A. Ballot Count Accuracy
Among our greatest concerns are the discrepancies reported.The table below shows all districts with ballot count discrepancies. In 10 of these districts the scanner counted more ballots than reported as counted by hand; in 6 of these districts, the scanner counted fewer ballots than were counted by hand.
|
Nov 2011 |
|||
| Scanner Counted Ballots | Hand Counted Ballots | Difference | Percent Difference |
|
208 |
205 |
3 |
1.4 % |
|
349 |
346 |
3 |
0.9 % |
|
1278 |
1269 |
9 |
0.7 % |
|
361 |
359 |
2 |
0.6 % |
|
269 |
270 |
-1 |
-0.4 % |
|
310 |
309 |
1 |
0.3 % |
|
711 |
713 |
-2 |
-0.3 % |
|
859 |
861 |
-2 |
-0.2 % |
|
987 |
989 |
-2 |
-0.2 % |
|
1103 |
1101 |
2 |
0.2 % |
|
2154 |
2152 |
2 |
0.1 % |
|
1092 |
1093 |
-1 |
-0.1 % |
|
1190 |
1191 |
-1 |
-0.1 % |
|
1209 |
1208 |
1 |
0.1 % |
|
1385 |
1384 |
1 |
0.1 % |
|
1459 |
1458 |
1 |
0.1 % |
Table 1: Discrepancies in Numbers of Ballots Counted by
Hand vs. Counted by Scanner[1] in Eighteen Districts, November 2011 Audits
We have omitted reports from the table for one town apparently reported total number of votes rather than ballots.
Based on observer reports, we do not believe that all of the hand counts are accurate because of the questionable counting methods observed. On the other hand, because of these differences, we also have no basis to conclude that the scanners counted all ballots accurately.
In general, when compared to the November 2010 and November 2009 reports except for two extreme cases in November 2010, there was only moderate, if any, improvement in the performance of the optical scanners and the election officials in counting ballots in November 2011.
B. Vote Count Accuracy
Even considering confusion over ballots with questionable votes, an analysis of the district reports submitted to the Secretary of the State indicates that vote count discrepancies remain.
|
Nov 2011 |
|||||
|
Col C Machine Totals (tape) |
Col D Undisputed Vote Totals |
Col E Questionable Vote Totals |
Col F Overall Hand Count Totals (D+E) |
Difference |
Percent Difference |
|
531 |
485 |
6 |
491 |
40 |
7.5 % |
|
62 |
76 |
1 |
77 |
-14 |
-22.6 % |
|
75 |
62 |
0 |
62 |
13 |
17.3 % |
|
404 |
386 |
5 |
391 |
13 |
3.2 % |
|
77 |
64 |
0 |
64 |
13 |
16.9 % |
|
50 |
63 |
1 |
64 |
-13 |
-26.0 % |
|
1686 |
1672 |
2 |
1674 |
12 |
0.7 % |
|
688 |
679 |
0 |
679 |
9 |
1.3 % |
|
971 |
959 |
3 |
962 |
9 |
0.9 % |
|
83 |
74 |
0 |
74 |
9 |
10.8 % |
|
963 |
951 |
3 |
954 |
9 |
0.9 % |
|
1559 |
1547 |
3 |
1550 |
9 |
0.6 % |
|
53 |
62 |
1 |
63 |
-9 |
-17.0 % |
|
916 |
903 |
5 |
908 |
8 |
0.9 % |
|
700 |
660 |
32 |
692 |
8 |
1.1 % |
|
189 |
182 |
0 |
182 |
7 |
3.7 % |
|
402 |
392 |
3 |
395 |
7 |
1.7 % |
|
400 |
391 |
2 |
393 |
7 |
1.8 % |
|
1084 |
1074 |
3 |
1077 |
7 |
0.6 % |
|
503 |
496 |
0 |
496 |
7 |
1.4 % |
|
910 |
897 |
6 |
903 |
7 |
0.8 % |
Table 2: Candidate counts where Hand-Counted Votes and Machine-Counted Votes Show Discrepancies of 7 Or More Votes 2011 Audit. [2]
The table on the previous page presents, by number and percentage, vote differences greater than 7 between hand-counted votes and machine-counted votes when all ballots with questionable votes are included[3] and all votes for cross-endorsed candidates are totaled.
Based on observer reports, we do not believe that all of the hand counts are accurate because of the questionable counting methods observed. On the other hand, because of these differences, we also have no basis to conclude that the scanners counted all votes accurately.
In general, the November 2011 data shows a performance in counting votes by the optical scanners, election officials, or both, improved over in 2010, but the 2011 was worse than 2009 .
The following table shows the number of candidate counts with various levels of count differences between the optical scanners and the hand counts, considering ballots with questionable votes:[4]
|
Nov 2011 |
||
|
Count Difference |
Number of Candidate Counts |
% Of All Counts |
|
0 |
498 |
56.1% |
|
1-3 |
306 |
34. 5% |
|
4-6 |
62 |
7.0% |
|
7-9 |
14 |
1.6% |
|
>=10 |
7 |
0.8% |
|
Total |
100.00% |
|
| Average Difference: |
1.12 votes |
|
Table 3: Distribution by Difference of Candidate Counts between
Hand-Counted Votes and Machine-Counted Votes 2011 Audit.
