It’s always nice to see evidence that supports your prior beliefs, which is why it’s important to avoid confirmation bias (seeking out confirming evidence, while ignoring or discounting the other kind). Since this ANU study of voters’ choices in the leadup to the May election is, AFAICT the only one to be published so far, I can cite it without fear of this bias.
I’m not usually keen on the excuse that “we lost because we didn’t get our message across”, but in relation to the last election, I said this before the election. Commenting at the halfway point, I said
The first half of the 2019 election campaign was the worst I’ve ever seen, especially relative to the possibility for real debate. Both sides ran continuous attack ads focusing on the opposing leader, playing into the gladiatorial model favoured by the Press Gallery. Labor, in particular, seemed to have forgotten it had any policy offer.
As the past tense indicates, when I wrote this I thought things had changed. For a day or two, Labor ran ads talking up its positive policies and focusing on the whole frontbench rather than Shorten alone. But that stopped almost as soon as it started and it was back to the fact that Morrison only cared about “the big end of town”. Apart from the clunky and dated rhetoric, we already knew that. By contrast, even as a close follower of politics I didn’t know Labor had an excellent dental policy until Tanya Plibersek mentioned it after the election.
The ANU study backs this up, first by saying that lots of people changed their minds at the last minute, which isn’t consistent with deepseated hostility to Labor policy, and second by saying that the big negative was reactions to Shorten, presumably driven in part by the Liberals’ negative campaigning.
The crucial point here is that, by playing the gladiatiorial leadership game, Labor was setting itself up to lose. Campaigning on policy would have reinforced the point, obvious since the election, that the coalition didn’t have any.