This document provides some answers to questions commonly asked by applicants and links to useful resources provided by other organizations to support providing access to data. Applicants are advised to read the Policy on Providing Public Access to Data Resulting from IES Funded Grants and the Implementation Guide for Public Access to Research Data for a full description of IES requirements. Questions are organized around four inter-related areas that must all be considered in the beginning stages of planning the study.

Access

How can I provide access to datasets?

There are a number of different methods for providing access to datasets. One approach is to send your data and documentation to an existing data repository (see the end of this document for examples).¹ Uploading of the dataset to a repository facilitates the capacity of other researchers to find available data. Some repositories offer researchers the choice of providing either full access to data or access through an application process. In cases where restrictions are placed on using the dataset, researchers applying for the use of data are typically asked to sign an agreement that stipulates the terms of use. Some researchers may want to create two files for public dissemination — one for that restricts access and one that is unrestricted.

Other options for making data available on the Internet include the use of websites at universities, state education agencies, and school districts as well as university repositories. However, IES encourages the practice of uploading data to a publicly accessible repository to facilitate the use of data and the provision of access after the grant has ended.

How can I obtain a DOI for my dataset?

Most repositories provide a Digital Object Identifier (DOI)² for uploaded datasets. IES encourages grantees to obtain a DOI for the purpose of public acknowledgement of their dataset and for the data to be easily found by other researchers.

What file format should I use when providing access to data?

It is advisable to provide the data in several electronic file formats (one of which is non-proprietary) in order to provide broad use for researchers with access to different analytic software. Note that data must be provided in an electronic format.

What does IES consider to be final data?

Refer to the IES Policy or the Fiscal Year 2015 RFA for the IES definition of "final data."

When am I expected to provide access to final research data?

Sharing of datasets is expected no later than the time of peer-reviewed publication of the final results from a study. However, data may also be made available earlier as appropriate. Researchers are encouraged to share data that will inform the field more broadly than may be feasible only through published studies.

How long should the data be made available?

IES expects data to be available for at least 10 years following publication of findings in a peer-reviewed scholarly publication.

What if I am not able to provide access to data?

Researchers are expected to fully explore viable options for data sharing, and in most cases will be able to share de-identified data under full or restricted access. It is crucial to consider data sharing requirements in recruiting schools and organizations that will contribute to the study so that adequate IRB and consent processes can be established prior to any data collection. If the researcher provides a data management plan which states that data cannot be shared, the DMP should include a compelling rationale which fully explains how all possible approaches to data sharing have been explored and are not feasible to employ. IES approval of the DMP is required prior to the commitment of funds for the grant.

In cases where access to administrative data cannot be provided directly, researchers may be able to provide access to the information used to create the datasets so that other researchers can contact the original source of the data to obtain a similar dataset. Alternatively, the dataset may be given back to the source (e.g., school district or state) and researchers can contact the source to obtain access.

What if my university/organization says they own the data and do not want to release it?

Data used in federally funded grants are considered to be products of the grant. EDGAR 34 CFR.74.36(C) states that: "The Federal Government has the right to: (1) Obtain, reproduce, publish or otherwise use the data first produced under an award; and (2) Authorize others to receive, reproduce, publish, or otherwise use such data for Federal purposes." See http://www.gpo.gov/fdsys/pkg/CFR-2014-title34-vol1/pdf/CFR-2014-title34-vol1-part74.pdf for further guidance.

Documentation

What kind of documentation is recommended?

There are several types of documentation that are needed in order to support the use of data: files that support discoverability (i.e., the likelihood that researchers will find the data) and files that describe how to use the data. The capacity to find data in an Internet search is enhanced by the provision of metadata files (as well as the addition of a DOI number). Metadata is 'data about data' and contains information on who created the data, where it is located, and characteristics of the file structure.

Metadata provide information about the dataset to help people find, understand, and use your data. There are emerging metadata standards in many fields, but currently there is not a set of standards in education research. Most importantly, researchers should document everything and strive to make notation as interpretable as possible. An example of a table of common elements for a metadata file can be found at http://www.lib.ncsu.edu/guides/datamanagement/metadata. Depending on the structure of the data and the file(s), this information can be embedded directly into the file or can be included with it as a separate file (e.g., a ReadMe.txt file, a .pdf, etc.).

Resources

What resources do I need to support providing access to data from my grant?

The DMP should describe staff responsibilities in creating, maintaining, and documenting datasets and support access. Staff responsible for creation and oversight of data (as well as regularly updating) should be clearly identified and their duties described. It is particularly important that documentation is updated on a regular basis to provide continuity in the event that project staff leave the grant prior to completion of the final dataset. It is also important to identify hardware and software that will be needed to support data management and sharing.

The costs of providing access to data should be included in the grant application budget. Such costs can include those associated with preparing the dataset and documentation as well as storing the data. The rationale for each of the costs needs to be provided in the grant application Budget Justification section.

How can I make data available at all (or for 10 years) if I have no funds?

Making the data accessible is an IES requirement. If costs are associated with making these data available after the grant has formally ended, those costs can be incorporated into the grant application budget (that is, you can pre-pay for this time before the end of your grant with grant funds). However, there are other low or no-cost options for making data available on the Internet including websites at universities, state education agencies, and school districts as well as university repositories. IES encourages the practice of uploading data to a publicly accessible repository to facilitate the use of data and the provision of access after the grant has ended.

Permissions

What processes should be in place to protect the identity of study participants?

Consent forms and IRB approvals should reference future sharing of data and stipulate the conditions that will be put in place to protect the privacy of participants. Researchers may use proxy IDs as one means of protecting direct disclosure, but should also attend to the possibility of deductive disclosure due to small numbers of participants or based on certain sample characteristics. In some cases, the potential for deductive disclosure may be decreased by limiting the number of variables that are made available to an individual researcher.

What should I include in consent forms to allow for future data sharing?

Participant consent forms should include appropriate language to support public sharing of data. The Interuniversity Consortium for Political and Social Research (ICPSR) includes some suggestions for language in consent forms at the following website: https://www.icpsr.umich.edu/icpsrweb/content/datamanagement/confientiality/conf-language.html.

Other Sources of Guidance

Expectations and standards in the field regarding data sharing are currently still evolving. IES does not endorse the use of a particular repository or standards for documentation. The following websites are provided only for reference as resources that may be useful to researchers in planning for data sharing and creating a DMP. This document will be updated as new resources are identified.

Constructing a DMP:

Interuniversity Consortium for Political and Social Research (ICPSR): https://www.icpsr.umich.edu/icpsrweb/landing.jsp

Data Management Tool (UC): https://dmptool.org/

University of Arizona: http://data.library.arizona.edu/data-management-plans

North Carolina State University: http://www.lib.ncsu.edu/guides/datamanagement/how_to_dmp

Research Data Alliance: https://rd-alliance.org/

Data Cite: https://www.datacite.org/

Center for Open Science: http://centerforopenscience.org/

State Longitudinal Data Systems: http://nces.ed.gov/programs/slds/

Data repositories:

Interuniversity Consortium for Political and Social Research (ICPSR): https://www.icpsr.umich.edu/icpsrweb/landing.jsp

Figshare: http://figshare.com/

Databrary: http://databrary.org/about.html

Harvard Dataverse: http://datascience.iq.harvard.edu/dataverse

MDRC: http://www.mdrc.org/public-use-data-files

University of California — Irvine: http://kdd.ics.uci.edu/

IES Guidance

Policy Statement on Public Access to Data Resulting From IES Funded Grants

Implementation Guide for Public Access to Research Data

---