Sample Counts - What exactly happened and what is it?

[Sammyboy RSS Feed]

An honorable member of the Coffee Shop Has Just Posted the Following:

http://www.eld.gov.sg/mediarelease/E...Count_1Sep.pdf

From the media, we hear that Sample counts had always been done in previous elections but not announced to the public. Only in this election 2015, that it is announced to the public.

I've the following questions:

1. Sample counts of 100 ballots are done at the polling stations in front of the candidates and counting agents, counted and a weightage given. The candidates are informed. The question here is there are many polling stations in a GRC and from what I know, in previous elections, the counting is done after the ballot boxes are driven to Counting Centers. In this pdf, there is no mention of ballot boxes driven from polling stations to Counting Centers. So, how are the candidates informed of the sample count result? Is this done in previous elections too?

2. Are the counting of the ballots done in the polling stations or at the Counting Centers? If done at the polling stations, then how are the candidates going to know the final result? With the 100 sample counts and if done at the polling stations, does it mean that the boxes are transported to the Counting Centers after the 100 sample counts are done?

3. Is the pdf info incorrect? That the 100 sample counts are done after the boxes are transported to the Counting Centers?

4. While the 100 sample counts are counted and tallied, are the counting for the rest of the ballots started? Or that has to wait until the result of the 100 sample counts are tallied and verified by the candidates and counting agents?

It is read that all counting of the 100 sample counts from all the electoral divisions are done by 10pm and announced by then to the public through the media and election website.

The 100 sample counts has an accuracy rate of less than 1% for almost all the contested wards. This typically defies statistical logic.

We need to know what sophisticated formula is used by the Election Dept to get such an acute accuracy. This has to be studied by independent mathematicians from renowned universities and statistics companies and research institutes to enlighten us how this is even mathematically possible. The Elections Dept can patent this formula as this can be used for other mission critical systems and gaming, etc.

On a cautionary note, it is of the opinion that the 100 sample counts for this election is very vague on its execution and process and we cannot accept any election result if we are not able to understand how transparent the processes and procedures are.

All parties really need to come clean and explain in details how this sample count is conducted and why it has a margin of error of 4% +-. This is a statistical breakthrough, as its formula application in other commercial and military systems is highly valuable.

It is not too far-fetched for anyone to be inclined to cast obvious doubt and skepticism of the election result until such a detailed disclosure of the 100 sample counts formula and its rationale are made by the Election Dept. That perhaps the 100 sample counts could be potentially a very needed quality-assurance mechanism built into the system, while something else that is time-sensitive may be happening with the technology deployed for the rest of the ballot papers.

The result of this GE15 defies perception, logic and common sense and is so contrary to existing ground sentiments.


Click here to view the whole thread at www.sammyboy.com.