SEON survey denominators

SEON’s November 2 release reports 65% would blame the agent provider for fraud. Its base: 885 opinion‑givers, excluding unsure respondents—not the full 2,026 U.S. adult sample.¹

Online fieldwork ran October 2–6 among adults 18+, with nine‑variable weighting. Guardrails used a random half (n=1,016). The stated ±3‑percentage‑point credibility interval covers only full‑sample questions, not a measured subgroup margin.¹

These opinions establish neither fraud incidence nor legal liability. Weighted percentages do not supply raw counts.

Card and laptop, August 2021; archive, not an AI‑agent payment or survey participant. Photo: Nenad Stojković / Shixart1985 / Wikimedia Commons, CC BY 2.0. Downscaled and display‑cropped.