2025 Immigration Message Test

July 14, 2025
Download Full Deck

Methodology: We tested 7 immigration messages with Swayable using randomized controlled trial (RCT) survey experiments to see which was best at shifting Trump approval. These messages were tested in English and Spanish. The tests ran from July 3 to July 9 nationally. We collected a total of 16,580 responses, with 11,852 English responses and 4,728 Spanish responses.

Key Takeaways:

1. Across languages and subgroups, the most effective message for lowering Trump approval overall and on immigration is focused on the impact of the administration’s actions on families and children. We recommend continuing to amplify stories of overreach as proof points. 

1. The Trump administration is going too far. Instead of prioritizing violent criminals, they’re separating mothers from their children and tearing families apart, many of whom have been in the US for a long time. They even detained a high school student with no criminal record while he was on his way to volleyball practice. Separating families and targeting children doesn’t make our communities safer. 

2. El gobierno de Trump está pasándose de la raya. En lugar de priorizar a los delincuentes violentos, separan a las madres de sus hijos y separan familias, muchas de las cuales llevan mucho tiempo en Estados Unidos. Incluso detuvieron a un estudiante de secundaria sin antecedentes penales mientras se dirigía a su entrenamiento de voleibol. Separar a las familias y atacar a los niños no hace que nuestras comunidades sean más seguras.

2. In English, we also found that focusing on the unfair impact that mass deportations are having on law-abiding, hard-working immigrants is an effective approach, though less effective in Spanish

1. Instead of prioritizing violent criminals, they are raiding workplaces, profiling people who are out shopping or driving, detaining immigrants who are doing the right thing by going to their scheduled court hearings, and even revoking permits and protections from legal immigrants to make them targets for deportation.

2. En lugar de priorizar a los delincuentes violentos, están allanando lugares de trabajo, discriminando a personas que salen de compras o conducen, deteniendo a inmigrantes que hacen lo correcto al asistir a sus audiencias judiciales programadas e incluso revocando permisos y protecciones a inmigrantes legales para ponerlos en peligro de deportación.

3. While not as effective overall as the others, a broader “Broken Promises” message was relatively effective across groups and in both languages at lowering Trump approval. It offers a model for how to incorporate immigration alongside other key issues

1. Trump is breaking his promises to the American people. He told us he’d bring down the cost of living, but his tariff policies continue to increase the cost of everyday goods. He told us he’d deport violent criminals, but instead is separating parents from their children and targeting hardworking immigrants who have been here for years. He told us he’d be a peacetime president, but instead dropped bombs on Iran. He promised he wouldn’t cut Medicaid, but his budget bill would do exactly that, taking away critical healthcare coverage from over 8 million Americans while giving tax breaks to billionaires. 

2. Trump está incumpliendo sus promesas a los estadounidenses. Nos dijo que reduciría el costo de la vida, pero sus políticas arancelarias siguen aumentando el precio de los bienes cotidianos. Nos dijo que deportaría a criminales violentos, pero en cambio está separando a padres e hijos y atacando a inmigrantes trabajadores que llevan años aquí. Nos dijo que sería un presidente en tiempos de paz, pero en cambio bombardeó Irán. Prometió que no recortaría Medicaid, pero su proyecto de ley de presupuesto haría precisamente eso: quitaría cobertura médica esencial a más de 8 millones de estadounidenses y daría exenciones fiscales a multimillonarios.

Overall Findings:

  • In English, the most effective immigration messages for lowering Trump approval overall and on immigration were the “Threat to all of us” message and the “Impact on families and kids” message. That is true for Latinos and non-Latinos. These messages were also most effective among the least politically engaged
  • In Spanish, every message except the “Threat to all of us” message significantly lowered Trump overall job approval, but none of them shifted Trump immigration approval. This suggests Spanish-dominant Latinos may be especially persuadable on their rating of Trump relative to other Latinos and voters generally. 
  • The broader “Broken Promises” message was also effective at lowering Trump approval overall and on immigration for all voters.  It also lowered Trump approval on immigration for Latinos in English. 
  • Our continued attempts to find an immigration message that connects to the economy and cost of living yielded mixed results. The messages were largely effective in Spanish, but not in English. 

For full memo, please download PDF below.

Read More

We work toward a more sophisticated understanding of the experiences, issue preferences, and political identities of Latino and Hispanic voters.

Blog title heading will go here

June 14th, 2023
Full name
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros.

Latino State of Play: 2025 Elections and New Equis Polling

November 17, 2025
Survey conducted 10/15 – 10/29 via phones and text-to-web with 2000 registered voters who identify as Hispanic or Latino nationally. The sample included an oversample in the following pooled competitive congressional districts (all 10%+ Latino CVAP): CA-03, CA-09, CA-13, CA-22, CA-27, CA-40, CA-41, CA-45, CA-47, AZ- 01, AZ-06, NV-03, NM-02, CO-08, TX-28, TX-34, TX-35, TX-15, NJ-07, NJ-09, NY-03, NY-04, NY-17, PA-07, PA-08. The national sample was weighted to match the composition of registered voters modeled as Latino by state, gender, age, education, and 2024 vote recall. The sample was also weighted by age, education, 2024 vote recall by gender, and geography, to match the composition of registered voters modeled as Latino in each state with a sample size of 175+ (CA, FL, TX), in the Mid-Atlantic region (DE, MD, PA, NJ, and NY) and the Southwest region (AZ, CO, NM, and NV). Language of interview, place of origin, education and other variables were monitored during the fielding. The survey was conducted in English and Spanish, according to the respondent’s preference. MoE is ~2.0 % for the full sample, ~6.0% for FL, the Mid-Atlantic, and Southwest, ~5.0% for TX, and ~4.0% for CA.

Memo: Latino Voters in Arizona September 2025

October 21, 2025
This memo summarizes key findings from a statewide poll of 600 registered Latino voters in Arizona, conducted September 18-23, 2025. The poll has a margin of error of ± 4.0 pp. This research was conducted by Equis Research in partnership with Progress Arizona and the Copper State Research Consortium. Arizona has approximately 1.3 million eligible Latino voters — 25% of the state’s potential electorate — positioning Latino voters to once again play a decisive role in Arizona elections. This poll explores how Arizona's Latino voters are evaluating the current political landscape, including the state's gubernatorial race, economic priorities, immigration enforcement, and views on Trump's presidency.

Memo: Summer 2025 Latino PulseCheck

August 19, 2025
This memo summarizes key findings from a national poll of 2,000 registered Hispanic voters, with an oversample in competitive congressional districts that are at least 10% Latino by eligible population. This poll was conducted July 21-31, 2025 and has a margin of error of ± 2 pp

July 2025 Poll on Latinos, Trump and the Economy

July 31, 2025
This memo summarizes key findings from a national poll of 1,614 registered Hispanic voters, conducted with Data for Progress from July 7 to 17, 2025. This poll has a margin of error of ± 2 pp.

Stay in the loop with Equis

Equis is a set of organizations working to create a better understanding of Latinos, innovate new approaches to reach and engage them, and invest in the leadership and infrastructure for long-term change and increased engagement.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.