The material housed in this password-protected website is confidential and only intended to be viewed and utilized by members of the Analyst Group (the “AG”). Keeping this material within the AG will ensure that partners can safely share research results and that our community can continue to grow our impact having benefited from our collective learning.

If you think others would benefit from the material we host,
invite them to join the AG. 

Confidentiality policies

These are our official confidentiality policies. Please review them carefully and see the following “Guidance for community members” section to understand what they mean in practice for your work.

Sharing information outside the Analyst Group

• No member is permitted to share any materials made available to AG members via this website, or otherwise made available to AG members at training programs or events or by Analyst Institute staff, with people outside the Analyst Group.
• No member is permitted to republish, redistribute, or otherwise share any such materials or their derivatives outside the AG.
• Any member who wishes to share such materials must obtain explicit permission from the owner(s) of said material. If ownership is unclear, please consult with Analyst Institute.

Sharing information within the Analyst Group

• With explicit permission from Analyst Institute, members may perform independent analyses and otherwise create derivatives of our materials to share within the AG.
• Members may only share AG content with other AG members by sending links to Analyst Institute’s secure, password-protected website. Do not bypass the secure login by sharing AG content directly in non-secure venues such as campaign or organizational materials.

Machine learning and generative AI

To ensure that materials remain exclusively within the AG:

• Do not data mine, scrape, or otherwise gather any material or data from this website for use in other platforms or systems.
• Do not use any such material or data in connection with the development of any software program, including, but not limited to, the training of a machine learning or artificial intelligence system. You may not otherwise input or load material into any such systems or tools.

Guidance for community members

In keeping with our confidentiality policies, below are examples of what is and is not allowed. Please reach out if you have any questions or want help referencing our research while adhering to these guidelines!

You may not:

• Post/publish/forward quotes, point estimates, charts/graphs, or the names of organizations that implemented research outside of the AG.
• Summarize, repost, or refer to specific results or syntheses of multiple research results (e.g., meta-analyses, literature reviews) outside of the AG.
• Disseminate independent analyses (quantitative or qualitative) outside of the AG that have been derived from data, results, and other website materials.
• Load material into an artificial intelligence system such as ChatGPT for any reason, such as to help you construct a campaign script or to create a summary of the content.

You may make general references to what has been studied and share secure links to pages on our website as long as the results aren’t referenced and the references are anonymized. You can mention the subject matter without revealing results or recommendations as demonstrated in the following examples:

• OK to share: “This test [linked to results on password-protected Analyst Group website] compares the effectiveness of plan-making to social pressure messages”
• Not OK to share: “Multiple tests of relational voter contact over the past decade have found positive effects in increasing turnout”
• Not OK to share: “A test of Partner X’s 2022 GOTV program found a XX pp increase in turnout following use of hard social pressure language”
• Not OK to share: “AI’s meta-analysis of GOTV programs, including X relational tests conducted over the past decade, found an average effect of XX in midterm election contexts.”

Sharing with conservatives and Republicans

Analyst Group members occasionally work on projects (often bipartisan) that involve conservative or Republican consultants or clients. In these instances, it is especially important that members do not expose community learnings to the other side.

You may not share knowledge (e.g., test results, best practices, recommendations) or materials without explicit permission from the contributor, and you must be transparent and contact AI staff about situations in which community learnings may be at risk of exposure through your work.