The most interesting thing about this poll is it demonstrates a cognitive bias.
Look at the poll results starting at the top, and running down to the bottom. Its ordered by time, starting with the earliest then running down to the most recent. It shows an exponential curve starting low, and ending high.
Rather than providing any comment on the relative merits of those PM's its more of a confirmation of what is called the 'Recency Effect' and also the Availability Heuristic. These biases mean that we have more associations with things that have happened recently. Given that we the question was 'Who was worst', we can immediately think of bad things done by the current PM, less bad things done by the previous and so on.
Pretty much my point in Post #9 but expressed better above.
By the time you get to McMahon (before many here were born, or at least old enough to remember politics), the rating is very low. I would guess that part of our makeup means that we are reluctant to condem a person as 'worst' when we don't know much about him, but because we know a lot abour Rudd and Howard, we are more comfortable condeming them.
The slight spike for Whitlam is interesting and I suspect there is are a couple of elements at play. Firstly his coming to power and losing power were events that are watermarks in Australian history, and are known by most informed people. He also introduced some of the widest social change in our political memory.
To test this more fully, I'd have included Curtin, Chifley, Menzies and Gorton (and if you insist, McEwan) in the list. And the same for the 'best PM' poll.
As personal memory fades or is confined to older people only, history tends to telescope the past. And memory becomes legend or one-line associations.
Eg Curtin = competent war time leader, Menzies = post war prosperity, Whitlam = social change but start of mass unemployment, etc. In this potted history some PMs are left out entirely, or the effects of significant ministers are forgotten (eg 'Blackjack' McEwan).
Certain PMs leave stronger impressions than others. In some cases this is due to longevity of office (Menzies), but in others it's due to percieved radicalness of policies, method of losing office (Whitlam) or that it broke a long period of rule by his opponents.
Whitlam accelerated rather than changed the country's social direction - the direction of change was set over 10 years previously (eg Menzies = abolish dictation test and Colombo Plan, Holt = Aboriginal referendum, Asian relations and abolish White Australia policy, Gorton = relax censorship and fund arts, etc). And similar changes ocurred in many other western countries at similar times to us.
Nevertheless, possibly due to his government's style (eg his relationship with the public service & self-indulgence of some ministers and the leader himself) and the unorthodox manner of its ejection, Whitlam polarised the public memory (both for and against) more than (say) the less significant Gorton and McMahon governments.
The difficulty is that the popular memory may not necessarily be correct or is an over-simplification. Eg multiculturalism and immigration are intertwined, but generally speaking governments that have promoted multiculturalism (Whitlam, Fraser, Keating) have presided over low or reduced immigration, while those who didn't care for it presided over high immigration (Menzies, Gorton, Howard).
Hawke is a partial exception, since he increased immigration in the 1980s. His multiculturalism was a balancing act between his own party (which includes ethnic branches, academics and radical sociologists) and the broader popular community opinion (which was nearer to Blainey, Howard and Ruxton).
This leads to the question: What makes the 'worst' prime minister...if you came from a very poor family and then went to University in the Whitlam era, you might recall him as being very good. (paid for your uni and might have stopped you going to Vietnam) Even though his Government ultimately failed.
Yep. Especially if you were a Canberra public servant who got a divorce and then went on to get an arts grant! But the farmer who lost their subsidies or the manufacturer who had their tariffs reduced and had to pay for the wages break-out wouldn't have been impressed though.
The high approval of Howard in the other poll might well have been because many following a modest investment program would have had their wealth increase maybe 10 fold since 1996 (for OOs with no IPs it might have been only 3 fold), and you'd have to go back to the Menzies period to find a similar time of strong growth and mild slow downs in between (though 1961 was a short, sharp shock).