One graph. Many stories.

[TruthIsAll]

[Tux]

1. Weird

[Wilms]

2. This one is especially thought provoking. n/t

[bleever]

3. Can Donna Brazile wrap her mind around this?

I sure as hell hope so.

Maybe just one crack in the dam will do it. Maybe MD.

How long til the DNC knows they lost a rigged carnival game?

[TruthIsAll]

5. She will never get Rove to approve.... nt

.

[Wilms]

4. kick n/t

[Bill Bored]

6. Uh, doesn't this show that the pre-election polls were more

in agreement with the results than the exit polls were?

[TruthIsAll]

8. The pre-election polls were within 0.14% of the exit polls, on average.

That is close enough for me.

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x330449

[Fly by night]

9. TIA, check your figures.

Using the numbers you provided in the graph, I came up with the following deviations (by determining the deviations for each pair of polls for each state, adding them up and then dividing by 51):

exit polls vs. "reported" vote -- 2.24%
pre-election polls vs. "reported" vote -- 2.28%
pre-election polls vs. exit polls -- 2.42%

Am I wrong? Maybe I didn't do this right, but the deviations were larger than you indicate.

[TruthIsAll]

12. You are calculating average deviations, not the weighted vote percentages.

1) Kerry's weighted Pre-election vote: 50.37% = 60979/121056 
60979 is his share of the two party-vote, or 60.979
million/121.056 million.

2) Kerry's exit poll: 50.51% = 61144/121056.

3) His recorded 2-party vote percentage: 48.76% =
59028/121056.

The difference between the weighted exit and pre-election
polls is a very small 0.14% = 50.51% - 50.37%.

So the final pre-election state polls (weighted to the
national vote) were confirmed by the exit polls.



	Vote	Pre	Exit	Act	Pre	Exit	Act
Total	121056	50.37%	50.51%	48.76%	60979	61144	59028
							
AK	302	34.48%	40.14%	36.77%	104	121	111
AL	1870	40.63%	41.08%	37.10%	760	768	694
AR	1043	50.00%	46.60%	45.07%	522	486	470
AZ	1998	47.37%	46.93%	44.72%	946	938	894
CA	12255	53.85%	55.73%	55.04%	6599	6830	6745

CO	2103	49.47%	49.07%	47.63%	1040	1032	1002
CT	1551	55.32%	58.47%	55.27%	858	907	857
DC	224	87.64%	91.63%	90.52%	197	205	203
DE	372	54.22%	58.44%	53.83%	202	217	200
FL	7548	51.55%	49.93%	47.48%	3891	3769	3584

GA	3280	44.68%	43.11%	41.65%	1466	1414	1366
HI	426	50.00%	53.32%	54.40%	213	227	232
IA	1494	53.19%	50.67%	49.66%	795	757	742
ID	590	33.71%	33.33%	30.68%	199	197	181
IL	5239	56.25%	57.13%	55.21%	2947	2993	2892

IN	2448	40.21%	40.97%	39.58%	984	1003	969
KS	1171	38.14%	34.60%	37.13%	447	405	435
KY	1782	41.05%	40.76%	39.99%	732	726	713
LA	1922	45.45%	44.50%	42.67%	874	855	820
MA	2875	70.33%	66.46%	62.74%	2022	1911	1804

MD	2359	55.67%	57.04%	56.57%	1313	1346	1334
ME	727	56.18%	54.83%	54.58%	408	399	397
MI	4793	53.61%	52.55%	51.73%	2569	2519	2479
MN	2792	54.17%	54.61%	51.76%	1512	1525	1445
MO	2715	47.31%	47.47%	46.38%	1284	1289	1259

MS	1130	45.16%	43.20%	40.49%	511	488	458
MT	440	38.71%	39.28%	39.50%	170	173	174
NC	3487	48.45%	47.31%	43.76%	1690	1650	1526
ND	308	38.89%	33.58%	36.09%	120	103	111
NE	767	34.41%	36.54%	33.15%	264	280	254

NH	672	50.00%	55.49%	50.69%	336	373	341
NJ	3581	54.35%	56.13%	53.37%	1946	2010	1911
NM	748	50.00%	51.34%	49.60%	374	384	371
NV	816	50.00%	50.66%	48.68%	408	413	397
NY	7277	59.38%	63.97%	59.29%	4321	4655	4314

OH	5599	51.55%	52.06%	48.94%	2886	2915	2740
OK	1464	31.46%	34.73%	34.43%	461	508	504
OR	1810	53.19%	51.22%	52.11%	963	927	943
PA	5732	52.63%	54.41%	51.26%	3017	3119	2938
RI	429	60.87%	64.24%	60.58%	261	275	260

SC	1600	43.30%	45.78%	41.36%	693	732	662
SD	382	44.68%	37.42%	39.09%	171	143	149
TN	2421	48.98%	41.15%	42.81%	1186	996	1036
TX	7360	38.54%	36.84%	38.49%	2837	2711	2833
UT	905	25.81%	29.93%	26.65%	234	271	241

VA	3172	47.96%	47.96%	45.87%	1521	1521	1455
VT	305	56.99%	65.69%	60.30%	174	201	184
WA	2815	54.17%	55.07%	53.65%	1525	1550	1510
WI	2968	53.68%	50.21%	50.19%	1593	1490	1490
WV	750	48.42%	45.19%	43.52%	363	339	327
WY	238	30.85%	32.07%	29.69%	74	76	71
							

[Bill Bored]

13. Should we use median or average?

Outliers will always skew an average or a standard deviation, but medians are robust statistics. They won't change if there's one or two states that are way off. I wonder if these deviations should even be looked at in terms of averages or not?

I see a lot of states on the graph where the blue and tan bars are pretty close and the exit poll is way off. Not all of course, but enough to give the impression that in some states, the exit polls may have been off while the other 2 agreed.

[TruthIsAll]

15. Focus on the national pre-election vs. exit poll vote - within 0.14%

We already know that 17 states exceeded the MOE for Bush.
None for Kerry.

The point here is to show that the national weighted average of pre-election state and exit polls are in confirmation.

There is no such thing as a weighted median. The simple ratio Kerry / (Kerry+Bush) is what we need for comparison.

Just sum up the votes for each state, based 1) on the pre-election poll, 2) the exit poll and 3) the actual vote.

Note the differences between the three percentages.

Which national percentage is the outlier and stands out like a sore thumb?

[kster]

7. kick...........NT

[iconoclastNYC]

10. Group them by voting method?

Is it still fact that the states with paper ballots were the the closest to the preelection polling and the exit polls?

[Wilms]

11. Dig it. n/t

;)

[TruthIsAll]

14. A graph of state exit poll deviations by polling methods

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=296774