Academic journal article Population

Providing Access to Quantitative Surveys for Social Research: The Example of INED

Academic journal article Population

Providing Access to Quantitative Surveys for Social Research: The Example of INED

Article excerpt

The importance of sharing survey data

Activities to facilitate access to survey data are of key importance for the social sciences. It is thanks to such efforts that surveys can be shared between the teams involved in their design and with the broader research community. Sharing survey data is essential for the social sciences as it provides opportunities for secondary analyses, and for the testing and replication of studies. This enables researchers to understand, evaluate and build upon existing research, thereby contributing to the progress of their disciplines (King, 1995, 2006; ICPSR, 2012; Silberman, 1999). Sharing survey data also discourages scientific fraud and is useful for teaching analytical methods. It benefits survey producers by raising awareness of their work through citation, provides justification, through data re-use, of the high costs of surveys, and allows for further testing of data collection methodologies (Silberman, 1999).

Data sharing activities concern both data producers and survey data archives (ICPSR, 2012). Producers process the data for subsequent use and create coherent data files. They also prepare survey documentation, grant access to data files and archive them to assure their reusability. Survey data archives, i.e. archives that mainly deal with data at the individual (micro) level, serve the entire research community, and must provide accompanying documentation that is as exhaustive as possible. The role of data archives is to review data quality, create exhaustive metadata records, publish data and metadata files in online catalogues, manage users' requests for data access, provide assistance to users and liaise with data producers to report on survey data use (ICPSR, 2012).(1)

Data documentation is paramount to data sharing, because "without adequate documentation, scholars often have trouble replicating their own results months later" (King, 1995, p. 444). However, there is generally no specific budget for the preparation of metadata, even though produced data can be used by other research teams, avoiding the need to duplicate data collection operations. Moreover, researchers are reluctant to dedicate much time to this activity. The majority hastily assemble the documentation just before depositing the data in an archive. Often "researchers still wrestle with the idea that others may benefit from using their painstakingly gathered datasets and, perhaps even more important, they also fear that by making their datasets public, mistakes in the collection or processing of their data and their (to be) published results may be discovered" (De Moor and Van Zanden, 2008, p. 68). At the same time, in most western countries, the practice of depositing data has become a mandatory requirement of funding agencies, and a growing number of journals require authors to make available all data referenced in an article (De Moor and Van Zanden, 2008; Mochmann and Vardigan, 2011). Nonetheless, archivists who assemble the documentation and manage data access are often confronted with issues of metadata availability.

This article acknowledges the importance of data sharing and describes the archival activities carried out by the Surveys Department (SES)(2) of the French Institute for Demographic Studies (Institut national d'études démographiques, INED) to provide access to INED surveys. These surveys consist of quantitative individual (micro) level data produced for non-commercial purposes by researchers with public funds, often with the collaboration of other public bodies. It examines INED's activities in both the international and French contexts of access to quantitative social science survey data (hereafter referred to as survey data). It does not cover other types of social science research data that fall beyond its scope, such as qualitative survey data, electronic texts, linguistic corpora, historical and archeological data, administrative data, or data produced for commercial purposes. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.