This page discusses expectations for approved researchers in the Secure Transfer, Restricted-Use Data Lake (STRUDL).
Each section has an accompanying video and text is provided as a visual aid. Documents referenced in each video are always formatted in bold and descriptions of each document can be found on STRUDL: Forms. We encourage potential applicants and program participants to reference the STRUDL Handbook for more detail on the full program process.
In Video 2.1, we will discuss the process for disclosure review, how approved researchers can develop a workplan based on that information, and general behavioral expectations for users of U.S. Department of Labor (DOL) restricted-use data. In particular, we will focus on expected behavior regarding releasing statistics and research results.
Video 2.1: Secure Transfer, Restricted-Use Data Lake Expectations: Disclosure Review and Developing a Workplan
Overview
In this section, potential applicants and program participants will learn about the following:
- Disclosure review process and timeline.
- How to develop a workplan based on these timelines.
- What the general behavioral expectations are for users of DOL restricted-use data.
What is disclosure review?
Disclosure review is a process where elements derived from confidential data (statistics, research output, etc.) are assessed for disclosure risk—that is, the risk that a malicious actor can use the product to derive confidential information.
If the risk is deemed too high, the derived product should not be released. Disclosure control methods can be applied to partially mitigate disclosure risk.
In the context of STRUDL, “disclosure review” refers to the formal process that all data and results must undergo before leaving DOL systems or being shared with anyone who is not an approved member of the project. This process assesses the disclosure risk for products derived from DOL data, as described above.
What falls under the disclosure review process?
Under the terms of STRUDL, all outputs derived from restricted-use data that are intended for dissemination (e.g., presentations, reports, publication, sharing with external collaborators, etc.) must undergo disclosure review.
STRUDL approved researchers should never share data or results outside of DOL systems unless they have undergone disclosure review, even if they are not being widely shared. This includes sharing unapproved results with anyone not on the approved project team, including with researchers on other STRUDL projects and colleagues who are on the project but are not approved to access STRUDL data.
In addition to results derived from the confidential data, STRUDL-approved researchers may not share the following information with anyone outside the approved project team, as it is considered confidential:
- Information about the presence or absence of a record in the data
- Any information about the disclosure control methods that were applied prior to clearing their results from DOL systems
Any information that has not been formally approved for release—including statistics and descriptive information—cannot be shared with researchers outside the approved project team. If approved researchers would like to share this information or other project notes with their unapproved colleagues, they must write up their notes and submit them through the STRUDL disclosure review process. Programs, program logs, and scripts (if applicable) must also be cleared through this process but may be shared after approval.
What does the disclosure review process look like at a high level?
The disclosure review process in STRUDL is discussed in detail in the STRUDL Handbook.
Three phases of the STRUDL program lifecycle involve disclosure review. These phases, depicted in Figure 1, are as follows:
- Applicants must identify the proposed products and results that will require review in their application to STRUDL;
- After applicants are approved and conduct their research, their project results must be reviewed. As part of this phase, STRUDL approved researchers will submit a disclosure review package with required forms and all proposed products at least 90 days before the project close date. The STRUDL team will review the proposed outputs in the disclosure review package and approve, make recommendations for adjustments or deny release of the outputs;
- Once outputs are approved, STRUDL approved researchers may create and submit dissemination materials for project results. The STRUDL team will again review these results and make a final determination on approval for dissemination.
What must approved researchers provide for the disclosure review process once results are ready for review?
The process for compiling a disclosure review package, including required forms, is discussed in detail in the STRUDL Handbook. In addition to these forms, which offer a useful way to centralize information about your project, outputs, and potential disclosure risks, approved researchers will need to provide the outputs in question, such as summary tables and research report drafts.
If elements of this review package are not completed, including all necessary forms, approved researchers will be automatically rejected. This package can be submitted to STRUDL@dol.gov.
All outputs must explicitly include a disclaimer that the research does not reflect the views of DOL; exact required language can be found in the STRUDL Handbook.
How do approved researchers request a modification to the project proposal?
If applicable, researchers may request modifications to their original proposal (e.g., if they need more time, are requesting access to different data, and/or are updating the research plan with new proposed outputs) via the Project Modification form (a description of which can be found on STRUDL: Forms).
If this modification is not approved, researchers must adhere to the original specifications of their proposal, including project timelines.
How should approved researchers plan for disclosure review in the overall project timeline?
The package should be submitted at least 90 days prior to the project close date to allow for iteration and adjustment. The amount of time required for review will depend on the type and number of outputs covered by the project.
Take note of the following to minimize the review time:
- Include clear documentation and formatted files.
- Only submit outputs for review that will appear in disseminations (reviewing intermediate outputs could slow down the review process).
How can approved researchers implement best practices for disclosure review throughout the project?
STRUDL approved researchers are encouraged to maintain an open dialogue with the DOL disclosure review team throughout their research project to ensure proper implementation of best practices. Videos 5.1 and 5.2 discuss these practices in more detail.
What are the general behavioral expectations of STRUDL approved researchers?
To reiterate, a major requirement of STRUDL approved researchers is that no results or details about restricted-use data may be shared outside of DOL systems without first going through the disclosure review process. This includes both written products of any type and verbal conversations about the results or data. Researchers are also prohibited from sharing or discussing data or results with other STRUDL researchers working on different projects.
Researchers may only use their data access in the context of the research plan submitted when they applied to STRUDL. Any other use of DOL data is a violation of the NDA signed when applying.
Approved researchers are also expected to:
- Report data breaches and other security breaches immediately according to DOL policy (more details: Video 3.1 and accompanying text);
- Follow physical security and computer practices (more details: Video 3.2 and accompanying text); and
- Ensure that any updates to submitted documents or research plans (made via the Project Modification form as described above) are re-approved.
At all stages of the project lifecycle, approved researchers are encouraged to reach out to the STRUDL team (STRUDL@dol.gov) with any questions or concerns.