Determining if you need PRA clearance
PRA exemptions
You do not need PRA clearance if:
Federal employees and contractors (acting in their official capacity) are excluded from the definition of "persons" under the PRA, so you do not need clearance to conduct research with them about matters regarding their work. (Citation)
The PRA does not apply when you are conducting research with less than 10 members of the public. This limit applies per "round" of research, over a 12 month period. For example, if you are interviewing participants, conducting usability testing and then sending out a post-launch feedback survey in the course of developing a new product, each of these would be considered different "rounds" of research because they are intended to collect different information. (Citation)
(Citation)
PRA for recruitment
Screener surveys for recruitment will need PRA clearance (if 10+ people will respond).
Never fear! You can get PRA clearance for a screener in 7-14 business days.
Simply follow the process to get Fast Track PRA clearance.
- You can also view an example of a screener that got clearance.
Please note that collecting Personally Identifiable Information (PII) has privacy implications.
Collecting and storing this information can be as simple as locking a folder with appropriate Controlled Unclassified Information (CUI) labels, but please work with GSAIT's Privacy Office (privacy.office@gsa.gov) or your supporting counsel in OGC's Tech Law Division to address these implications.
PRA for research
The following research methods do not need PRA clearance (regardless of the number of participants).
Each of these methods involve direct observation of participants (in-person, virtual, or on a recording) and/or use open-ended questioning, making them exempt from needing PRA clearance regardless of the number of participants.
Usability testing
Observing participants as they attempt to use a product or service while thinking out loud. View OIRA's usability testing guidance for more information.
Interviews
A wide-spanning set of open-ended or semi-structured questions posed to participants to discover more about a problem space.
Open-ended feedback forms
Forms that ask open-ended questions like "Please let us know your thoughts". PRA-friendly templates are available through Touchpoints.
Card sorting
A categorization exercise in which participants divide concepts into different groups based on their understanding of those concepts. Can be moderated or unmoderated.
A/B or multivariate testing
A method that asks participants to compare the performance of two or more versions of content to see which one appeals more.
Visual preference testing
A method that allows participants to review and provide feedback on a solution's visual direction.
Contextual inquiry
A method in which participants go about their work as researchers observe and ask questions.
Cognitive walkthrough
An method in which participants work through a set of representative tasks and ask questions as they go.
Close-ended UX/CX feedback surveys will need PRA clearance and are eligible for the Fast Track process.
Quantitative feedback surveys
This could include, for example, feedback surveys that are based on a Likert scale and/or any other questioning that will produce numerical data for analysis.
Close-ended, qualitative feedback surveys
This could include surveys that ask a set of identically worded questions with close-ended responses to participants.
- Example: USWDS top tasks research study
Learn how to get Fast Track clearance for your research.
Larger-scale surveys are not eligible for the Fast Track process.
Any larger-scale research that implies statistical rigor or relates to the development of surveys, program effectiveness metrics, or outcome measures and program-level decision-making
This kind of research will need PRA clearance, but will not be covered by the Fast Track process.
Please contact GSARegSec@gsa.gov and consult pra.digital.gov for more information.
Other methods
If you are planning to use another method not listed, please contact tts-research@gsa.gov.
You can also consult the OMB CX Team's page "What type of CX research does not require PRA approval?" (Requires login with PIV card.)
Example scenarios
Common research scenarios and PRA implications
| Description of research scenario | Needs PRA clearance for recruitment? | Needs PRA clearance for research? |
|---|---|---|
| A TTS UX Researcher is recruiting a few participants through word of mouth by connecting with a community organization they had previous contact with. They are planning to have unstructured conversations with 4-5 participants about whether they might find a new website feature useful or not. | NO (recruiting <10 people) | NO (<10 people, unstructured conversation) |
| A TTS Content Designer needs to recruit 20-30 people that are representative of the US population and will send out a screener survey asking demographic questions to some email listservs. They are planning to conduct moderated card sorting tests to get a better sense of participant's mental models. | YES (recruiting 10+ people w/ structured survey) | NO (direct observation, unstructured conversation) |
| A TTS Product Manager is conducting an anonymous customer satisfaction survey (meaning they won't collect any Personally Identifiable Information) and expects between 50-100 participants to fill it out. | NO (not collecting PII) | YES (10+ members of the public, no direct observation, structured) |
| A TTS UX Designer has to test the accessibility of their website and needs to recruit participants with and without disabilities. They are planning to conduct 12-15 usability tests (which will not require PRA clearance) and send a follow-up survey to participants afterwards (which will require PRA clearance). | YES (recruiting 10+ people) | YES (for the structured survey, not for usability testing) |