Welcome to European Tribune. It's gone a bit quiet around here these days, but it's still going.
Display:
You only need to do it in the largest municipality, or even the largest single ward in the state, really.

We have met the enemy, and he is us — Pogo
by Migeru (migeru at eurotrib dot com) on Wed Jan 16th, 2008 at 09:03:30 AM EST
[ Parent ]
If that would be so, statistical analysis would have sput out one single ward that is off, rather than find a debated effect across-the-board.

*Lunatic*, n.
One whose delusions are out of fashion.
by DoDo on Wed Jan 16th, 2008 at 06:06:46 PM EST
[ Parent ]
Look at the chart I made of the county-level effect. No county shows a vote reversal except for Hillsborough, which is the largest county of all.

We have met the enemy, and he is us — Pogo
by Migeru (migeru at eurotrib dot com) on Wed Jan 16th, 2008 at 06:12:10 PM EST
[ Parent ]
But all but two counties show a machine-counted vote surplus, and Coös County displays a difference as big as Hillsborough.

*Lunatic*, n.
One whose delusions are out of fashion.
by DoDo on Wed Jan 16th, 2008 at 06:14:28 PM EST
[ Parent ]
I discuss this in the other thread:
Let's do a nonparametric test here. We observe that only two of the ten counties show machine counts favouring Obama. What are the odds of that? This is like tossing a two-headed coin (Clinton on one side, Obama on the other) ten times and getting two or fewer Obamas. The odds of this are 1 + 10 + 10 * 9 /2 divided by 2 to the 10th power, or 56/1024, or 7/128, or about 1/18. This is not quite significant at 95%. Moreover, it would be equally suspicious if machines favoured Clinton in only two counties or less. But for a two-sided alternative the odds of an extreme result are 1/9, so not quite significant even at 90%.


We have met the enemy, and he is us — Pogo
by Migeru (migeru at eurotrib dot com) on Wed Jan 16th, 2008 at 06:19:26 PM EST
[ Parent ]
What's the relevance?

Maybe if the hypothesis is that there was vote reversal in a single county, the odds to calculate is of Obama ahead only in 3. But I don't think such a calculation is realistic by using a binary variable (Obama ahead/Clinton ahead).

*Lunatic*, n.
One whose delusions are out of fashion.

by DoDo on Wed Jan 16th, 2008 at 06:26:13 PM EST
[ Parent ]
This is difference from the question of who's ahead. It's the question of which direction the machine effect goes. There are counties in which the machines don't change the lead, but here I'm looking at how many times the machines favour obama or Clinton (regardless of whether it is by enough to change the lead). In fact, the lead only changes in 3 out of the 10 counties. But to study the odds of the lead changing you need a model of the average lead and the variability of the vote, so it is no longer a nonparametric test.

We have met the enemy, and he is us — Pogo
by Migeru (migeru at eurotrib dot com) on Wed Jan 16th, 2008 at 06:30:30 PM EST
[ Parent ]

Display:

Occasional Series