 42% of the random sample of public libraries responded to the survey, and only 38% of the special libraries.
In any survey in which a response rate of this approximate size is obtained, the questions have to be asked:
With this survey, it seems fairly clear that the libraries who responded were those who were involved in some form of thesaurus activity. A minority of respondents reported that they did not have a need for a thesaurus of Australian terms. In this case, it does seem clear that the non-respondents were almost certainly not interested in the subject of this survey.
We conclude that, despite its 50% response rate, this survey does provide a reliable and useable estimate of the likely extent of thesaurus activity in Australian libraries. Most smaller libraries - those from whom we obtained the lowest response rate - simply do not have the resources to undertake activity of this kind. Indeed, several of those from smaller libraries who did take the trouble to respond to our survey commented in just these terms upon it:
"This library is a very small, part-time run one."
"Would have minimal time to assist."
Special Libraries, from whom the response was lowest, also recognised that their particular needs were likely to be highly specific and possibly unique:
"Our needs are subject specific."
"A general thesaurus would probably be of little use to us."
In this section, the responses to each of the questions we asked will be discussed in turn. A summary table of the complete results will be provided to ACOC as an Excel spreadsheet on disk and is printed in Appendix E, along with the coding scheme used to enter the data. Every survey response will also be provided.
Additional written comments were invited as part of the questionnaire and a number are reported here. The full text of all additional comments is provided in Appendix F. So too, are the contact details for the individual librarians who completed the survey should ACOC wish to follow up any particular suggestions or observations with these. [Note that appendices and charts illustrating responses are not included here].
Question 1
Do you have a need for any purpose, for a thesaurus of Australian terms beyond those included in the international lists of subject headings or thesauri such as LCSH, MeSH?
Ninety-five respondents or 50% did see a need for a thesaurus of Australian terms; 56 or 33% did not. Five respondents omitted to answer this particular question. Non-respondents to a particular question have been noted throughout as we believe they show the full picture of potential interest or disinterest in a particular area.
This overall finding was noted above, in the discussion of the response rate. It suggests that those libraries most likely to be involved in thesauri activity were also most likely to have responded to this survey.
As noted in Section 3 Survey Design, respondents who answered 'no' to this question were not required to complete the remainder of the questionnaire.
Question 2
Do you use a thesaurus of Australian terms to meet this need?
Sixty-four respondents or 50% did use a thesaurus of Australian terms; 32 or 33% did not. There were four non-respondents or 17% to this question.
In other words, of those institutions which had a need for a thesaurus of Australian terms beyond those included in the international lists, two thirds had identified an existing thesaurus which met all or some of their needs.
Question 3(i)
Do you use a published thesaurus of Australian terms?
Over three quarters or 75% of respondents to this question were using a published thesaurus (48 out of 63 respondents).
Question 3 (ii)
List the one/s you use, eg APAIS or LASH (List of Australian Subject Headings 2nd ed).
When asked to nominate the published thesauri involved, 31 different thesauri were nominated - see Table 1.
It will be apparent that three thesauri are widely used, APAIS, LASH and ABN Authorities. No other thesaurus was mentioned more than four times, with most thesauri being mentioned a single time only.
This clustering is highly significant. If it were possible to develop a single Australian thesaurus which reflected the use of APAIS, LASH and ABN Authorities, this would meet an overwhelming proportion of the expressed need for a thesaurus of Australian terms beyond those included in the standard international listings.
However, the existing published Australian thesauri do not meet all of the respondents' needs. Only 20 respondents or 17% indicated that they did; 43 respondents or 33% indicated that they did not meet all their needs. However, it should be noted that a very large number of respondents (37 or 50%) did not answer this question. Overall, responses to this question do suggest that further enhancement of the existing Australian thesauri is seen as desirable by many respondents.
Name Number of Times
Listed
APAIS 27
LASH 17
ABN Authorities 13
Family 4
Australian Education Descriptors 4
MAIS 3
Legal Thesaurus 2
SCIS 2
IREL 2
Australian Ephemera Thesaurus 1
LCSH 1
SAGE 1
John Oxley Library Pictorial Thesaurus 1
Australian Tourism Thesaurus 1
TACT 1
Taxindex 1
Macquarie Concise 1
LOCOM 1
Local History Headings 1
Gippsdoc 1
Guidelines 1
AIFS 1
OPAC 1
NCJRS 1
AIDS 1
EIM 1
AABT 1
NCBA 1
BTE 1
PICMAN 1
Aust. Industrial Relations Thesaurus 1
Human Services Thesaurus 1
Nat. Gallery of Aust. Working Names Authority 1
AGLIBs 1
Australian 1:250,000 & 1:100,000 Master Names 1
Gazetteer
Question 4
Do you supplement published thesauri for Australian materials?
