Ted Koppel Demands Candidates Drop Out 42 days before a Vote is Cast

Misinterpreting Public Opinion Polls

Polls no longer just measure the political landscape -- they also shape it. The Pew Research Center is a reputable source. However, polls are being misused.

On December 8, 2003 Pew Research Center released National and state polls on the Democratic presidential race that had been conducted between November 18 and December 4. Based on these poll results, the next day during a presidential debate he was hosting, ABC's Ted Koppel demanded that three of the candidates (Dennis Kucinich, Al Sharpton and Carol Moseley Braun) drop out of the race

(At the time of Koppel's demand, in Pew's national poll, John Kerry had 6%, John Edwards 5%, Al Sharpton 5%, Carol Moseley Braun 4% and Dennis Kucinich 2%) *

I have put together this page summarizing the Pew Resarch Center results compared with how voters actually voted. It also summarizes how Koppel's analysis and ABC's subsequent decision to discontinue coverage influenced the outcome.

* The cited margin of sampling error was.plus or minus 5 percentage points for the national poll. For the state polls,: Iowa: plus or minus 6 percentage points; New Hampshire and South Carolina: plus or minus 5 percentage points.

the context

Poll Compared to Actual Results

 
National
Iowa
New Hampshire
South Carolina
 

Pew Poll -- Dec 8
Pew Poll Dec 8
Actual  caucus delegates -- Jan 19
Pew Poll Dec 8
Actual Vote--Jan 27
Pew Poll --Dec 8
Actual Vote --Feb 3
Howard Dean
15%
29%
18%
34%
28%
7%
5%
Wesley Clark
15%
3%
0%
8%
13%
11%
7%
Dick Gephardt
12%
21%
11%
5%
withdrawn
10%
withdrawn
Joe Lieberman
12%
1%
0%
8%
9%
9%
2%
John Kerry
6%
18%
38%
20%
39%
3%
30%
John Edwards
5%
5%
32% *
4%
12%
16%
45%
Al Sharpton
5%
5%
0%
1%
0%
8%
10%
Carol Moseley Braun
4%
4%
withdrawn
1%
withdrawn
2%
withdrawn
Dennis Kucinich
2%
2%
1% *
1%
1%
0%
1%

* Kucinich's actual totals are no doubt higher and Edward's lower because of a deal between the two camps

On actual election day,

In Iowa, the prediction missed Edwards by 27 percentage points, Kerry by 20 points, Dean by 11 points, and Gephardt by 10 points.

In New Hampshire, the prediction missed Kerry by 19 percentage points, Edwards by 8 points.

In South Carolina, the prediction missed Edwards by 29 percentage points, Kerry by 27 points.

 

It Had An Effect

Pew Research Center conducted research for Democratic presidential race between November 18 and December 4

December 8, 2003 -- Pew Research released their poll results.

December 9, 2003 -- Based on these poll results (the day following their release) during a presidential debate he was hosting, ABC's Ted Koppel demanded that three of the candidates drop out of the race. (Sharpton, Moseley Braun and Kucinich) (view debate remarks)

December 10 -- ABC would announce the day following Koppel's analysis that it was withdrawing all coverage from the Sharpton, Mosely Braun and Kucinich campaigns.

1. John Kerry who was designated viable by Ted Koppel had only 6% nationally (and 18% in Iowa) at the time of Koppel's designation. He would finish with 38% in Iowa.

1. John Edwards had 5% both in Iowa and nationally at the time of Koppel's comments. However, he would be designated by Koppel as a viable candidate. He would ultimately total 32% in Iowa.

2. Al Sharpton had 5% both in Iowa and national at the time of Koppel's comments. Following the comments and media designation, he would get 0% in Iowa.

3. Carol Mosely Braun who had 4% both in Iowa and nationally at the time of the comments would withdraw before the caucuses took place. qsw

4. Dennis Kucinich the candidate with probably the least name recognition had 2% in Iowa and nationally. He would gain 2% in Iowa (and probably more because of a behind the scenes deal between Kucinich and Edward supporters).

January 19 -- The Iowa caucuses took place -- 42 days after Koppel's analysis. During exit polls, most voters in Iowa said they had made up their minds in the last 3 days.

January 27 -- The first vote took place in New Hampshire 49 days after Koppel's analysis.

 

To View Koppel's comments during the December 9 Presidential Debate

 

Pew Research Center.

National and state polls on the Democratic presidential race

complete report is at http://people-press.org/reports/display.php3?ReportID=198

from press release:-- December 8, 2003

The polls were taken Nov. 18-Dec. 4.

The Iowa poll of 394 likely caucus voters has a margin of sampling error of plus or minus 6 percentage points. The New Hampshire poll of 585 likely voters, the South Carolina poll of 566 likely voters and a separate national poll of 469 voters who are Democratic or lean Democratic had margins of error of plus or minus 5 percentage points.

When results don't total 100 percent, the remainder didn't know or refused to answer.

NATIONAL: Howard Dean, 15 percent; Wesley Clark, 15 percent; Dick Gephardt; 12 percent; Joe Lieberman, 12 percent; John Kerry, 6 percent; John Edwards, 5 percent; Al Sharpton, 5 percent; Moseley Braun, 4 percent; Kucinich, 2 percent.

IOWA: Dean, 29 percent; Gephardt, 21 percent; Kerry, 18 percent; Edwards, 5 percent, Kucinich, 4 percent; Clark, 3 percent; Lieberman, 1 percent; Moseley Braun, 1 percent; Sharpton, 1 percent.

NEW HAMPSHIRE: Dean, 34 percent; Kerry, 20 percent; Clark, 8 percent; Lieberman, 8 percent; Gephardt, 5 percent; Edwards, 4 percent; Kucinich, 1 percent; Moseley Braun, 1 percent; Sharpton, 1 percent.

SOUTH CAROLINA: Edwards, 16 percent; Clark, 11 percent; Gephardt, 10 percent; Lieberman, 9 percent; Sharpton, 8 percent; Dean, 7 percent; Kerry, 3 percent; Moseley Braun, 2 percent; Kucinich, 0 percent.