Estimating Senate Election Outcomes (Research Note 30 1997-98)

Research Note 30 1997-98

Estimating Senate Election Outcomes

Gerard Newman
Statistics Group
9 March 1998

Introduction

Recent speculation regarding the possibility of a double dissolution election being held during 1998 has heightened interest in the likely outcome of such an election for the Senate.

This Research Note provides some estimates of the likely outcome of a double dissolution election for the Senate using the 1993 and 1996 Senate election results as examples. While the 1993 and 1996 election results provide a good example of both high and low votes for the Coalition and the Labor Party, they do not take into account any new political party that may arise. For this reason the figures in this paper should be regarded as indicative only and do not attempt to predict the outcome of the actual election.

In order to calculate the estimates in this paper a computer- based Senate election results ready reckoner has been developed. See the box section for details of the ready reckoner.

Results

Table 1 provides a summary of the results based on the 1993 and 1996 voting figures together with the numbers in the current Senate. Details of the theoretical results based on the 1993 and 1996 voting figures are shown in Table 2.

Table 1: State of the Senate

	Current Senate	1996(a) base	1993 base
	Current Senate	Indicative Senate Result
LP	31	30	30
NP	6	6	6
ALP	28	28	33
AD	7	10(9)	4
Grn	2	2	2
Ind	2	(1)	1
Total	76	76	76

Figures in brackets assume Sen. Harradine is re-elected.

Based on the 1993 and 1996 election results a double dissolution election is unlikely to provide any joy for the Coalition or the Labor Party. Under both election scenarios neither of the two major parties would have a majority in the Senate. The balance of power in the Senate would remain in the hands of the minor parties and independents. If either the 1993 or 1996 voting figures were replicated, the Coalition would need three extra seats for a majority. For the Labor Party the situation is worse: nine extra seats would be required for a majority if the 1996 voting figures were replicated and six extra seats would be required if the 1993 votes were replicated.

Under a Coalition government the balance of power in the Senate would be shared by the Democrats, the Greens and an independent if either the 1993 or 1996 voting figures were replicated.

Senate Ready Reckoner

The Senate Ready Reckoner is an Excel spreadsheet based system that can be used to estimate likely Senate election outcomes given certain levels of electoral support.

The Ready Reckoner calculates the quotas received by the major parties for a given level of support and preference distribution. Both the level of support and preference distribution can be varied by the user.

The Ready Reckoner consists of a number of templates for each State, for full and half Senate elections, and for three vote scenarios - 1993 election results, 1996 election results and opinion poll data.

The Ready Reckoner is relatively easy to use and only requires a basic knowledge of Excel and some understanding of the Senate electoral system.

Senators and Members can download a copy of the Ready Reckoner from their Departmental common drive. The spreadsheet can be found in the Reckoner directory of the x drive.

Table 2 Indicative Double Dissolution Senate Election Results

NSW

VIC

QLD

SA

WA

TAS

NT

ACT

AUST

Based on 1996 Election

Percent

ALP
37.2

39.8

30.3

32.2

34.0

39.1

45.1

42.6

36.2

LP
41.4

41.4

35.4

45.8

45.4

42.3

46.7

39.0

44.0

NP (a)
..

..

15.0

..

2.1

1.2

..

..

..

AD
9.5

10.9

13.2

14.5

9.4

7.1

1.8

10.2

10.8

Greens
2.7

2.9

2.4

2.1

5.7

8.7

6.4

5.9

3.2

Others
9.2

5.0

3.7

5.3

3.4

1.7

..

2.3

5.9

Quotas (b)

ALP
4.836

5.174

3.939

4.186

4.420

5.083

1.353

1.278

LP
5.382

5.382

4.602

5.954

5.902

5.499

1.401

1.170

NP
..

..

1.950

..

0.273

0.156

..

..

AD
1.235

1.417

1.716

1.885

1.222

0.923

0.054

0.306

Greens
0.351

0.377

0.312

0.273

0.741

1.131

0.192

0.177

Others
1.196

0.650

0.481

0.689

0.442

0.221

..

0.069

Senators

ALP
4

5

4

4

4

5

1

1

28

LP
4

4

4

6

6

5

..

1

30

NP
2

1

2

..

..

..

1

..

6

AD
2

2

2

2

1

1

..

..

10

Greens
..

..

..

..

1

1

..

..

2

Total
12

12

12

12

12

12

2

2

76

Based on 1993 Election

Votes

ALP
46.9

45.0

39.4

38.0

38.3

42.7

55.3

48.6

43.5

LP
38.9

44.1

31.5

45.6

48.4

36.7

44.7

35.0

43.0

NP (a)
..

..

14.5

0.5

1.7

..

..

..

..

AD
4.9

4.0

7.0

9.9

4.1

1.7

..

6.9

5.3

Greens
3.4

1.2

3.2

1.6

5.5

6.8

..

6.1

2.9

Harradine
..

..

..

..

..

10.4

..

..

0.3

Others
5.9

5.7

4.3

4.4

2.0

1.6

..

3.4

4.9

Quotas (b)

ALP
6.097

5.850

5.122

4.940

4.979

5.551

1.659

1.458

LP
5.057

5.733

4.095

5.928

6.292

4.771

1.341

1.050

NP
..

..

1.885

0.065

0.221

..

..

..

AD
0.637

0.520

0.910

1.287

0.533

0.221

..

0.207

Greens
0.442

0.156

0.416

0.208

0.715

0.884

..

0.183

Harradine
..

..

..

..

..

1.352

..

..

Others
0.767

0.741

0.559

0.572

0.260

0.208

..

0.102

Senators

ALP
6

5

5

5

5

5

1

1

33

LP
3

5

4

6

6

5

..

1

30

NP
2

1

2

..

..

..

1

..

6

AD
1

1

1

1

..

..

..

..

4

Greens
..

..

..

..

1

1

..

..

2

Harradine
..

..

..

..

..

1

..

..

1

Total
12

12

12

12

12

12

2

2

76

(a) Included in LP for Australian total.
(b) Quota for States - 7.69% of votes: Quota for Territories - 33.33% of votes.

Research Note 30 1997-98

Estimating Senate Election Outcomes

Gerard Newman Statistics Group 9 March 1998

Gerard Newman
Statistics Group
9 March 1998