Thirty-six respondents or 36% indicated that they supplement the published thesauri to meet additional Australian needs; 30 respondents or 30% do not do this. There was a large number, 34 or 34% of respondents who did not answer this question.
Question 5
Do you create or use in-house/local headings for Australian materials?
As can be seen from Chart 5, many respondents create or use in-house or local headings for Australian materials: 53 or 53%. Although only 21 or 21% indicated they did not create or use local headings, a further 26 did not answer this question.
Question 6
What is the subject and/or format emphasis of these headings? (eg social history, photographs).
Fifty-four respondents nominated general subject areas in which they created additional in house or local headings for Australian material. Many of these patterns were, of course, predictable: law libraries suggested legal headings, and so on. A very wide scatter of topic areas were suggested, from which it is difficult to discern meaningful patterns - some small libraries suggested a large number of subject areas; some larger libraries did not suggest any. As noted above we will make available to ACOC all responses so that the range of suggestions in this area can be investigated further if desired. Some issues which were raised by several respondents will also be noted in the conclusions and recommendations which follow in Section 6.
Question 7 (i)
Are these headings developed by your own organisation?
Fifty respondents indicated that the headings were developed within their own organisations.
Are these headings developed cooperatively with other organisations?
Only 12 organisations indicated that they developed additional subject headings cooperatively with other organisations.
Question 7 (ii)
What format are these headings in?
The formats reported were as follows:
- electronic 39
- paper 16
- card 5
- other 7.
The formats nominated in the 'other' category were:
- "informal listings"
- "library catalogue authorities on-line"
- "subject list in catalogue system"
- "on-line authority file is its own thesaurus".
Overall, these findings suggest there is minimal cooperative activity in the development of additional Australian subject headings. This seems a very small base upon which future collaborative thesaurus development could be built.
Question 8
Are you satisfied with this thesaurus?
Only 50 respondents or 50% of the possible respondents answered this question, of whom 30 or 30% were satisfied with the thesaurus which they were developing themselves, 20 or 20% were not satisfied. Forty-nine respondents or 50% did not respond.
Question 9
If a commonly accepted thesaurus of Australian terms was to be developed would your organisation use it?
Eighty-two respondents would use a thesaurus of Australian terms were one to be developed. (Note that this group also represented 82% of the 99 who might have been expected to reply to this question). Only two respondents would not use an Australian thesaurus; however, 16 did not respond to this question.
It has to be said that this suggests there would appear to be moderate support for the development of a new thesaurus of Australian terms.
Question 10
[If a commonly accepted thesaurus of Australian terms was to be developed would your organisation].... Purchase it?
Here, 80 of the 82 institutions which would use such a thesaurus were it to be developed would be prepared to purchase it. The two respondents who would use such a thesaurus but did not indicate they would also be prepared to purchase it, in fact did not respond to this question.
Questions 11 and 12
[If a commonly accepted thesaurus of Australian terms was to be developed would your organisation].... Assist with its compilation and enhancement? Contribute your own in-house database to its development?
Fewer libraries were prepared to commit themselves to this: only 50, about one third of all respondents. However, only 27 indicated they would not be prepared to contribute to its development. Examination of the individual returns reveals that many respondents have written in:
- "possibly"
- "we are not sure that our headings would add any depth to the thesaurus"
- "would consider positive responses to questions 11 and 12 when more detail is available"
and so on. Given the large number of smaller special libraries participating in this survey, such responses are understandable.
Question 13
Additional comments
Finally, respondents were given an opportunity to make any additional comments they wished upon this survey. Nearly a third (48) availed themselves of this opportunity. As would be expected, many respondents commented on how closely their library systems were tied in with ABN, on a need to maintain consistency with national agency data, and on the need to avoid expensive in-house manipulation of standard records. Typical comments included:
"We conform very closely to LCSH, supplemented by ABN, many of whose headings have become part of LCSH. We occasionally are forced to created topical headings when nothing fits, but we avoid this as far as possible. Unless an Australian thesaurus was adopted by LCSH we would not use it, except to create see references."
"We prefer to use a nationally accepted thesaurus to minimise changes to copy cataloguing and simplify processing. The proposed changes to make World 1 headings for Australian subjects will be very useful."
