electionline Weekly
In Focus This Week
Caltech 2024 Election Integrity Project
Election security collab between researchers and election officials to combat misinformation
By Michael Alvarez, PhD
Flintridge Foundation Professor of Political and Computational Social Science, Caltech
Co-Director, Caltech Linde Center for Science, Society, and Policy
Co-Director, Caltech/MIT Voting Technology Project
It’s hard to believe but it’s been over two decades since the Caltech/MIT Voting Technology Project was founded. During that time, we have seen the dramatic evolution and growth of academic research about election administration and technology. In the immediate aftermath of the 2000 presidential election, the founders of this project scoured the research literature and found that there was not a great deal of work we could draw upon to guide the project’s initial efforts.
But that has now changed considerably, with important new work appearing weekly in peer-reviewed journals, with academic conferences devoted to advancing the study of election science, and with scholars throughout the United States and other nations devoting considerable time to studying many aspects of the election process.
And even better, much of this research is very applied, intended to help improve election administration and technology. Academic researchers are talking about their work at workshops and conferences of election officials, helping to disseminate the results of our work to those who can use it to make important changes to how elections are conducted. We continue to see the development of direct collaborations and partnerships between academic researchers and election officials, which are generating important synergies for both those conducting elections and for those who study elections.
To further these direct collaborations and partnerships, Caltech recently received a research grant from the Haynes Foundation for our research through the Linde Center for Science, Society, and Policy at Caltech. This grant is focused on developing new and deepening existing partnerships between academic researchers and election officials, in particular with respect to providing research-based factual material to help mitigate or prevent the spread of election misinformation. The grant continues similar partnerships that Caltech has built, especially in Southern California, since 2018. We’ve named this effort The Caltech 2024 Election Integrity Project.
An important example of our work work cites a common election security rumor about the voter list maintenance. To quote an election security rumor from the Cybersecurity & Infrastructure Security Agency (CISA) Election Security Rumor vs. Reality area of its site, “Election officials don’t clean the voter rolls.” Election officials and those who closely study election administration know that this is not true, as many steps are taken by election officials before, during, and after elections to make sure that the voter rolls contain accurate, timely, and complete information about all registered voters in their jurisdiction.
One way to help stop these rumors is for local election officials to work directly with researchers to provide access to the voter registration databases, so that academic researchers can independently and transparently conduct detailed quantitative examinations of the accuracy and completeness of the data in those files.
Such partnerships were conducted by Caltech in 2018 and 2020, in association with the Registrar of Voters’ Offices in California’s Orange County and Los Angeles County. The process was straightforward. The local election officials provided daily “snapshots” of their voter registration data. The Caltech team built algorithms that scanned each snapshot for potential duplicate records, and which compared daily snapshots to past daily snapshots to look for evidence of anomalous record changes. Detailed reporting dashboards provided crucial information for the local election officials, allowing them to take action to further examine potential duplicate records and also to be aware of situations where there might be anomalous record changes (either in the form of records added or dropped, or in the form of changes to the records themselves).
Other than providing actionable information for the election officials concerned, what is valuable about such a partnership? Obviously, during the preparation period for a presidential election, few jurisdictions have the personnel, resources, or computational facilities that could be used for these types of detailed studies. But equally important, these reports provide independent confirmation that the voter file in a jurisdiction is as accurate, complete, and timely as possible.
This can help when certain inquiries arise about rumors like this. For example, if a local newspaper calls the election official asking about voter file maintenance, the election official will have the intelligence on hand to respond in detail to the questions. And, the reporters can be referred to the academic partner for their independent evaluation. If candidates running in the jurisdiction make these same inquiries, again, they can also be referred to the academic partner. This may not totally prevent the spread of such rumors, but it can help to mitigate their origin and spread.
We are currently initiating these same voter file reports for a number of Southern California counties, and we welcome inquiries from other jurisdictions which might find them useful in 2024 or 2026.
However, an exciting new dimension of our research project in 2024 might provide even further support to election officials to help them deal with election misinformation and rumors when they arise.
Over the past few years, there has been considerable research seeking to understand how to talk to those who might be misinformed, or who might be concerned about a particular rumor they have heard. More intriguingly, there has also been considerable research that has sought to determine how to “pre-bunk” or “inoculate” people so that they are better able to spot rumors and misinformation, and thus to not be as susceptible to misinformation in the first place. (Interested readers might want to pick up a copy of Lee McIntyre’s “How to Talk to a Science Denier: Conversations with Flat Earthers, Climate Deniers, and Others Who Defy Reason” and Sander van der Linden’s “Foolproof” to learn more about this research).
Our research group is currently working to understand how to use these techniques, along with fact-based, scientific research, to better respond to misinformed statements about election administration and technology, and to possibly “pre-bunk” some of this misinformation and rumor.
“Inoculation against election misinformation is especially useful because there are countless variations of false information that people might encounter. Debunking is naturally reactive, and it’s not possible to fact check each and every election rumor people encounter. Pre-bunking presents a way to be proactive,” said Mitchell Linegar, a Caltech elections researcher who is working closely with me and the rest of our team on this important effort.
We plan to provide election officials with best practices in the very near future about how to talk about election rumors and misinformation, and how to frame responses about misinformation and rumor to their stakeholders and communities. Please visit our project website – https://protectingtheelection.us – where we will share our research in the coming weeks and months. Stay tuned and please reach out to us with any questions!