BP blogged an introductory post last week looking at the March 2, 2013 Bangkok gubernatorial race. This post will look at 4 NIDA polls published over the last month or so, namely:

  • NIDA Poll (Dec 20-21) that surveyed 1,254 people in all 50 Districts;
  • NIDA Poll(Dec 25-26) that surveyed 557 people in 7 Districts;
  • NIDA Poll (Dec 25-Jan 4) that surveyed 3,356 people in all 50 Districts;
  • NIDA Poll(Jan 17-19) that surveyed 1,500 people in all 50 Districts;

Below is a chart of the 4 polls:

Microsoft Excel

BP: As you will see the number of undecided voters is increasing even if the choice of candidates has grown. NIDA polls are well-known for including large number of undecided voters – see this post from 2011. Hence, BP has created another chart removing the “not vote” and “undecided” categories and adjusting the other candidates votes accordingly:

Microsoft Excel

NOTE: Of course, this is assuming those in the ”undecided” category will vote for candidates in the same proportion as those who have already made up their mind. However, in support of the general accuracy of this assumption is the reality that voter turn-out is unlikely to exceed 60%. Hence, many of the undecideds are likely to be those who won’t vote. In addition, you can take the above as reflecting voter intentions as of the time period of the poll in a more readable way. How the undecideds will vote will also depend on future events so it is hard to know for certain so weighting in another way would be pointless.

1. Based on the latest poll, things are quite close with only a 3% margin separating the two main candidates.

2. The independents, including Seripisut, have no chance but are still likely to attract up to 20% of  votes cast.

Will look at other polls and then do some analysis on the state of the race.

*Note: Eagle-eyed readers may have added up the December 25-January 4 NIDA polls figures up and only got 97% and wondered where the missing 3% went. No idea either. BP inputted all the figures and only get 97%. So for the 2nd chart have assumed, the missing 3% are in the ”not vote” and “undecided” categories.