skip to main content
Control of U.S. House Looks Too Close to Call

Control of U.S. House Looks Too Close to Call

Republicans likely to lose seats

GALLUP NEWS SERVICE

PRINCETON, NJ -- The CNN/USA Today/Gallup final pre-election poll, conducted Oct. 29-31, finds the Democrats and Republicans virtually tied on an important indicator of each party's strength in the race for control of the U.S. House of Representatives. According to the "generic ballot" -- a question asking voters whether they plan to vote for the Republican or Democratic candidate in their local congressional district -- the Democrats lead by 50.5% to 49.5%, a lead that is well within the poll's margin of sampling error.

The two parties have been neck and neck on this measure all year:

 

Democratic
candidate

Republican
candidate

Undecided/
other

Likely Voters

%

%

%

2004 Oct 29-31

48

47

5

Final Allocated Estimate

50.5

49.5

--

       

2004 Oct 22-24

47

50

3

2004 Sep 3-5

46

48

6

2004 Jul 30-Aug 1

47

47

6

2004 Jan 2-5

45

50

5

The Republican Party enters Tuesday's election with only a 22-seat lead in the U.S. House, 227 to 205. It is unclear whether the closeness of the generic ballot spells a handover of power in the U.S. House to the Democrats. The final estimate is identical to what Gallup recorded four years ago when Republicans managed to hang on to a majority of seats in the House. But they did so in 2000 by the slimmest of margins (only nine seats).

 

Gallup Final Pre-Election Generic Ballot, 1994-2004

 

% Voting Democratic

% Voting Republican

Net % Voting Republican

       

2004

50.5

49.5

-1.0

2002

47.0

53.0

+6.0

2000

50.5

49.5

-1.0

1998

50.0

50.0

0.0

1996

49.5

50.5

+1.0

1994

46.5

53.5

+7.0

Historically, Gallup's final pre-election result of the generic ballot (asked solely in mid-term elections prior to 1996) has been a highly accurate predictor of the actual number of votes cast nationally for the two parties across all congressional districts -- a statistic generally known as the national two-party vote.

 

Gallup Accuracy in Forecasting Republican Share of
National Two-Party Vote for U.S. House of Representatives

Actual Republican
Two-Party Vote

Republican % in
Final Gallup Poll

Poll Error

%

%

2002

53.1

53.0

-0.1

2000

50.2

49.5

-0.7

1998

50.5

50.0

-0.5

1996

50.2

50.5

0.3

1994

53.5

53.5

0.0

1990

45.9

46.0

0.1

1982

43.8

45.0

1.2

1978

45.5

45.0

-0.5

1974

41.3

40.0

-1.3

1970

45.8

47.0

1.2

1966

48.7

47.5

-1.2

1962

47.3

45.5

-1.8

1958

43.6

43.0

-0.6

1954

47.2

48.5

1.3

1950

50.0

49.0

-1.0

However, since the House is elected on the basis of 435 individual races, what matters in forecasting which party will win the majority of seats on Tuesday is how the national two-party vote relates to the percentage of seats won by each party.

The most comparable elections to look at span the Republican majority in the U.S. House from 1994 through 2002. In each of these years there was nearly a one-to-one correlation between the percentage of all votes cast for Republican candidates nationally (the GOP two-party vote) and the percentage of seats won by Republican candidates in the House. However, in three of those years (1996, 1998, and 2000) Republicans won a slightly higher proportion of seats compared with the two-party vote, while in two of those years (1994 and 2002), the Republican seat share lagged the GOP two-party vote share by a small amount.

 

Two-Party Vote vs. House Seats

 

GOP Two-Party Vote

GOP Seats
in U.S. House

Gap

 

%

%

 

2002

53.1

52.6

-0.5

2000

50.2

51.0

0.8

1998

50.5

51.3

0.8

1996

50.2

52.0

1.8

1994

53.5

52.9

-0.6

If the Republicans indeed receive only 49.5% of the national two-party vote this year, as the latest poll projects, that would be the lowest share of the two-party vote the Republicans have received since taking majority control of the House in 1994. And given the recent election patterns, such a result could very well put the continuance of the already-slim Republican majority in jeopardy.

