Child pages
  • Party Data Survey

ARDC Party Infrastructure Project

Party Data Survey: Summary of findings

By Natasha Simons

Business Analyst, Australian Research Data Commons Party Infrastructure Project

National Library of Australia


About the survey


The Australian Research Data Commons Party Infrastructure Project team designed and conducted an online survey to gather information about party data from managers of this data in the Australian research sector. The survey was conducted over the internet, contained 14 questions and was open from Wednesday May 19 to Friday June 11, 2010. Emails inviting people to participate were sent to the Australian National Data Service (ANDS) and Institutional Repository (IR) communities. There were 43 responses. 24 completed the survey. Some of the questions were quite complex and respondents were asked to answer to the best of their knowledge. Responses to the survey will be used to better inform the project team about the current state of party data in the research sector and to identify potential contributor sources.


Brief summary of main conclusions


  1. Party data to build comprehensive profiles of researchers does not exist in one single source. Aspects of party data are spread across systems.
  2. The aspects of party data that is stored vary widely between systems and between institutions.
  3. There is minimal, if any, authority control for party records outside of library databases though survey participants recognised the need for it.
  4. On the whole, party data identifiers take the form of a name or an email address over a URL or URI.
  5. Research Office party data tends to be private.
  6. Party data exists in varying formats. Most survey participants were unsure or did not know in what format their party records could be provided to the NLA.
  7. The majority of survey participants could provide party records via OAI.
  8. Institutional privacy policies and/or requirements for individual researcher approval may restrict contribution of party records to the NLA.
  9. There is a general willingness to share party data that is tempered by institutional requirements and approval.


Related documents

  • PartyInfrastructureSurveyQuestions.pdf

A note on analysis logic

The survey was conducted over the internet using Survey Monkey (see ) to design, collect and analyse responses. There were 14 questions. There were no mandatory questions and no ‘either/or’ questions.

Survey Monkey analysed the number of respondents who participated in the survey and the number of respondents who completed the survey based on the following logic (as described in Survey Monkey’s Frequently Asked Questions):

“Those that have clicked the [Done] button [this is the button that concludes the survey – NS] AND answered at least one question will be included in the Total Completed Survey number.   However, it does not mean that ALL questions have been answered on every page. The respondent may have only answered a few questions on each page. It means that the respondent moved through all pages using the navigation buttons, clicked the [Next] button on every page, answered at least one question, and finally clicked the [Done] button.”

The percentages shown in this summary were calculated by Survey Monkey. They were calculated using the following logic:


  1. Where there were multiple choice questions (e.g. ‘yes’, ‘no’, ‘unsure/unknown’) the percentage was calculated using the total number of number of respondents to the question and the number of respondents selecting each possible answer to the question. The number of respondents who skipped the question was noted but not included in the percentage total. For example, if there were 12 respondents and 3 of them answered ‘yes’ then the percentage was 25% for ‘yes’. As none of the questions in the survey were ‘either/or’ questions, it was possible for the percentage total to exceed 100%.


  1. Where there were matrix questions, the percentage was calculated for each row of choices based on the total number of respondents for each row. For example, if there were 25 total respondents to a matrix question, and 6 of these responded to the first row of the matrix question and 3 of these selected ‘institutional repository’ then the percentage was 50% for ‘institutional repository’. The number of respondents for the matrix question as a whole was noted, as was the number who skipped the question.


Summary by section


Section one: about the survey


Only one respondent said they did not collect party data for their researchers or research organisations (and were therefore advised to discontinue the survey).


Section two: about you


Of those who listed their institution, these included:

  • 17 Universities [Note: none from Western Australia, South Australia or Northern Territory]
  • Education Services Australia
  • Northern Territory Department of Health and Families


Of those who gave a brief description of their position or role within their institution, this included:

  • Project Managers
  • Repository Managers
  • eResearch specialist
  • IT support for research
  • Senior Data Specialist
  • Deputy Director
  • Digital/Electronic Services Librarian
  • Research Services Librarian
  • Director of eMarine information infrastructure
  • Digital Infrastructure Manager
  • Data Management Coordinator


Section three: about your party data


Which systems or databases are used to store party data (participants could select more than one where applicable):


  • Institutional Repository 75.9%   (22 of 29 respondents)
  • Human Resources database 72.4%   (21 of 29 respondents)
  • Research Office database 65.5%   (19 of 29 respondents)
  • Staff/researcher web directory 58.6%   (17 of 29 respondents)
  • Library Management System 34.5%   (10 of 29 respondents)


Other sources for storage of party data mentioned in (optional) further comments included:

  • IT system
  • Student system
  • Social networking service [provides Australian education and training professionals with an online networking and profile space.]
  • Integrated system (finance+HR+research management processes)
  • NicNames software
  • Research Supervisor Connect
  • Faculty and department web pages
  • Research management system (research office+information services+faculty research coordinators)
  • Two developing projects: staff profile project (draws on various systems); custom solution (identity management systems+repository data)


What party data is PUBLICLY available and where it is stored (participants could select more than one where applicable):

[Note: only the top 2 responses are included below]




Name variants:

  • Institutional Repository 53.8%
  • Human Resources database 38.5%



  • Staff/researcher web directory 85.7%
  • Human Resources database 23.8%


Dates – birth or death

  • Library Management System 62.5%
  • Human Resources database 50%

[note: 0% stored in IRs or staff/researcher web directories]


Field of Research (FoR) codes:

  • Institutional Repository 100%
  • Research Office database 30%


Research field/keywords:

  • Staff/researcher web directory 60%
  • Institutional Repository 53.3%


Other research classification:

