JWS Research: 54.4-45.6 to Liberal in Frankston

With the numbers in the Victorian Legislative Assembly at 44 seats for the Coalition, 43 seats for the Labor and one seat for troubled independent Geoff Shaw in Frankston, the latter personage and the electorate he serves have been looming large in the state’s political life of late. Shaw resigned from the Liberal Party in March and last month had 23 charges of misconduct in public office and obtaining financial advantage laid against him. It is thus of interest to relate, courtesy as always of GhostWhoVotes, that robo-pollster JWS Research has tested the waters in an electorate that could potentially decide if the government survives to the end of its term.

The prognosis for the government is favourable: despite Geoff Shaw having a disapproval rating of 64% and an approval rating of 20%, the poll credits the Liberals with a 54.4-45.6 lead on two-party preferred, suggesting a 2.3% swing in their favour (Shaw having won the seat from Labor in 2010 by a margin of 2.1% off a swing of 5.3%). The primary votes are 49% for the Liberals (46.8% at the election), 35% for Labor (36.4%) and 9% for the Greens (8.4%). The sample for the poll was 535, with a theoretical margin of error of 4.2%.

It should be noted though that JWS Research appeared to have a Liberal skew in its federal election campaign polling. Two sets of polling were conducted, one targeting eight seats four weeks out from the election, the other targeting six seats two weeks out. The average two-party result from the fourteen polls was 3.1% higher for the Coalition than the eventual election result, with eleven of the 14 poll results leaning in their direction. While the Labor primary vote tended to be accurate, Coalition primary votes were exaggerated at the expense of “others” (which can partly be explained as the polling being conducted too early to catch the late-campaign surge for the Palmer United Party) and to a lesser extent the Greens.

UPDATE: The Herald-Sun has a full set of results from the poll, including a thick slice of attitudinal questions. The poll also asked separately how voters would vote at a by-election as distinct from a state election (the latter results being those featured above), though with much the same result (the Liberals one point higher and the Greens one point lower). Eighteen per cent said the knowledge that a Labor by-election win might trigger a state election would case their vote to change, which is either a lot or a little depending on your perspective. The no response was at 73%. The results also suggest that Denis Napthine is more popular than Daniel Andrews in this part of the world.

Author: William Bowe

William Bowe is a Perth-based election analyst and occasional teacher of political science. His blog, The Poll Bludger, has existed in one form or another since 2004, and is one of the most heavily trafficked websites on Australian politics. View all posts by William Bowe