In my post-election conversations and eavesdropping I have heard several people refer to informal votes as ‘donkey votes’.
In standard usage – still supported by the Macquarie Dictionary and several random Australian politics books I checked – a donkey vote is defined as the practice of numbering all candidates in the order they appear in the ballot paper, rather than according to the voter’s political preference. This is a formal vote, which will eventually go to whichever serious candidate appears first in the list of candidates. However Wikipedia is wobbling, suggesting that informal votes can also be classified as ‘donkey votes’.
In Bryan Garner’s five stages of language change, ‘donkey vote’ is at stage two or three. Several of the people I have heard use ‘donkey vote’ when they mean informal vote have university degrees.
Stage 2: The form spreads to a significant fraction of the language community but remains unacceptable in standard usage.
Stage 3: The form becomes commonplace even among many well-educated people but is still avoided in careful usage.
While I tend towards language conservatism, I am less opposed to this change than others (eg reticent/reluctant, uninterested/disinterested) where I think important distinctions are being lost through clumsy usage. I take it that the key idea being conveyed in the expression ‘donkey vote’ is that the voter is a ‘donkey’ who lacks the intelligence or interest to cast an informed vote. Arguably someone who cannot fill in the ballot paper correctly is a bigger ‘donkey’ than a person who knows how to do it but does not care who wins.