  • Research Office database 50%
  • Other 50%

[note: 0% stored in Library Management System]



  • Staff/researcher web directory 76.2%
  • Institutional Repository 57.1%




  • Staff/researcher web directory 60%
  • Other 33.3%



  • Staff/researcher web directory 62.5%
  • Other 37.5%


Biographical information:

  • Staff/researcher web directory 78.6%
  • Institutional Repository/Other 21.4%


Contact details:

  • Staff/researcher web directory 85%
  • Human Resources database 25%


Staff identifier that is a URI or URL:

  • Institutional Repository 66.7%
  • Staff/researcher web directory 50%


Staff identifier that is NOT  a URI or URL (e.g. email address):

  • Staff/researcher web directory 93.3%
  • Institutional Repository 33.3%



  • Institutional Repository 69.2%
  • Staff/researcher web directory 53.8%



  • Institutional Repository 77.3%
  • Staff/researcher web directory 45.5%



  • Staff/researcher web directory 54.5%
  • Institutional Repository 36.4%


Dates – Grants:

  • Staff/researcher web directory 66.7%
  • Research Office database/other 33.3%

[note: 0% stored in Institutional Repositories, Library Management Systems and Human Resources databases]


Dates – Teaching or employment period

  • Human Resources Database 50%
  • Staff/researcher web directory; Research Office database; Other 33.3%


Professional membership:

  • Staff/researcher web directory 54.5%
  • Other 27.3%


Personal web page or blog:

  • Staff/researcher web directory 50%
  • Other 41.7%





  • Staff/researcher web directory 61.5%
  • Institutional Repository/Other 23.1%



The form of the PUBLIC party identifier and where it is stored (participants could select more than one where applicable):

[Note: only the top response is included below]


An identifier is stored but it is PRIVATE (9 of 12 respondents):

  • Research Office Database 75%


The identifier is an email address (6 of 7 responses):

  • Staff/researcher web directory 85.7%


The identifier is a name (6 of 8 respondents):

  • Staff/researcher web directory 75%


The identifier is a URL (2 of 4 responses):

  • Institutional Repository/Other 50%


An identifier is not stored (2 of 2 responses):

  • Staff/researcher web directory 50%
  • Institutional Repository 50%


The identifier is a URI (1 of 1 response):

  • Institutional Repository







Use of authority control for party data:


  • Yes 54.5% (12 of 22 respondents)
  • No 50% (11 of 22 respondents)

[Note: the total percentage is more than 100%. This can be explained by one respondent who answered yes (for institutional repository) and also no (for Library management system).]


Further comments (optional):

  • There were a number of comments about plans to do this or efforts to do this though little on what could be used for authority control or types of authority files in use.


Section four: sharing your party data


PUBLIC party data can be made available to the NLA:


  • Unsure/unknown 56%   (14 of 25 respondents)
  • Yes 40%   (10 of 25 respondents)
  • No 4.0%   (1 of 25 respondents)


Further comments (optional):

  • Subject to institutional signoff
  • Subject to privacy policy check
  • Subject to individual researcher approval
  • Subject to be able to provide the data in a suitable format
  • Subject to provision of further information/consultation


PUBLIC party data can be provided to the NLA in the following formats (participants could select more than one where applicable):


  • Unsure/unknown 54.5%   (12 of 22 respondents)
  • MARCXML or Excel spreadsheet 27.3%   (6 for each out of 22 respondents)
  • RIF-CS 13.6%   (3 of 22 respondents)
  • RDF 9.1%  (2 of 22 respondents)
  • EAC 4.5%   (1 of 22 respondents)


Other formats suggested in optional further comments:

  • MODS
  • Dublin Core
  • A couple of people commented that they could probably provide in any format required subject to consultation and time/effort required.




PUBLIC party data can be provided to the NLA via the following:


  • OAI 93.8%   (15 of 16 respondents)
  • SRU 6.3%   (1 of 16 respondents)


Other formats specified in (optional) comments:

  • Libraries Australia upload from catalogue of MARC authority data


Institutional privacy policy that may impact on sharing party data:


  • Unsure/unknown 50%   (12 of 24 respondents)
  • Yes 37.5%   (9 of 24 respondents)
  • No 12.5%  (3 of 24 respondents)


Further (optional) comments included:

  • Some general comments about requiring institutional approval and selection of what aspects of party data to share.
  • Some referrals to institutional privacy policy documents available on the web.


Section five: size of your party data


Quantity of PUBLIC party records [Note: there was a blank box beside each option and participants could complete both options where applicable]:


  • Contemporary identities: 100% (20 of 20 respondents). Numbers ranged from 120 to 40, 000 records.
  • Historical identities: 75% (15 of 20 respondents). Numbers ranged from 0 to 33, 000.


Section six: further comments


Further (optional) comments included:


  • Party data to build comprehensive profiles of researchers does not exist in one single source.
  • A consistent way to manage party data would be much appreciated.
  • It needs to be clear who has the responsibility for maintaining party data.




Appendix - Table of Acronyms



Australian National Data Service


Commonwealth Scientific and Industrial Research Organisation


Encoded Archival Context


Human Resources


Institutional Repository


Information Technology


MAchine-Readable Cataloging


MARC in Extensible Markup Language (XML) format


Metadata Object Description Standard


National Library of Australia


Open Archives Initiative


Pacific And Regional Archive for Digital Sources in Endangered Cultures


Resource Description Framework


Registry Interchange Format - Collections and Services


Search/Retrieval via URL


Uniform Resource Locator/Uniform Resource Identifier