Below is an excerpt from the full report, which is viewable from this link: http://bit.ly/YR6XoY
On page 7:
On page 7:
Some Questions for the Comelec and Smartmatic
What could explain these error rates? What possible underlying root causes might explain these? How should the candidates and political parties interpret these discrepancies? How are the various RMA results consolidated? If a variance of 16.67 percent, similar to the one in this RMA, is noted in the 2013 polls, say in the senatorial position, what would the Comelec and Smartmatic do?
How are checks made on the canvassing process? How will the public or candidates and parties know if the canvass results are accurate? How should the audit of the canvassing results look like?
Discrepancy Between AES and RMA
Section 13 of the G.I. BEI states that “In the event of discrepancy between the AES and RMA results still exceeds the allowable margin of ten (10) votes per candidate per position (emphasis supplied), the RMA committee shall...”.
What does “allowable margin” mean? What is the basis for the “10 votes per candidate per position”?
Comment by MM on 11 February 2012
ReplyDeleteI am in favor of Namfrel involvement, even if only as an observer, in the RMA, so that the general public will finally have, through Namfrel, access to the actual raw data results of the RMA, and thus be able to evaluate it.
In the first place, the sample size (i.e. 231 precincts) of the RMA is NOT MINUTE. The so-called "sampling proportion" of 0.26% is totally irrelevant. First of all, a standard should be set for judging if the PCOS count of any one precinct is a success or not. The error margin for successful counting by PCOS depends on the proportion of successes in the test of 231 cases.
If the success rate in the test of 231 cases is 95%, then the error margin is about plus/minus 2.8%, i.e., the success-rate for the entire population can be estimated as between 92.2% and 97.8%.
If the success-rate in the test of 231 cases is 97%, then the error margin is about plus/minus 2.2 percent, i.e., the success-rate for the entire population can be estimated as between 94.8% and 99.2%.
My reading of the table of Telibert on p.8 is that the 1st column corresponds to the votes for each candidate, with one line per candidate, and the 2nd column (called "variance" but which I prefer to call "deviation") is equal to the RMA count minus the PCOS count. The table on p.8 apparently only includes cases of candidates with non-zero deviation, since the photo on p.7 shows many cases of zero deviation that are not in the table on p.8. The cases of deviations are almost all by 1 vote only, plus one case of a deviation by 2 votes. Whether this would be rated as a "success" or a "failure" of the PCOS count in the test-precinct depends on the pre-set tolerance for the deviations. Personally, I would like to compare the election result per candidate according to the PCOS with the result according to the RMA -- are the candidate-ranks the same? What is the vote-margin of victory per PCOS, compared to the vote-margin per RMA? For the example on p.8, I think the difference in the margin of victory in the test-precinct would not be more than 3 votes, which I doubt would deserve a rating of failure.