surveys and statistics and politics

A final Age article for tonight:

First, let me just say that I don't like statistics. Or rather, I don't trust people who try to convince me of something by quoting figures. In order to trust statistics I need to know about sample sizes, about questions asked, about demographics . . . I need to know about the methodology behind the figures. So do 7 out of 10 dentists. Or chiropractors. Or window cleaners.

So anyway, what we have here is a situation where two opposing opinions are being justified by different sets of figures. Dig it:

In 2005 the Australian Government introduced WorkChoices; an amendment to the Workplace Relations Act 1996. You can read all about that here.

So anyway, there was a study on the effect of WorkChoices on the Australian working population: The study was titled Australia@Work: the Benchmark Report (click here if you want to read the report (.PDF)) and it was authored by the Workplace Research Centre at the University of Sydney. The report was funded by both the Australian Government and Unions NSW.

Now, it seems that the Government was not happy with the report's findings:

The Age reported yesterday that the survey of 8343 people had found employees on individual contracts introduced under the Government's WorkChoices scheme worked longer hours for $100 less weekly pay than those on collectively negotiated agreements.

Why would the Government be unhappy about that? On the eve (we expect) of a federal election?

The story prompted a political storm, with Mr Hockey [Minister for Employment and Workplace Relations] attacking the authors, from the Workplace Research Centre in Sydney.

He said centre director John Buchanan and report author Brigid Van Wanrooy were "former trade union officials who are parading as academics" and that "I'm not sure that this institution is known for academic rigour".

Well, he is a politician. If you don't like the message, you can probably get political mileage from attacking the messenger.

Mr Hockey said that ABS data showed that workers on Australian Workplace Agreements "earn nearly twice as much as people on awards". Prime Minister John Howard also said that the ABS "tell us that people are better off under AWAs".

Hmm. Well, if the Australian Bureau of Statistics says you're right, you must be right! Or are you?

But the ABS's assistant director of labour employee surveys, Valerie Pearson, said the survey Mr Hockey had referred to "was conducted only six weeks post WorkChoices".

"The only thing that would have been picked up in the survey was any AWAs negotiated in that six weeks," Ms Pearson said.

Oops! So much for using figures! Anyway, one good thing to come out of all this:

Dr Buchanan

[director of the Workplace Research Centre]

said yesterday that he would consult a defamation lawyer over the comments to "explore legal options".

"Saying we've concocted stuff is very serious for academics … I'd prefer a retraction, but I'm sick and tired of the way they're bagging academics," he said.

Good on him. And who knows? Maybe he'll end up owning Joe Hockey's website!