U.S. House Vote is Predictably Partisan

Not surprisingly, a high degree of Republicans and Democrats are supporting their own party's congressional candidate in this election -- 93% for Republicans and 92% for Democrats. Political independents are breaking more heavily toward the Democrats, with 50% voting Democratic and only 39% voting Republican.

It also appears that not much ticket splitting is likely to occur: 90% of voters backing Sen. John Kerry for president plan to support the Democratic candidate for the U.S. House in their district; 88% of voters backing President George W. Bush plan to vote for the Republican candidate.

Survey Methods

Results are based on telephone interviews with 2,014 national adults, aged 18 and older, conducted Oct. 29-31, 2004. For results based on the total sample of national adults, one can say with 95% confidence that the maximum margin of sampling error is ±2 percentage points.

Results based on likely voters are based on the subsample of 1,573 survey respondents deemed most likely to vote in the November 2004 general election, according to a series of questions measuring current voting intentions and past voting behavior. For results based on the total sample of likely voters, one can say with 95% confidence that the maximum margin of sampling error is ±3 percentage points. The likely voter model assumes a turnout of 60% of national adults, based on past U.S. voting history and the current level of interest in the election. The likely voter sample is weighted down to match this assumption.

5. (Asked of those who have not already voted) If the elections for Congress were being held today, which party's candidate would you vote for in your Congressional district -- [ROTATE: The Democratic Party's candidate or The Republican Party's candidate]?

5A. As of today, do you lean more toward -- [ROTATE: The Democratic Party's candidate or The Republican Party's candidate]?

5-1. (Asked of those who already voted) In the election for Congress, which party's candidate did you vote for in your Congressional district -- [ROTATE: The Democratic Party's candidate or The Republican Party's candidate]?

 

Democratic
candidate

Republican
candidate

Undecided/
other

Likely Voters

%

%

%

2004 Oct 29-31

48

47

5

Final Allocated Estimate

50.5

49.5

--

       

2004 Oct 22-24

47

50

3

2004 Sep 3-5

46

48

6

2004 Jul 30-Aug 1

47

47

6

2004 Jan 2-5

45

50

5

 

Democratic
candidate

Republican
candidate

Undecided/
other

Likely Voters

%

%

%

2004 Oct 29-31

48

47

5

2004 Oct 22-24

47

50

3

2004 Sep 3-5

46

48

6

2004 Jul 30-Aug 1

47

47

6

2004 Jan 2-5

45

50

5

       

Registered Voters

     

2004 Oct 29-31

49

45

6

2004 Oct 22-24

49

47

4

2004 Sep 3-5

48

44

8

2004 Jul 30-Aug 1

49

44

7

2004 Jan 2-5

45

46

9

2003 Nov 14-16

46

47

7

       

National Adults

     

2004 Oct 29-31

50

44

6

2004 Oct 22-24

49

46

5

2004 Sep 3-5

48

45

7

2004 Jul 30-Aug 1

50

42

8

2004 Jan 2-5

45

46

9

2003 Nov 14-16

47

45

8

Trends for Comparison: Final pre-election poll trends

 

Democratic
candidate

Republican
candidate

Undecided/
other

Likely Voters

%

%

%

2000 Nov 5-6

49

45

6

1996 Nov 3-4

48

45

7

1984 Sep 28-Oct 1

48

46

6

1976 Oct 28-30

53

39

8

       

Registered Voters

     

2000 Nov 5-6

50

42

8

1996 Nov 3-4

47

47

6

1992 Oct 23-25

51

41

8

1988 May 13-15

52

37

11

1984 Sep 28-Oct 1

47

46

7

1980 Oct 30-Nov 1

53

38

9

1972 Sep 22-25

51

38

11

1968 Oct 17-22

48

41

11

1964 Oct 8-13

55

32

13

1960 Oct 18-23

53

34

14

1956 Oct 15-20

45

34

21


Gallup https://news.gallup.com/poll/13879/Control-US-House-Looks-Too-Close-Call.aspx
Gallup World Headquarters, 901 F Street, Washington, D.C., 20001, U.S.A
+1 202.715.3030