10976RE: [GreenLeft_discussion] Re: Latham's Leichhardt meeting
- Nov 24, 2004Bob Gould wrote:
>Michael Berrell['s] maths are pretty selective. When you look at Malcolm Mackerras's pendulum, the 2004 election 47.5 per cent two-party preferred vote for Labor wasn't as low as its 1996, 1975 or 1966 votes.To assist the debate, I attach my Excel spreadsheet workbook file, which includes spreadsheets of figures and three charts of the ALP vote, the first being the historical first preference vote from 1901, the second being the first preference vote from 1983 (the period of big decline under the impact of economic rationalist policies) with a couple of stabs at doing regression lines (both linear and polynomial), the third being a look at the postwar first preference and two party preferred votes. If you move your mouse over any of the data points on the chart lines, it will display the value of that point.
I think Bob Gould's maths are selective, basing himself on the ALP two party preferred vote, which includes all preferences beyond the first preference vote. However, even here it must be said that it is declining, faster than the first preference vote since 1998. And I read on the Australian Electoral Commission website that the ALP's two party preferred vote was 47.26%. Slightly selective stats, Bob.
However, I think the underlying argument is about the core support of a party, ie its first preference vote. The 2004 result, at 37.6%, is the worst result since the Great Depression and the related elections of 1931 (27.1%) and 1934 (26.8%), outside of the initial rise of the ALP in the 1901 - 1910 period. Granted, it is only 0.2% worse than the 2001 result, and at best can be said to almost hold the line, but at worst can be said to confirm a sliding trend. In fact if you have a look at the post 1983 graph of the first preference vote, in the post 1983 era there were only two rallies, a big spike in 1993, and small hump in 1998. The 2004 first preference result doesn't even do that, and for this reason confirms a downward trend.
Speaking about ALP first preference vote trends (see my 'Extrapolation' and 'Chart 2, post 1983' spreadsheet tabs), earlier this year I looked at the linear trend between the 1983 and 2001 elections, and found that the falling trend in this period was a slow decline of the order of [ALP First Preference Vote] = -1.6[Number of Elections since 1983] +50.175, with which I predicted the 2004 ALP first preference result to be 37.375%. The final result was a bit higher than that by only 0.265 percentage points.
With the 2004 results to add into my spreadsheets, I now come up with a slightly amended and lesser falling trend for the ALP first preference vote, of the order of [ALP First Preference Vote] = -1.4783[Number of Elections since 1983] + 49.769. On the basis of this, I predict that at the next election the ALP first preference vote will be somewhere around 36.46%, and the election after that it will be around 34.99%.
The two party preferred vote for the ALP historically (see my 'First pref vs 2 party pref' and 'Chart3, 1st pref & 2 party pref' spreadsheet tabs) has seen the two party preferred vote grow over the years on top of the first preference vote, jumping in 1990 to 10.5 percentage points above, and hovering in the five to ten percentage points above figure since then (this could reflect Greens preferences). However, if you look at my chart (Chart 3) where I plot both first preference and two party preferred ALP votes, the two party preferred vote has generally followed the fortunes of the first preference vote, with a falling trend evident post 1983, despite peaks in 1993 and 1998, and especially since 1998.
I have analysed the relationship between the ALP first preference vote and the two party preferred vote (see my 'First pref vs 2 party pref' and 'Chart4 1st pref-2 party scatter' spreadsheet tabs) for the post war period, and found a weak relationship of the order of [ALP Two Party Preferred Vote] = 0.4151[ALP First Preference Vote] + 30.578, with a strength of R squared = 0.3725 (0 = no relationship, and 1 = a perfect 1 to 1 relationship), meaning that if the first preference vote was zero, there still would be an ALP (sympathy?) two party preferred vote of 30.578%. However, this has to be read against recent historical trends, and the question becomes, at what stage of decline of the ALP first preference vote would the ALP two party preferred vote collapse?
To partially answer this question, I did the same analysis for the 1983 and onwards data (see my 'Chart5 1st pref-2 party scatter' spreadsheet tab). This comes up with a strong relationship of [ALP Two Party Preferred Vote] = 0.4023[ALP First Preference Vote] + 33.039, with a strength of R squared = 0.7008.
Though if you look at the pretty steep decline in the ALP two party preferred vote since 1998, a trend line based on the three elections starting with 1998 suggests an extremely strong relationship of [next election ALP two party preferred vote] = -1.85[the number of elections since 1996] + 52.8, which has an R squared of 0.9978. This suggests that the next two federal elections ALP two party preferred vote will be 45.4% and 43.55% respectively.
Thus, when taken together, the analyses suggest that in addition to the predictions I provide above for the ALP first preference vote for the next two elections, I also predict that the ALP two party preferred vote will be between 45.4% and 47.71% for the next election, and 43.55% and 47.11% for the one after that.
Though you never know what a recession will do to these predictions, this being the wild card, and I think just around the corner.
The information contained in this e-mail message and any attached files may
be confidential information, and may also be the subject of legal
professional privilege. If you are not the intended recipient any use,
disclosure or copying of this e-mail is unauthorised. If you have received
this e-mail in error, please notify the sender immediately by reply e-mail
and delete all copies of this transmission together with any attachments.
- << Previous post in topic Next post in topic >>