As much as it pains me to do this while I still haven’t even started digesting last week’s election, I have been forced by UDN and China Times to start a page showing presidential race polling results. Can we wait a while before we start that race? Please? I only want a month or two. I’m just not psychologically ready for the next election to start yet.
In a related story, both polls asked questions about the shooting, and both found that about 4% of respondents claimed that they changed their behavior in reaction to the shooting. However, this doesn’t mean you can just subtract 4% from KMT candidates’ vote shares to get the “real” vote (ie: what would have happened if the shooting had never occurred). Some of the people who were mobilized to vote or changed their votes ended up voting for the DPP. The net effect may have benefited the KMT, but these results suggest the effect was very small. (Surveys can’t tell you whether the effect was 1% or 2% because those are statistically not different from each other. However, we can probably be sure that the effect wasn’t over 5%. At any rate, I’m going to stick with the qualitative conclusion of “small.”
Frozen Garlic 
December 1, 2010 at 1:52 am |
“Some of the people who were mobilized to vote or changed their votes ended up voting for the DPP”
Since you always insisted on solid evidence, the above sentence, written as a statement but not an opinion like “I believe …,” I would guess that you do have solid evidence for this, too ?
December 1, 2010 at 1:09 pm |
The China Times survey asked people how they changed their behavior. There were some who weren’t going to vote but then decided to vote for the DPP. There were also some who changed their vote from KMT to DPP. Again, these are very small numbers and so we can’t really measure them very accurately with surveys. However, there certainly were people who said that they behaved in these manners.
December 3, 2010 at 3:50 am |
I believe the news below from the China Times is the one you quoted:
http://news.chinatimes.com/focus/0,5243,50107372×112010112900064,00.html
in which it says,
連勝文槍擊案被視為影響五都選情的關鍵事件，由事後調查回溯五七九位五都市長選民的投票意向，自認受此槍擊案影響的票源約有四％。調查反映，有二％的選票流向國民黨候選人，但也有一％轉向民進黨，另有一％則是其他狀況。
December 3, 2010 at 4:38 pm |
There was also a table in the printed version. I haven’t looked at the online version, but they often don’t include tables.
December 5, 2010 at 1:30 pm |
The Thursday before the election, Hau was leading 52-48. A poll done after the shooting was reported at 8:50 P.M. on the news showed that Hau increased his lead to 54-46. An increase of 4 points.
A poll was not done for Xinbei City after 8:50 P.M. on Friday but the poll the day before on Thursday showed 53-47 and the result was about 52.6% in favor of Chu so it appears the shooting did not impact Xinbei City but impacted the Taipei City race because of Lien’s close relationship with Hau Long-Bing. Also Lien Chan showed up at the Taipei City rally and made a plea so it’s plausible that it would help Hau’s support rate.
It’s reasonable to believe that the shooting of Sean Lien would not impact races evenly all over Taiwan. He has nothing to do with candidates running in Taichung and Southern Taiwan.
So it’s open to interpretation whether the DPP would have won or not in Taichung without the shooting. I don’t think the shooting would have impacted Taichung voters much because it doesn’t have anything to do with them. I think Hu just barely hung on, but if the election campaign had continued another week, perhaps the DPP candidate Su Jia-Chuan may have pulled ahead. He clearly won Taichung County and may have gotten even more support in Taichung City if he had more time.