Poll on the ROK–U.S. Investment and Shipbuilding Agreement

Key Findings

Topline Results

Results by Respondent Characteristics

Figures are weighted row percentages. Completed-interview counts and weighting factors are reported separately in the respondent profile below.

Note. Respondents in their 40s were the only group in which "a good negotiation" reached a majority (53.1%). Among those aged 60 and over, "don't know" rose to 34.5% (60s) and 41.1% (70 and over).

Caveat on interpretation. Gangwon·Jeju has 37 completed interviews (weighted base 43). Figures for that region are reported for completeness and should not be interpreted on their own.

Note. Disclosure was the majority position in every age group and every region. It was highest among respondents in their 30s (80.7%) and lowest among those in their 50s (62.0%).

Caveat on interpretation. The response options did not provide a rationale for either position. Because the item did not specify the timing or scope of disclosure, the result is best understood as a general preference on the question as worded. It does not necessarily indicate opposition to the negotiations themselves.

Note. Renegotiation was the most-chosen option in every age group and every region. It was highest among respondents in their 50s (56.8%) and 60s (56.5%).

Cross-Question Analysis

All figures in this section are weighted row percentages; n is the number of completed interviews in each group.

A majority favored disclosure in both assessment groups. Among respondents who described the agreement as a good negotiation, 54.5% favored disclosure; the corresponding share among those who described it as a bad negotiation was 87.5%.

Among respondents who judged the agreement a good negotiation, renegotiation (46.1%) was chosen more often than implementing it as agreed (41.0%).

Caveats. These are cross-sectional associations measured at a single point in time and do not establish causal direction. Q3 was asked after Q1 and Q2, and order effects cannot be excluded.

On the "Don't Know" Response to Q1

The 27.9% "don't know" share on Q1 was the highest of the three items. It was substantially lower on disclosure (8.3%) and on the future course (15.3%). Among respondents who withheld judgment on Q1, 62.5% also said the terms should be disclosed and 38.2% chose renegotiation.

One possible interpretation is that some respondents were not ready to assess the agreement while still holding views on disclosure and next steps. The survey did not directly ask why respondents selected "don't know," however, so this interpretation should be regarded as tentative.

Survey Overview

Respondent Profile

Completed interviews versus the weighted base, with the resulting weighting factor.

The weighting factor is the weighted base divided by the number of completed interviews in each category. Final RIM weights adjust sex, age and region simultaneously. Combining the three dimensions, individual final weights range from 0.52 to 1.92. The design effect is 1.06 and the effective sample size is 948.

Notes on Method

Weighting. RIM weighting (raking) on sex, age and region, benchmarked to Ministry of the Interior and Safety resident registration figures for July 2026 (population aged 18 and over: 44,546,499). Weights were not trimmed. The actual range of individual weights (0.52–1.92), the design effect (1.06) and the effective sample size (948) are disclosed above.

Margin of error. The ±3.2-point figure is calculated on the effective sample size of 948, not on the 1,006 completed interviews.

Subgroup figures. All subgroup figures in this report are weighted. Cells with fewer than 30 completed interviews are excluded from interpretation.

Single-call design. Each number was dialled once. Demographic weighting does not eliminate the possibility of non-response bias associated with a single-call design, and this limitation should be considered when interpreting the results.

Separate policy matter. On 23 July 2026, the Office of the U.S. Trade Representative announced Section 301 duties of 10% or 12.5%, net of the most-favored-nation rate, for certain Korean products, subject to applicable exemptions, in connection with forced-labor import measures. This action was not referenced in the questionnaire and should be considered separately from respondents' views of the investment agreement.

Scope. This survey examines public attitudes toward a policy issue. It does not measure voting intention, candidate preference, or party support. The questionnaire and methodological information are published here to support independent assessment of the findings.

Questionnaire

Response options were read in forward and reverse order across split samples; "don't know" was always read last. Sex, age group and region of residence were asked as background items.