Looking at the data this way and comparing with 2010 and 2009, we observe that 2011 shows significant improvement even considering that there were fewer votes counted overall.
Using the same data as the previous table, omitting small counts with small differences[5], the following table also shows the number of candidate counts with various levels of percentages of differences between the optical scanners and the hand counts, considering ballots with questionable votes:
|
Nov 2011 |
||
|
Range of % of Count Difference |
Number of Candidate Counts |
% Of All Counts In Range |
|
0 |
498 |
56.1% |
|
> 0 and < 0.5 % |
167 |
18.8% |
|
0.5 % and < 1.0 % |
128 |
14.4% |
|
1.0 % and < 2.0 % |
57 |
6.4% |
|
2.0 % and < 5.0 % |
23 |
2.5% |
|
5.0 % and < 10.0 % |
4 |
0.5% |
|
10.0 % and greater |
10 |
1.3% |
|
Total |
100.0% |
|
|
Average Difference % |
0.28% |
|
Table 4: Distribution by Difference of Significant Candidate Counts (30 or more votes) between Hand-Counted Votes and Machine-Counted Votes
By Ranges Of Percent 2011 Audit.
We note that if we were to trust these counts as an accurate representation of the optical scanner’s counting:
- For over 25% of candidate vote counts, the machine count difference is greater than or equal to 0.5% which is the maximum level for requiring a close vote recanvass.
We do not believe that all of these counts are accurate. But we have no reason to believe that all the hand count differences can be attributed to human counting error. For public confidence it would seem important that all unsatisfactorily or unexplained discrepancies between machine counts and official, final audit results should be significantly lower than the maximum threshold for automatic recanvasses.
We continue to support investigations and recounting in public of all unsatisfactorily explained differences over a prior agreed upon threshold per count.
These counts do show an improvement in optical scanner and human counting from past years.
C. “Questionable” Votes and “Undisputed” Ballots
Observations and comments from election officials indicate confusion about classifying “undisputed ballots” and about counting “questionable votes.”[6] An undisputed ballot is a ballot with no apparent problem or questionable votes on it. A questionable vote is a mark on a ballot that may not have been read properly by the optical scanner. Audits exhibited a variety of interpretations of what constitutes “undisputed” and “ballots with questionable votes.” Audit statistics confirm these observations.
The following table has some examples of candidate counts with the largest percentages of questionable votes. Comparing Column C and Column F, in general, we observe the optical scanners seem to have counted accurately many of the votes classified by officials as questionable.[7]
|
Nov 2011 |
||||
|
Col C Machine Totals (tape) |
Col D Undisputed Vote Totals |
Col E Questionable Vote Totals |
Col F Overall Hand Count Totals (D+E) |
Percent Questionable |
|
407 |
385 |
22 |
407 |
5.4% |
|
615 |
580 |
32 |
612 |
5.2% |
|
359 |
344 |
15 |
359 |
4.2% |
|
443 |
424 |
18 |
442 |
4.1% |
|
958 |
930 |
31 |
961 |
3.2% |
|
661 |
650 |
15 |
665 |
2.3% |
|
713 |
702 |
15 |
717 |
2.1% |
|
754 |
742 |
14 |
756 |
1.9% |
|
953 |
937 |
17 |
954 |
1.8% |
Table 5. Examples of Candidate Counts with the Largest
Percentage of Questionable Votes 2011 Audit
This is consistent with the 2010 audits where results improved over 2009, yet on average, 2011 audits reported 0.77% questionable votes vs. 0.46% 2010 and 2% in 2009. Observations show that officials often classify too many votes as questionable and the results of counting show that most votes classified as questionable are indeed counted by the scanners.
[1] The law and audit procedures often use the term “Tabulator” to refer to election machines. We use the terms “Scanner” or “Optical Scanner” to make the report clearer.
[2] In 2011, there were about 50% fewer ballots counted in as counted in 2010 and about 15% fewer in 2011 than in 2009.
[3] This is the most favorable interpretation of the audit reports, giving every benefit of the doubt to the accuracy of machine counts and the accuracy of hand counts. When Total Hand Count Totals is less than or equal to the Machine Totals, then the Questionable Vote Totals are included. When Undisputed Totals is greater than or equal to the Machine Totals then all Questionable Vote Totals are excluded. In the remaining cases enough Questionable Vote Totals are included to make the difference zero.
[4] This table and the following two tables provide data similar to that provided by the University of Connecticut in analyzing post-election audit data. We use the same groupings to facilitate comparisons between our report and theirs.
[5] The table omits candidate counts with tape counts less than 30 votes that have differences less than 3 votes.
[6] Part of the confusion comes from as the terms “Undisputed Ballots” and “Questionable Votes.” One term refers to ballots, the other to votes, where the process must focus at different times between classifying ballots and classifying votes. Also the terms can add to the confusion between votes which might have been read two different ways by the scanner and votes that should have been read one way by the scanner, yet that reading would not accurately reflect the voter’s intent.
[7] [7] We display at most one count per district audited. Many other counts in the same districts had similar percentages of overvotes.
