Social Science Data that have historical/scientific significance should be preserved even when immediate dissemination is not possible. This policy can be applied to either an entire study or to selected variables in a data set. Data whose preservation, cataloguing or storage has been supported in any way by membership funds must become available at an agreed upon date in the future. The following document was designed to address the issues surrounding the possible institution of an expanded Collection Development Policy to accommodate preservation with delayed dissemination.
Human Subjects Protections - When there is potential for disclosure of human subjects, ICPSR may wish to preserve data without dissemination until disclosure risk is minimized. For example, smaller longitudinal studies may pose disclosure risk even with the removal of direct and indirect identifiers. In this instance, ICPSR will work with the data collector to determine when disclosure risk will be minimized (e.g. when all human subjects are deceased) and plan to disseminate data at that agreed upon time. ICPSR may make a preservation copy when there are human subject protection issues for a number of reasons (1) to develop expertise with the data, (2) to process the data while there is knowledgeable staff to work with, and (3) for general safekeeping. ICPSR will be more likely to accept data with human subjects protection issues when a viable subset of the data can be released earlier through one of ICPSR dissemination streams (e.g. restricted release and/or secure data enclave). Before release, data will be subjected to review by an IRB (or an IRB-type organization.)
Self-Dissemination - Data collectors may choose to disseminate their own data especially while funding is available to support this activity. Also, in some cases data collectors may be required to maintain a user registration system. Since ICPSR collects registration information, but does not share user information with the original data collector, this may also be a barrier to immediate dissemination by ICPSR. ICPSR may make a preservation copy during the period of self-dissemination for a number of reasons (1) to develop expertise with the data, (2) to process the data while there is knowledgeable staff to work with, and (3) for general safekeeping. ICPSR will enter into an agreement to make an archival copy for later dissemination only when there is an agreed upon date for eventual dissemination. For example, ICPSR may request to begin dissemination when grant funding ceases to support dissemination.
Other - Occasionally, depositors will request an embargo to be placed on their data until a book or article is published or a report is written and submitted to a funder. Embargoes will typically last no longer than one year from the time ICPSR receives the study. During the embargo, ICPSR will process the data in the usual manner, but not release the data to the public. When the embargo is lifted, data would be ready for immediate dissemination.
Not all data have equal merit regarding preservation without dissemination. It is in ICPSR's interest to make an archival copy when the following criteria are met:
Data have high social science significance as outlined in the ICPSR Collection Development Policy.
The data have an undue risk of being lost. Examples include data that are not being disseminated widely and data not having an adequate backup/recovery plan.
A compelling reason exists why the data cannot be disseminated by ICPSR. (see above discussion on Human Subjects Protection, Self-Dissemination, and Other)
The PI and ICPSR can work out an agreeable plan for the eventual dissemination of data.
Issues regarding embargo periods and human subject protections specific to preservation-only studies will be settled as part of the original negotiations with the data collector prior to deposit. In addition, the negotiation process will include ground rules on the extent of processing which will take place by ICPSR staff while the study remains in the "preservation-only category". Possible items to address may include discussion about whether data will be subject to "routine" or "intensive" processing. Highest priority will be given to activities relating to the original reasons for inclusion of the study for "preservation-only". These original reasons include the need to develop expertise with the data and to process the data while there is knowledgeable staff to work with.
Like all data in ICPSR, preservation-only studies will be preserved, cataloged (by PI, title, summary, and ICPSR-assigned collection/study number) and will be searchable by the tools residing on the ICPSR homepage. Unlike other studies, these preservation-only data will not be available for ordering in the normal manner. A new field will be added to the study record - Anticipated Release Date. For data that are being self-disseminated - a link to the website where data can be downloaded/order will appear (as is already customary for self-dissemination datasets)
Data will be disseminated at the predetermined end of the preservation-only period. Dissemination dates may vary according to the original reason for non-dissemination. In all cases, the agreed upon date will be negotiated by the data collector and ICPSR prior to submission.
Human Subjects Protections - When there are unusual human subjects protection issues, ICPSR will work with the data collector to determine when disclosure risk will be minimized (e.g. when all human subjects are deceased) and plan to disseminate data at that agreed upon time.
Self-Dissemination - When data are being preserved by ICPSR and disseminated by the data collector, ICPSR will begin to disseminate data when one of the following criteria is met: (1) when grant funding ends and dissemination activities are no longer supported or (2) ICPSR confirms that data are no longer being disseminated by the data collector.
Other - Data embargoed because of impending publications and/or funder requirements will be released in a timely manner following the agreed upon release date.
Membership and external grants cover the cost of processing data that are disseminated to the membership and public-at-large. Studies accepted into the Archive under the "preservation-only" category will need to be carefully tracked in order to study the need for cost recovery and the viability of continuing to accept studies of this category. It is anticipated that the archiving and preservation of data with delayed dissemination (preservation-only data) will be more costly than that of unrestricted public-use data. Contracts must be drawn up and reviewed outlining agreements with data collectors thus increasing the initial negotiation period and requiring additional staff time. In addition, unusual costs are expected to accumulate when preservation-only periods are long in duration or updates to preservation-only data must be made before the data are ever disseminated.
Careers | Contact Us | Privacy Policy | Official Representative Site | Accessibility
© 2007 Regents of the University of Michigan. ICPSR is part of the Institute for Social Research at the University of Michigan.