"A thesaurus of Australian terms is good in principle. However, local enhancement of catalogue records using any additional thesaurus, has workflow and resource implications. This enhancement would have to be done centrally for it to be practicable/feasible."
What has this survey found? If one accepts that non-respondents to it were not particularly interested in further development of standard Australian subject thesauri, then only about a third of Australian libraries have a need for a thesaurus of Australian terms beyond those included in the standard international listings (95 respondents out of a total sample sent of 280). Of these 95, slightly fewer would use, purchase or contribute to the development of such a thesaurus (82, 80 and 50 respondents respectively).
Given that our respondents included most of the largest libraries in Australia, and all of the national and state libraries, this suggests that the market of a thesaurus of Australian terms is real but limited. No doubt if one were published, schools of librarianship, not included in this survey, would also contribute to it and purchase it - presumably in multiple copies for student use.
There can be no doubt that to be commonly accepted, a thesaurus of Australian terms would need to be based upon the three existing published Australian thesauri most widely used: APAIS, LASH and ABN Authorities. It is noteworthy that, in response to our question 3(ii), many libraries reporting that they used published thesauri of Australian terms in fact listed several.
Compatibility with standard bibliographic data provided by ABN, and by World 1 in future, is understandably a priority for most Australian libraries. This strikes us a sine qua non: given full compatibility with World 1, and that the records which the National Library of Australia generates for its own original Australian cataloguing utilise the headings in the new Australian thesaurus, then it will be welcomed by many Australian libraries. A substantial number of these libraries will also purchase it, and almost certainly it will be accepted by all. However, if it is not compatible or not fully compatible, at best it will only have a limited impact.
Accordingly, our first recommendation is:
That a single thesaurus of Australian terms be developed, which draws in part upon the terms appearing in the present APAIS, LASH and ABN Authorities listings.
However, we see such a project as viable only on the basis that it is completely compatible with World 1, and that libraries in Australia using World One records for copy cataloguing will be able to receive records for newly catalogued items which incorporate subject headings based upon this thesaurus. If this condition cannot be met, then in our judgment, the project should not proceed.
In considering whether to proceed with such a project, it should be noted that our survey found only very limited collaborative thesaurus building activity taking place in Australian libraries.
If the development of a new Australian subject thesaurus is decided upon then we would offer two additional recommendations based upon some of the comments in responses to the survey we undertook:
Some respondents, including the Northern Territory University, drew attention to the need for more adequate coverage of terms relating to Aboriginals and Torres Strait Islanders. We understand that the Northern Territory University is in fact developing an Aboriginal and Torres Strait island thesaurus for publication later this year. Another respondent commented,
"there is a need for thesaurus of Aboriginal communities, language groups, clans etc to assist with subject description of works."
We concur, and recommend:
That any Australian subject thesaurus must include, as an integral part, terms which cover Aboriginal communities, language groups, clans etc.
Appendix D. contains a brochure provided by the Chief Librarian, Northern Territory University on the INTAN MAS database. [Brochure not included here].
We note that several public libraries commented on their need for enhanced descriptors for assistance with their maintenance of local history collections. Comments included:
"Our local history collection (photos, documents, maps, newspaper indexes, historical society publications, indexes) uses LCSH - style headings. We create a heading where there is not a suitable LCSH equivalent eg produce merchants, fruit shops."
"We are unsure of the extent to which a standardised list could provide headings of value to those working with such collections - who will inevitably have very limited bibliographic and other resources upon which to draw. How much of the material which they need will be unique to their own collection?"
Such questions cannot be answered from the limited data provided by this survey. Instead, we recommend:
That the questions these respondents raised be explored with those working in this area, including respondents to our survey. It would also seem appropriate to contact the Local Studies Section of the Australian Library and Information Association.
The concerns noted above were each raised by several respondents to our survey. In addition, individual respondents raised particular issues which, although not otherwise mentioned may seem worthy of further consideration. As an example, the University of Adelaide response commented that:
"Australian plants and animals are poorly represented in LCSH; an Australian list would make searching for specific terms using subject terms easier. Such a list would have to consider the use of scientific names rather than the LCSH practice of preferring common names."
A quick perusal of the survey returns may suggest a small number of other, specific matters which any group working upon the development of a new Australian thesaurus may wish to take into account.
In conclusion we should like to thank ACOC for the opportunity of working upon this project. We hope that it will assist in the development of library and information services in this country which provide greater ease of access to specifically Australian material. This will be in the interests of all Australians, as well those interested in our country, and can only help in the further development of our national identity and culture.