2007 New South Wales Election: Final Analysis
Download the full paper as PDF 363Kb
Advice on legislation or legal policy issues contained in this paper is provided for use in parliamentary debate and for related parliamentary purposes. This paper is not professional legal opinion.
Backgound Paper No. 01/2008 by Antony Green
| This paper provides an analysis of the results of the 2007 New South Wales election, updating the preliminary analysis published as Background Paper No. 2/07 in May 2007. This final version completes the analysis of two-party preferred vote and includes new research on the distribution of preferences under optional preferential voting.|
New Electoral Boundaries
All results from the 2003 election have been adjusted to match the new electoral boundaries used for the 2007 election. Details of the redistribution can be found in 2004 NSW Redistribution: Analysis of Final Boundaries, NSW Parliamentary Library Research Service Background Paper No 1/05.
By-Elections and Sitting MP Indicators
By-elections have been ignored in preparing the results for each electorate and all swings are shown compared to the results of the 2003 election, adjusted for the new boundaries. Details of the five byelections conducted between 2003 and 2007 can be found on page 43.
A distinction has been drawn between which party is notionally assigned as holding a seat and who is the sitting MP. An asterisk "*" is used to indicate any candidate who was a sitting Legislative Assembly MP. Both Steven Pringle in Hawkesbury and Bruce Gaudry are shown as sitting MPs, though both contested the 2007 election as Independents. Peter Black is shown as a sitting MP in Murray-Darling, though Murray-Darling had become a notional National Party seat on the new electoral boundaries.
Format for Electorate Results
For each Legislative Assembly electorate, full details of primary and two-candidate preferred votes
are provided. The format and calculations used in the electorate results are as follows.
First Count: The votes shown for each candidate are the total primary votes received. Percentage
votes are calculated as a percentage of the formal vote for each electorate. Swing is calculated by
subtracting the percentage vote received by a party at the previous election from the percentage
received at the current election. Where the parties contesting the district differ from the previous
election, ‘ghost’ candidates (indicated by "....") have been included representing candidates not
contesting the current election. As a result, all primary swings add to zero, subject to rounding errors.
Final Count: Represents the two-candidate preferred count after the final distribution of preferences in
an electorate. All votes that did not indicate a preference to the two remaining candidates are included
in the total listed as Exhausted. Two-candidate preferred percentages are calculated by dividing the
two-candidate preferred vote by the votes remaining in the count, that is the formal vote minus the
exhausted vote. Two-candidate preferred swings are shown compared to the final candidates from
the previous election. Where the party composition of the final two candidates differs from the
previous election (e.g. Newcastle), ghost candidates appear and more than two swing figures are
Two-Party Preferred: Represents a separate distribution of preferences conducted in the 21 districts that did not finish as a two-party contest between the Labor Party and the Liberal-National Party Coalition.
Two-Candidate versus Two-Party Preferred results
Two-candidate Preferred Count: To win an electorate, a candidate must receive more than 50% of the vote after the distribution of preferences. The distribution of preferences is performed by successively excluding candidates with the lowest primary vote and distributing that candidate’s votes as preferences to other candidates remaining in the count. At the end of the count, the vote for the two final candidates is referred to as the 'two-candidate preferred vote'.
NSW Election 2007
Two-Party Preferred Count: Most Australian elections finish as a contest between candidates representing the Labor Party and the Liberal-National Party Coalition. The final distribution of preferences between Labor and Coalition candidates is referred to as the 'two-party preferred count'. At the 2007 election, there were 72 districts where the final contest was a two-party preferred count between Labor and Coalition candidates. Of the 21 districts that did not finish as two-party contests, 11 were Coalition-Independent contests, six Labor-Independent contests, two Labor-Green contests and two Coalition-Green contests.
In the 21 seats that did not finish as two-party preferred contests, a second two-party preferred count for information purposes was undertaken by the NSW Electoral Commission.
Inconsistencies in Two-Party Preferred Results
The Electoral Commission's "Report on the 2007 State Election" includes a table of two-party preferred counts by electorate. The table, published on pages 99-100 of the report, includes the additional two-party preferred counts for the 21 districts that did not finish as two-party contests.
Unfortunately, the other 72 entries in the table do not use the actual two-candidate preferred count in each district. The figures are instead derived from unaudited preference counts conducted on election night. While all primary votes were re-counted after election night, the indicative preference counts
were not re-counted. As a result, the entries for all 72 electorates that finished as two-party preferred contests differ from the final published distribution of preferences. This publication has been produced by using the actual distribution of preferences in the 72 electorates that finished as two-party preferred counts. This produces minor discrepancies between the totals used in this publication and those released by the Electoral Commission's election report.
None of these discrepancies significantly alter the margins in seats.
Analysis of Preference Flows
A new feature provided in this publication is detailed analysis of preference flows by party. This has been accomplished by building a database of preference flows from the tally sheets used to conduct the indicative preference count on election night. An explanation of how the preference flows were conducted along with analysis of the flows can be found on page 55. Full details of preference flows by booth and by candidate have also been produced and can be found on the Parliamentary Library web site.
Australian Labor Party and Country Labor
Labor candidates nominated as Country Labor are indicated by the underlining ALP. Some comparison of Labor/Country Labor vote has been provided with state wide totals, but the redistribution has made it too difficult to compare Country Labor vote by seat.
All results are based on details made available by the NSW State Electoral Office. However, responsibility for all calculations and omissions is taken by the author. Note that due to rounding errors, some tables do not add to exactly 100%.
My thanks to the NSW Electoral Commissioner, Mr Colin Barry, for all his assistance in providing the results of the state election, and in particular the Electoral Commission's Ian Brightwell for making the data available in electronic format.