How do independent polls (YouGov, Civiqs, Pew) differ in question wording and timing when measuring support for abolishing ICE?
Executive summary
Three independent pollsters—YouGov (often paired with The Economist), Civiqs, and Pew—show rising but different estimates of U.S. support for “abolishing ICE,” and those differences track both how questions were worded and when the surveys were fielded; YouGov’s January 2026 Economist partnership found a near-plurality favoring abolition after a high-profile shooting (46%), Civiqs’ tracking shows a slower rise from 19% in September 2024 to the low‑40s in 2025, and Pew is cited for related attitudes toward DHS and deportation but is not represented in the provided sources with a direct, comparable question about abolishing ICE [1] [2] [3].
1. How timing shifted the headline numbers
Poll timing produced sharp swings: The Economist–YouGov snapshot conducted Jan. 9–12, 2026 captured public reaction after the Minneapolis shooting and found 46% in favor of abolishing ICE versus 43% opposed (sample n=1,602) [1] [4]. By contrast, Civiqs’ September 2024 measure reported just 19% support for abolition, a baseline frequently cited to show the scale of change [2] [5]. Civiqs’ daily tracking then documents a steady climb through 2025—rising from about 21% on Election Day 2024 to roughly 37% by July 2025, and later reporting figures in the low 40s—showing how sustained political events and media attention moved respondents over months, not just days [6] [7].
2. Question wording: “abolish,” “eliminate,” and replacement caveats matter
Wording choices altered responses. YouGov’s reporting frames answers as “support or oppose eliminating the agency” and reports both “somewhat or strongly” categories, language that can register more intensity than a single binary [8]. Some historical polls explicitly asked whether ICE should be abolished “and replaced with a different agency,” a qualifier that earlier polls used and that produced lower support numbers (for example, 2018–2019 surveys that included replacement language yielded support figures in the 29–33% range) [8] [1]. Civiqs’ tracking often frames the question as support for “getting rid of the agency” or “abolishing ICE,” and reporting sometimes specifies the universe—registered voters versus all adults—which can shift percentages [6] [7]. The presence or absence of a replacement clause, and the intensity gradations (somewhat/strongly), thus pull responses in measurable ways [8] [7].
3. Sample frames and methodology: registered voters, opt‑in panels, and weighting
Methodological differences also change the headline. YouGov draws from an opt‑in online panel and weights to demographic and political benchmarks—YouGov’s brief notes weighting by gender, age, race, education, 2024 presidential vote and other factors—while The Economist–YouGov result cited a web‑based survey of 1,602 adults [8] [4]. Civiqs offers daily tracking of registered voters, a narrower electorate-focused frame that historically yields different percentages than adult samples and can show faster partisan movement around election cycles [6] [7]. These sample-frame choices—adults versus registered voters, online opt‑in versus tracking panels—explain part of the numeric divergence even when questions appear similar [8] [6].
4. Event sensitivity: why a single incident changed the snapshot
Multiple outlets trace the January 2026 uptick directly to public outrage after an ICE agent shot a Minneapolis resident; YouGov’s contemporaneous poll found rising disapproval of ICE tactics and linked the spike to the shooting, while Civiqs’ rolling data shows a more gradual trend that predated and was accelerated by the incident [9] [5] [6]. In short, short‑window polls taken immediately after a salience‑raising event (YouGov Jan. 9–12) show higher support for abolition than broader, longitudinal tracking that smooths momentary spikes [4] [6].
5. Pew’s role and the limits of available reporting
Pew Research Center is cited in the sources for ranking DHS low in agency popularity and for asking related deportation questions, but the provided materials do not include a direct Pew questionnaire or a Pew estimate expressly measuring “abolish ICE,” so no definitive comparison of Pew wording or timing on that exact question can be made from these sources [3]. That gap is important: absent a cited Pew question text and field dates, claims that “Pew finds X on abolishing ICE” cannot be substantiated here [3].
6. Bottom line for interpreting polls
Readers should treat the headline percentages as products of timing, framing, and sampling: immediate post‑incident snapshots (Economist–YouGov Jan. 9–12, 2026) can show sizable surges, Civiqs’ daily tracking reveals slower trend lines and different baselines (registered voters vs. adults), and prior question wording—especially whether abolition would be followed by a replacement agency—has historically depressed support numbers compared with blunt “abolish” phrasing [4] [6] [8]. Any direct comparison requires attention to the exact question text, the dates fielded, and whether the sample is adults or registered voters; the sources provide that detail for YouGov and Civiqs but not for Pew within this dataset [4] [6] [3].