The Democratic Party of Korea said on the 23rd regarding a recently released poll indicating that the gap between Lee Jae-myung, the Democratic Party candidate, and Kim Moon-soo of the People Power Party has narrowed, that "allegations of manipulation have been raised." It added, "The election campaign committee will take strong actions to ensure that public opinion manipulation cannot take root, including filing a complaint regarding the allegations of manipulation of the poll in question."

On the 23rd, just ten days before the 21st presidential election, election candidate posters are hung on the sidewalk near Gongdeok Intersection in Mapo District, Seoul. According to a survey conducted by Korean Gallup from the 20th to the 22nd of 1,002 eligible voters aged 18 and over nationwide (with a 95% confidence level and a margin of error of ±3.1 percentage points), Democratic Party of Korea candidate Lee Jae-myung records 45%, People Power Party candidate Kim Moon-soo receives 36%, and Reform Party candidate Lee Jun-seok achieves 10% in approval ratings. Among voters, 8% hold undecided opinions. /Courtesy of Yonhap News Agency

Spokesperson Shin Hyun-jeong noted in an afternoon written briefing that "allegations of poll manipulation have emerged through media reports. We must decisively judge the forces that hinder fair elections through various rule violations and illegalities and trample on public sentiment," stating this.

The poll pointed out by Spokesperson Shin was conducted by Every Day, commissioned by Every News and Media Local (Korea Regional Journalists Club), targeting 1,000 men and women aged 18 and older, and announced on the 20th. The poll reported that the approval rating for Lee was 46% and for Kim was 41.6% (with a sampling error of ±3.1 percentage points at a 95% confidence level). The gap between the two candidates is 4.4 percentage points.

Spokesperson Shin said, "The poll that announced the gap between the Democratic Party candidate Lee Jae-myung and the People Power Party candidate Kim Moon-soo as being within a 4 percentage point margin of error was conducted using the RDD (Random Digit Dialing) method with rim weighting, raising allegations of manipulation."

This study was conducted with the RDD method applying rim weighting instead of using virtual numbers, highlighting this point. The RDD method conducts polls using phone numbers randomly selected. Therefore, it has been pointed out that it may be less accurate compared to the virtual number method, which reflects regions and ages in advance.

It is also pointed out that the method of assigning weights to polls is problematic. Unlike the cell weighting method, which assigns weights based on gender, age, and regional variables, the rim weighting method assigns weights only to basic variables from the overall sample.

Spokesperson Shin said, "The bigger issue is the high likelihood that the results were leaked prior to the public announcement of the poll, noting that the comment by lawyer Seo Jeong-wook the day before, stating that 'a poll with a 4 percentage point margin of error will be released tomorrow,’ has been confirmed." Seo is a conservative commentator.

He added, "Especially if the Kim Moon-soo camp was aware of the results before the announcement, it is highly likely they intervened in the poll manipulation."

He emphasized, "Poll manipulation is a serious crime that undermines the fundamental order of democracy. It cannot be tolerated under any circumstances," stating, "The campaign committee will respond strongly to the allegations of manipulation of this poll, including filing complaints, ensuring that public opinion manipulation cannot take root. We strongly urge investigative agencies to conduct thorough and swift investigations."

Meanwhile, the Democratic Party claimed that the narrowing of support rates between candidate Lee and candidate Kim in recent polls indicates that "a conservative oversampling phenomenon is occurring."

Cheon Jun-ho, head of the campaign strategy office, said in a briefing at the party's office in Yeouido, Seoul, in the afternoon, "The support ratings for the candidates are expected to converge with party support ratings," adding, "We must pay attention to the fact that a conservative oversampling phenomenon is occurring in this process."

He added, "We should keep in mind that polls may not reflect actual public sentiment."

The poll cited in this article was conducted using a 100% ARS telephone survey method utilizing RDD. The sampling error is ±3.1 percentage points at a 95% confidence level, and the response rate is 5.5%. For detailed information, refer to the Central Election Poll Review Committee.