Respectable people can think there is human-induced climate change. And that, on balance, it is a bad thing (likely more harmful the greater and the faster it is). And that it is worth considering whether, and if so, how, public policy might respond (Pyrmonter has engaged with several cats on this issue in discussion fora). Others – especially some around Catallaxy – take strongly contrary views.
What respectable people cannot do, if they wish to retain their respectability, is engage in dishonest argument, such as passing off opinion surveys derived from biased samples of the population as representative of it. Yet, again, this is what the ABC is doing with a sample taken from its ‘Australia Talks’ project undertaken last year.
Recently the ABC told us, based on the Australia Talks data*, that most Australians are troubled (going so far as to say, ‘losing sleep’, though that is surely journalistic cliché) about climate change.
Cats will recall that Australia Talks involved a survey using a large sample (somewhere over 50,000 responses) of respondents to an earlier ABC data gathering exercise, Australia Votes. The promoters of the project say they have weighted the responses by various social characteristics so that it can fairly represent the entire population. What they can’t do however is exclude the inevitable bias involved in selecting their data from that subset of the population that is motivated to interact with the ABC.
While the survey can find respondents who are coalition, PHON or LDP voters who interact with the ABC (including Pyrmonter), extrapolating from those sub-sets of the population assumes that those who interact with the ABC are representative of the population. Such an assumption is quickly contradicted by consideration of the ABC’s media channels: left-leaning youth radio networks (the ‘J’ brand); generally older, upper middle class listeners to western art music (ABC FM) (though, presumably biased against those who now use more modern platforms like Spotify or Apple Music); middle class listeners to the ABC metropolitan stations; the mostly left leaning listenership of RN; and that small share of the population that continues to watch ABC television, now almost shorn of its once staple feed of BBC dramas and documentaries.
Australia Talks tells us something interesting. It tells us what those who interact with the ABC (and who, generally, are that part of the population likely to resist changes to it) think. What it doesn’t tell us is what the national population thinks. The ABC should stop pretending it does. And those in a position to inquire about the ABC’s conduct – its staff, its management, its Board and their paymasters – should be asking why it continues to do it.
* The ABC also refers in passing to a survey undertaken by the Australia Institute. Cats are referred to the writings of Kevin Bonham, Anthony Wells and Nate Silver about the usefulness (or not) of campaign polling.