Summary

Power for Democracies (P4Dem) is trying to develop donor recommendations to most effectively counter 2026 US election integrity threats. To do so, we searched for (and believe we found) all the main categories of threats, identified a fairly comprehensive list of counter-tactics, and then evaluated those tactics based on evidence, a theory of change analysis, and expert interviews. 

Per our analyses, the following tactics are most promising to address the high-severity threats: 

• Volunteers, lawyers & independent poll monitors deployed to contested districts
• Proactive communication by election officials or other trusted local officials about technical security measures
• Prebunking & trusted local media fact-checking campaigns — inoculate voters before they encounter disinformation and build coalitions for real-time correction
• Deploying trusted community intermediaries as "safety ambassadors" to reduce support for political violence

What we did 

We set out to identify which threats to the 2026 midterm elections are most severe and which counter-tactics are best supported by evidence — and then to narrow those down to a short list of interventions where philanthropy can add the most value. 

The work involved two parallel tracks: a threat severity assessment and a tactic prioritisation process. Both drew on desk research and structured expert interviews and were designed to produce actionable recommendations under time and resource constraints rather than an exhaustive review. 

How we assessed threat severity 

We assessed eight election threats against a common rubric, rating each as High, Medium, or Low. The core question for each threat was: What is the expected impact of this threat on the integrity of competitive 2026 elections — specifically on voter participation, election certification integrity, and a peaceful transfer of power? 

The assessment included four components 

1. US precedent: Has this threat materialised in the US in the last 30 years? Where it has, we documented the evidence of its effect on voter participation — using quantitative turnout data where available and structured estimates where not.
2. International precedent: Where US evidence was absent or thin, we reviewed international cases, focusing on the effect on voter participation, not just whether the threat occurred.
3. US transferability: A qualitative judgment of whether the effect would be larger, similar or smaller in the US context, or qualitatively different— considering the state of democracy in the US, federalism, the judiciary, civil society, and the media landscape.
4. Current trajectory: Whether the threat is escalating, stable, contained, or declining as of early 2025 through March 2026, based on reporting, legal trackers, and expert commentary. 

A threat was considered "meaningful" if it could plausibly: (a) shift turnout or vote share by ≥1 percentage point in competitive 2026 districts or states, or (b) materially compromise election certification integrity or the peaceful transfer of power in competitive jurisdictions.  

A High rating required all three conditions to be met: the threat has demonstrably suppressed voter participation, compromised certification integrity, or disrupted the peaceful transfer of power in the US or comparable democracies at a meaningful scale; its current trajectory is escalating or uncontained; and there is no strong reason to believe US institutional safeguards would neutralise the effect. A medium rating required only one of those conditions to be met to avoid prematurely dismissing threats. 

How we prioritised tactics 

Tactic prioritisation ran in three phases. First, we applied two quick filters requiring no original research: we dropped tactics that already seemed dominated by well-funded organisations (e.g. litigation, large-scale voter mobilisation) and those that could not realistically be implemented before the 2026 midterms. Second, we applied a deliberately low evidence floor — at least one credible evaluation or well-documented real-world example with clear outcomes; tactics failed only where evidence was purely theoretical or anecdotal. Third, we assessed each surviving tactic’s theory of change: who it reaches, the mechanism by which it counters the threat, the conditions required, and whether its realistic scale is meaningful relative to margins in competitive races. 

Expert interviews 

Independently, we conducted structured interviews with five subject-matter experts, asking each to identify the most significant threats to the 2026 midterms and which two or three they would most want civil society to focus on. Rankings were aggregated with consideration for independence of perspective and each expert's field of expertise (e.g. cyber security experts, election interference experts). Expert views on individual tactics were recorded separately and considered alongside the filtering results, with expert enthusiasm serving as a tiebreaker where multiple tactics survived all filters. (Where an expert did not endorse a particular tactic, this reflects only that they did not independently raise it — not that they considered it ineffective.) 

Narrowing down to the most consequential threat-tactic pairs 

The threat severity ratings and tactic filtering results were combined to identify high-severity threat and counter-tactic pairs — interventions that address a serious threat, are not already well-resourced, can be implemented before 2026, have at least some evidential support, and have a plausible theory of change. 

For the rest of this post, we will focus on the threats categorised as high severity and the evidence-based counter-tactics that correspond to them.  

ICE and security forces interfering on and around Election Day 

The threat of ICE or military personnel being deployed to or near polling places could significantly reduce turnout — not just through direct enforcement action, but also through the fear and intimidation their presence creates, particularly in immigrant communities. All five of our experts rated this as a serious and escalating threat. 

Tactic: Deployment of volunteers, lawyers, and independent poll monitors to highly contested districts 

Evidence base:  

• A case study of volunteer poll watching groups -- during the 2026 Illinois primaries, a coalition of civil rights organisations doubled their volunteer numbers and deployed nonpartisan poll watchers, legal volunteers, and rapid response teams to counter the chilling effect of federal immigration enforcement on voter turnout. In the past their volunteers have been able to provide support for election judges (election officials are often inexperienced), ensure polling places are accessible to individuals with language barriers, and commonly address voter questions about election procedures.
• Multiple survey experiments show that information about both partisan and non-partisan election monitors can increase public confidence in elections or the lack of them can decrease trust in results
• In a large-scale field experiment, a broad-based citizen electoral observation program in Colombia led to increased reporting of election irregularities. Furthermore, when campaigns were aware of this program, it appears that the campaigns were deterred from manipulating votes (causing irregularities). 

