[General Election Poll Verification - Busanjin-gap] Only Busan Ilbo and Busan MBC Out of 11 Polls in April Correctly Predicted Seobyungsoo's Victory

[Asia Economy Reporter Hwang Jin-young] Various opinion polls conducted to gauge local public sentiment ahead of the general election have been mired in controversy. The core of the debate was that opinion polls failed to accurately capture public sentiment and instead distorted public opinion. Asia Economy is publishing a series verifying how accurate the opinion polls were, focusing on constituencies where many polls were conducted.

Busanjin-gap, where 'presidential hopeful' Kim Young-chun of the Democratic Party of Korea, a former Minister of Oceans and Fisheries, faced off against Seo Byung-soo of the United Future Party, a former mayor of Busan, was the fiercest battleground in Busan and Gyeongnam (PK) region.

To counter Kim, who was running for a fourth term, the United Future Party nominated Seo, who had served three consecutive terms in Haeundae-Gijang-gap, as a 'challenger candidate,' making the district a nationally watched area.

Opinion polls showed Kim leading Seo either within or slightly beyond the margin of error.

The conservative vote was split to Seo's disadvantage as Jeong Geun, a doctor who received 24.7% of the vote in the 19th general election, ran as an independent candidate.

However, when the ballot boxes were opened, Seo received 52,037 votes (48.5%) and Kim received 48,287 votes (45.0%), with Seo winning by a margin of 3,750 votes (3.5 percentage points).

Among 11 opinion polls conducted since the official campaign began on the 2nd, only two predicted Seo's victory, while the other nine showed Kim ahead.

A poll commissioned by Busan MBC to Hangil Research from April 4-5 surveyed 708 eligible voters aged 18 and over residing in Busanjin-gu Gap constituency, showing Seo Byung-soo at 44.7% and Kim Young-chun at 40.5%. (For detailed information, refer to the Central Election Poll Deliberation Commission website)

This poll predicted Seo's lead, with a 4.2 percentage point difference between the two candidates, close to the actual vote difference of 3.5 percentage points.

Comparing the difference between support rates and actual vote shares for each candidate, Seo was off by -3.8 percentage points and Kim by -4.5 percentage points, slightly exceeding the margin of error (±3.7 percentage points).

A poll conducted by KSOI for Busan Ilbo on the 6th predicted Seo's lead with Seo at 44.2% and Kim at 39.2%.

The difference between poll support rates and actual vote shares was -4.3 percentage points for Seo and -5.8 percentage points for Kim.

Seo's vote share was within the margin of error (±4.3%), while Kim's vote share exceeded the margin of error.

Media	Polling Agency	Polling Period	Kim Young-chun	Seo Byung-soo
SBS	Ipsos	5-8	40.5	38.8
Newsis	Realmeter	7-8	46.1	36.3
Munhwa Ilbo	Embrain Public	7	42.1	35.5
Busan Ilbo	KSOI	6	39.2	44.2
Seoul Economic Daily	Embrain Public	6	38.5	34.7
Kookmin Ilbo·CBS	Realmeter	4-5	44.5	36.2
Busan MBC	Hangil Research	4-5	40.5	44.7
JTBC	Korea Gallup	4-5	39.1	36.8
KBS	Korea Research	2-4	43.7	35.4
TV Chosun	Matrix Research	2	39.5	36.9
Dong-A Ilbo	Research & Research	31-1	38.9	35.9

Unit: %

However, the other nine polls all predicted Kim's victory.

Among the nine polls predicting Kim's win, the poll commissioned by Newsis and conducted by Realmeter on the 7th-8th showed a support rate gap between the two candidates exceeding the margin of error (±4.4%).

In this poll, Kim had 46.1% and Seo 36.3%, with a 9.8 percentage point gap.

This poll was close to Kim's actual vote share with a 1.1 percentage point difference but differed from Seo's actual vote share by 12.2 percentage points.

Polls commissioned by Kookmin Ilbo·CBS and conducted by Realmeter on the 4th-5th (Kim Young-chun 44.5%, Seo Byung-soo 36.2%) and by KBS and conducted by Korea Research on the 2nd-4th (Kim Young-chun 43.7%, Seo Byung-soo 35.4%) also showed support rate gaps of 8.3 percentage points each.

Both polls were within about 1 percentage point of Kim's actual vote share but differed from Seo's actual vote share by 12.3 and 13.1 percentage points, respectively.

The difference of more than 10 percentage points between the poll support rates and Seo's actual vote share resulted in outcomes far from the actual voting results.

The lower accuracy of opinion polls in Busanjin-gap is also interpreted as partly due to Jeong Geun's poll support rate not translating into actual votes.

Jeong's support rate in polls hovered above 10%, but his actual vote share was only 5.5%.

It is analyzed that some of Jeong's supporters switched their votes to Seo, who shares a conservative stance, due to strategic voting to avoid wasted votes, influencing the election outcome.