TVBS poll on Kaohsiung 9

So far, I have only seen one poll on a specific legislative race.  No suprises: it is Kaohsiung 9, the district in which Chen Chih-chung 陳致中, the former president’s son, is running.

Kaohsiung 9 is a strong green district.  Four years ago, Kuo Wen-cheng 郭玟成 of the DPP won this district by 6% (or about 8000 votes).  Of course, 2008 was a terrible year for the DPP.  Last year, Chen Chu 陳菊 won about 58% in this district (or about 30000 votes over the combined vote of the other two candidates).[1]  This year, Kuo is running for re-election as the DPP nominee against the same KMT candidate, Lin Kuo-cheng 林國正.  In a two way race, we might expect Kuo to win re-election rather easily.  However, with Chen in the race, it is a different story.  Chen’s vote should presumably come entirely from Kuo’s support, so there is a real chance that Chen’s entrance in the race could allow Lin to win.

Of course, most people are not really interested in Kuo and Lin or even about the younger Chen.  We care about this race primarily because of what it tells us about how much support the former president still has in Taiwan.  Will Chen Shui-bian (CSB) continue to be a major presence lurking over the political scene, or have his supporters already moved on?

Fortunately, we have some very interesting poll data.  TVBS released a poll, conducted between Sept 1-5, with sample size n=804.  (Once again, Frozen Garlic is here to provide you with an instant reaction two months later!)  TVBS asked how voters would vote in a three way race, in a two way race if Chen dropped out, and in a two way race if Kuo dropped out.  Here are the results:


  Lin Kuo Chen DK
3 way 33 34 14 19
2 way, Chen out 38 51   12
2 way, Kuo out 56   29 16


In a two way race between the KMT and DPP, the DPP wins easily, as expected.  However, when Chen enters, things change.  Chen doesn’t have anywhere near enough support to win, but he does draw off just enough support from Kuo to leave the top two candidates roughly tied.  Most stunning, if Kuo were to yield, Lin would win in a landslide over Chen.  Of the voters who support Kuo in the three way race, far more would choose to support the KMT candidate over the ex-president’s son.  Presumably, these people are overwhelmingly green supporters and will probably vote overwhelmingly for Tsai Ing-wen over Ma Ying-jeou.  This might be the clearest rejection of CSB that I have seen thus far.  He is being rejected by green supporters in the deep south.  He clearly has a hard core of supporters, but the mainstream has not only found new champions, they clearly do not want him back at all.

I don’t know what the Kuo campaign has planned, but it seems to me that they ought to make sure that voters see a lot of poll results like this.  I don’t know if the media will publish any more polls on this race.  If they don’t there are not many more cost effective paths to victory than blowing a bit of change commissioning your own polls.  Certainly, he should be able to convince the DPP polling center to do a poll.  Many voters might be wary of poll results coming from the DPP, but Kuo is trying to communicate with voters who support the DPP and are more likely to believe that the DPP polls are honest and fair.

Finally, a little note on reading poll results.  We generally think of voters and candidates lining up on one dimension, so let’s put the independence people on the left and the unification people on the right.  Chen should be at the far left end of the spectrum, Kuo in the center left, and Lin in the center right.  If Chen drops out of the race all of his supporters should go to Kuo.  (Equivalently, all of Chen’s support is stolen from Kuo.)  So it is commonly believed that you can simply add Chen’s and Kuo’s support to get the green support.  Unfortunately, reality is a bit messier.  People don’t really all line up on one dimension, and some people have strange, seemingly illogical explanations for their vote.  So when Chen (and his 14%) drops out, somehow Lin’s support goes up by 5%.  In fact, this isn’t necessarily because Chen’s supporters turned to Lin, because Kuo goes up by 17%.  What happens is really that far fewer people are undecided.  Apparently, it is much easier to make a decision in a two way race.  However, that also doesn’t make much sense.  We can imagine people who are undecided between Chen and Kuo having a much easier time deciding how to vote when only one is in the race, but these results seem to imply that there are also a chunk of people undecided between Chen and Lin, who are on opposite sides of the spectrum.  In fact, the TVBS report also indicates that there are some people who support Chen in the first situation but Lin in the second.  The point of all this is that these results are messy.  Moreover, this is NORMAL.  You can never simply add two candidates’ support together and assume that is what one would get if the other dropped out.  1+1 never quite equals two.  Keep all this in mind the next time you read a poll and think about what all those Soong supporters will eventually do.  We like clean and simple assumptions, but real life is messy.

[1] This is a quick and dirty number.  Kaohsiung 9 includes all of Xiaogang District and most of Qianzhen.  However, about a sixth of the population in Qianzhen is in District 8, so it takes a bit of work to figure out exactly how the mayoral vote breaks out into the legislative districts.

Tags: , , ,

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: