- What are the Major Issues which Affect Access?
- The Argument for Data Sharing
- Preserving and Disseminating Data
- Basic Procedures of Archiving Data
- Good Practice Guidelines and Codes
Data Access and Data Sharing
This section deals with those issues which are involved in the sharing and preservation of data. There are numerous sources of information which interested researchers can, and should, consult at the very start of a data collection exercise. One of the more comprehensive is “Preserving & Sharing Statistical Material”, produced in 2002 by the Working Group on the Preservation and Sharing of Statistical Material: Information for Data Producers, of The Royal Statistical Society and the UK Data Archive which can be purchased or downloaded from the data archive site.
What are the Major Issues which Affect Access?
Although the desirability of sharing research data is generally accepted, there are a number of factors which still act as barriers to wide-spread data sharing. These can affect access to all types of quantitative data, whatever its source or coverage. Among these are:
- Structural, Institutional and Management Issues.
- Concerns of Data Providors. This can include concerns about deliberate or inadvertent misuse of the data, apprehension about resources needed to document and disseminate data, necessity to generate revenue, concern to protect intellectual capital and competitive advantage.
- Legal and Policy Issues
- Data Protection and Confidentiality Concerns.
The Argument for Data Sharing
These were discussed in an early publication, Sharing Research Data (edited by Stephen Feinberg, Martin and Miron Straf for the Committee on National Statistics Commission on Behaivoral and Social Sciences and Education National Research Council in 1985) and repeated in the ICPSR Guide to Social Science Data Preparation and Archiving, and also discussed in many other publications since then. Although the benefits of sharing data are manifest and widely accepted, it is still perhaps worth their reiteration here.
Sharing data:
- Reinforces open scientific inquiry
- Facilitates high-quality, policy-relevant research
- Encourages diversity of analysis and opinions, and of a multiplicity of perspectives
- Promotes new research and allows for the testing of new or alternative methods
- Improves methods of data collection and measurements through the scrutiny of others
- Reduces costs by avoiding duplicate data collection efforts
- Allows the creation of new datasets through the merging or linkage of two or more existing sources of information.
- Provides an important resource for training in research
- Fulfils the social or custodial duty of the original data collector to the respondents
- Can reduce the burden on respondents caused by multiple data collection efforts.
Preserving and Disseminating Data
Reasons to Preserve Data
- Fulfilling contractual obligations to funding bodies
- Fulfilling custodial responsibility
- Justifying Costs of Data Collection
- Meeting Legal Requirements
- Social and Intellectual Responsibility
Benefits of Archiving Data
Archiving is a time-consuming and sometimes expensive task. It is therefore important that the benefits to the original data collector of depositing data in a Data Archive are also stressed. These are numerous and varied:
- Data are preserved and protected against technical obsolescence
- Archiving ensures the safekeeping of data
- Data are cleaned and documented to a high standard
- Confidentiality of data is ensured by the archive
- Data are described and publicised by the archive
- Obligations regarding making funded research available to the research community are fulfilled.
- Researchers are able to demonstrate continued use of the data after the original research is completed, which can influence funding agencies to provide further research money
- Costs of disseminating data are borne by the archive, freeing time and resources of data collectors and permitting other users to carry out secondary analysis on the data.
- Procedures required for archiving of data impose research management standards can be useful during primary analysis of the data.
Basic Procedures of Archiving Data
Preserving data for later re-use and sharing is not simply a matter of a decision to do so, nor is it a technical issue alone. It should also be an organisational and strategic issue, which is best approached by establishing appropriate policies and adopting best practice in this area.
It is also an important part of any strategy for the preservation and sharing of data that a code of best practice is adopted. This will imply that decisions on eventual preservation will inform all aspects of the project, including design decisions, management of metada and an acceptance of the principles of data re-use.
Archives typically provide detailed information on the archiving of data. Examples are those provided by the Inter-University Consortium for Political and Social Research in its Guide to Social Science Data Preparation and Archiving (which can be downladed from the ICPSR website at http://www.icpsr.umich.edu/access/dpm.html and that from the UK Data Archive at http://www.esds.ac.uk/aandp/create/research.asp. It is best to check the websites of the intended archive as procedures might well vary.
Depositing a data collection is straight forward if the project is effectively managed and well ordered. Just as materials need to be collated and organised by a researcher for their own fieldwork and analytical purposes, the same systematic approach to housekeeping one's research and data will form the basis of a data deposit. Good research practice relies upon an organised approach to work where each stage of the research process is documented in a clear way. Confusion over file contents, consent or a change in the direction a project took can be difficult to resolve as a project nears completion or if it is being reviewed at a much later date. Maintaining lists of data and metadata is a crucial method of keeping a research project under control.
It is important that the eventual deposit of the data is bourne in mind from the very beginning of a project. Complete documentation is both a prerequisite for deposit, and an important aspect of responsible research management. Archives generally provide guidance to investigators as to what constitutes good and complete documentation (see, for example, Good Practice In Data Documentation produced by the UK Data Archive).
In stressing the importance of a data management plan, the ICPSR Guide highlights the following aspects of creation of documentation as part of the project plan:Documentation should be as much a part of project planning as questionnaire construction or analysis plans. At a minimum, a project plan should involve decisions on the following topics:
- File structure.
- Naming conventions.
- Data integrity
- Codebook preparation. (with a reference to the “Characteristics of a “Good”
- Dataset” of the Data Documentation Initiative (DDI) for guidance in using a standards-based approach to codebook production.
- Variable construction.
- Documentation on project decisions, field procedures, coding decisions, variable construction and so on.
- Integration of documentation into project procedures.
Good Practice Guidelines and Codes
Increasingly, research organisations, centres and organisations have recognised the need for guidelines on good research practice. These cover some basic issues in research practice in common; others relate specifically and solely to the handling of data as part of the research process; others are much broader and cover all aspects of research. Below is a list of examples of these types of Guidelines and codes of practice. The categories, with examples, are:
Guidelines on the Handling of Statistical Data
- UK Government statistical services
- UK Data Archive and Royal Statistical Society
National Institutions
- University of Dublin
Funding Bodies
International Organisations
Europe-wide Guidelines
Promoting Easy, Effective and Economical Access to Essential European Data
©NESSIE 2004