Experts: 5/5 endorsed  

• Experts emphasised the importance of legal support networks for under-resourced county officials, with several pointing to organisations such as Protect Democracy, the Election Safeguards Response Network, and Democracy Docket as vehicles for scaling this work. 

The evidence available does not directly cover voter intimidation or even physical violence that could ensue from security force presence at polling stations. However, the evidence suggests that legal support, poll observers, and other volunteers are likely to effectively document incidents; deter illegal activity; increase citizen confidence in election integrity and results; and respond in real time in the face of intimidation. 

Spreading lies about election integrity 

Experts also rated misinformation as a severe threat, particularly due to the risk of foreign interference, rapid online spread, and narratives delegitimising election integrity. When combined with misinformation about ICE presence at polling places, this could significantly suppress turnout, especially among immigrants, naturalised citizens, and other vulnerable or targeted groups. 

Tactic: Prebunking and trusted local media fact-checking campaigns 

Evidence base: 

• Prebunking literature has a strong academic evidence base, with dozens of pre-registered survey-based experiments demonstrating that inoculation against disinformation can be effective at discerning untrustworthy content and improving online sharing decisions (e.g. Biddlestone et al 2025 and 2026, Roozenbeek et al 2022, Lewandowsky and Linden 2021). 

Experts: 3/5 endorsed. 

Tactic: Proactive communication by election officials or other trusted local officials about technical security measures (a specific version of prebunking) 

Evidence base: 

• A pre-registered survey experiment with 10,000 Americans shows that proactive communication increases trust and mitigates distrust caused by delays in results reporting. 

Experts: 3/5 endorsed.  

• Experts noted that election officials see proactive communication as part of their role but often lack the resources for effective public outreach — a gap illustrated by the confusion that followed informational changes during the Texas primaries. They also emphasised training election officials as a key enabler of this tactic. 

Prebunking programs have been found to be successful in many survey-based experiments, which might give the impression that they are the most promising of the tactics we reviewed. However, changing behaviour at a large scale in the physical world, like voting, is different and more difficult than changing attitudes in a survey or even changing online information sharing behaviour. We do think prebunking is likely a valuable tactic to deploy, but we are not confident that it will be as robustly effective as the extensive research suggests. 

Multiple high severity threats 

Tactic: Deploying trusted community intermediaries as "safety ambassadors" (addressing disinformation, ICE intimidation, and voter restriction threats)  

Evidence base:  

• A high-quality nationwide RCT from Nigeria that demonstrated reductions in electoral violence and increased turnout through community-based campaigns focused on reducing support for and preparing responses to political violence — with effects spreading through social networks even to individuals not directly reached by the campaign. Nevertheless, US-specific evidence is lacking. 

Experts: 2/5 endorsed 

Limitations and uncertainties 

AI Disclaimer: The evidence review was conducted in less than one month with substantial assistance from Claude (Anthropic's AI assistant). We have manually verified the sources it identified and checked its characterisations of findings, but there remains a risk of inaccuracies we did not catch — and Claude's retrieval is not equivalent to a systematic literature review. There are likely relevant studies, working papers, and practitioner reports that were not surfaced. We plan to run a more thorough and consistent parallel hand-search through Google Scholar in the next phase. 

We also deliberately set a low bar for passing the evidence filter — at least one credible case study, survey, or documented real-world outcome was sufficient. This means several tactics that passed have a thin empirical base, and a "Pass" should not be read as strong confidence in effectiveness. 

The boundary between high and medium threats is not finalised — election official intimidation, certification coercion, and voter data exploitation may be reclassified as the analysis develops. And several litigation-based or policy change-related tactics were excluded not because they lack evidence but because they are already well-resourced (our quick assessment was that a few million additional dollars in 2026 were unlikely to make a counterfactual difference) or could not feasibly be completed ahead of the 2026 midterm elections. Both areas warrant reconsideration, particularly in preparation for 2028. 

Next steps and final thoughts 

The goal of this work is not to be exhaustive or perfectly accurate but to quickly narrow down what decision-makers should pay attention to in as evidence-based a way as possible, though we recognize the evidence base is generally quite limited in the non-voter engagement democracy space. Even for the most promising tactics, this remains the case. It underscores a clear need for more field research and rigorous evaluations on these types of programs. 

We'll be following up with our analysis of medium-severity threats and will publish our full methodology and threat-tactic prioritisation table on our website in the coming weeks. If you're a donor, researcher, or practitioner working in this space, we'd welcome the chance to walk you through our full analysis and hear your thoughts — reach out at samantha.sekar@powerfordemocracies.org. 

Power for Democracies authorship note: Peter Lang, Research Fellow, was the primary researcher on this project and author of this piece; Jon Helfers, Sr. Consultant for Impact Evaluation and Serra Sutekin, Student Assistant, also made significant research contributions.