MyDD April Reader Straw Poll
| Candidate |
1st Choice |
March |
2nd Choice |
1st + 2nd Choices |
Last Choice |
| Edwards |
43.2 |
33.1 |
29.7 |
72.9 |
1.5 |
| Obama |
34.3 |
36.1 |
29.1 |
63.4 |
2.7 |
| Richardson |
7.9 |
9.7 |
19.4 |
27.3 |
0.4 |
| Clinton |
3.8 |
4.7 |
7.1 |
10.8 |
18.4 |
| Kucinich |
1.4 |
2.0 |
2.6 |
4.0 |
26.2 |
| Dodd |
0.7 |
1.1 |
2.5 |
3.2 |
1.7 |
| Biden |
0.5 |
0.5 |
1.3 |
1.8 |
18.3 |
| Gravel |
0.4 |
0.3 |
0.7 |
1.0 |
30.9 |
| Other |
5.9 |
10.6 |
5.5 |
11.4 |
-- |
| Unsure |
2.0 |
1.9 |
2.1 |
4.1 |
-- |
It was a good month for Edwards, who rose ten points, while everyone else dropped slightly. The main source of Edwards' rise came from "Other," and he still dominates the second place votes among the remaining "Other" voters, where he leads Obama and Richardson 55-13-13. Other factors: was this due to my methodology, the way we cover the candidates on MyDD, or the general upward trend for Edwards? Who knows, but I'm sure we will have those discussions in the comments.
There were some interesting tidbits about the last place votes:
- Obama supporters were 12 of the 13 people who ranked Edwards last, but Edwards supporters only represented 10 of the 27 who ranked Obama last. Then again, all three people who ranked Richardson last were Edwards supporters.
- 93 Edwards votes ranked Biden last, while only 22 Obama voters ranked Biden last. This is stunning, because there was virtually no difference between Edwards and Obama supporters when it came to last place votes for Clinton (62-59), Gravel (107-100) and Kucinich (90-83).
- Dodd continues to have the fewest supporters and haters. Combined, I only counted twenty-five first and last place votes for Dodd. To put this in perspective, Richardson, with 134, had the second fewest first and second place votes combined.
Anyway, I do not have a huge amount of commentary on poll that was fairly static. The methodological statement is in the extended entry.
You can view the results of the poll yourself
by clicking here, and using zxa1q5ge as the password. I removed stuffed votes
according to the same rules I followed last month. Overall, the poll was open from 6:50 pm eastern, until 12:35 a.m. eastern, which was a change from March when the poll was held in the afternoon. I counted 1652 non-stuffed votes.
Just like in March, I removed all votes that only ranked one or two candidates. Unlike March, I had to deal with a new stuffing problem: repeating sequences. Here are just some of the repeating sequences that were clearly stuffed votes:
7,1,6,3,0,8,2,9,4,5: 15 times, all near the end
1,6,3,7,9,4,0,2,5,8: 7 times, all near the end
6,1,5,0,9,7,2,8,4,3: 5 times, all near the end
6,3,8,7,2,1,0,9,4,5: 21 times, all near the end
1,6,3,7,8,2,0,4,5,9: 14 times, all near the end
2,1,7,6,3,0,8,8,4,5: 10 times, all near the end
3,6,1,8,2,5,0,9,7,4: 14 times, all near the end
6,1,3,2,7,8,9,0,4,5: 21 times, all near the end
6,8,1,2,7,9,3,0,4,5: 8 times, all near the end
6,8,2,3,1,0,7,9,4,5: 7 times, all near the end
8,6,3,2,7,0,1,5,4,9: 11 times, all near the end
There were so many of these repeating sequences, all of them appearing toward the end of the data stream, that I fond a point just before the appearance of the first repeating sequence, and deleted all votes coming after that point. Once that was done, I looked at the data in an Access database, and was satisfied that the remaining votes were legitimate.
Cry foul as much as you want. If you know me, you know that I did my best to produce as accurate results as possible. This is not a scientific poll, but it is the best I can do.
