Zhao Shaokang conducted a survey on Beijing citizens at his own expense in September. He revealed that he did not dare to publish the reasons at the time.

The United Daily News recently released the latest poll for the Taipei mayoral election. The Democratic Progressive Party’s Shen Boyang’s support was 45%, while Chiang Wanan’s support was 41%. Senior media person Zhao Shaokang broke the news today that he actually spent his own money to conduct a poll in September. When the Taipei poll was conducted, Jiang Wanan only won 2%, which made him afraid to publish it at the time.

Zhao Shaokang said on the radio program "Chao Shaokang Time" today that he previously spent 1 to 2 million on Taipei, New Taipei, and Kaohsiung mayoral election polls. As a result, when Taipei did it, Jiang Wanan only won by 2 percentage points. "I didn't dare to publish it at the time," because he only won 2 percentage points.

However, Zhao Shaokang emphasized that the reason why he did not publish was not because he did not win much, but because of the sample problem. 20% supported blue and 30% supported green. This is different from the voter structure in Taipei City that he knows. "The worst and worst blue and green in Taipei City are about the same? How can it be possible that 10 or 20% more people support green than blue."

Zhao Shaokang said that more than 1,000 samples were taken at that time, and it was like this. When he saw that the samples were like this, he thought it was not right. He thought that there was a problem with the sample's voice rate, and therefore there was a problem with the entire poll, so he did not publish it. Hundreds of thousands were spent on this part of the poll, "but it was all in vain!"

The Lianhe Daily survey was conducted from October 20 to 22 and 25, successfully interviewing 1,115 adults registered in Taipei City, and 264 people refused to be interviewed; at a 95% confidence level, the sampling error was within plus or minus 3.0 percentage points. The survey uses Taipei City's residential and mobile phone dual-telephone records as the parent population for random sampling. It improves the shortcomings of traditional local phone sampling by adding samples of mobile phone users who are inaccessible to local calls. It is weighted according to the gender, age and administrative district structure of voters over 20 years old in Taipei City. The source of funding for the survey is Lianhe Daily.