NAPLAN prepares us for neither the real world nor Trump

Google NAPLAN and you'll get an advertisement taking you to a website that will help you cram for it.

Of course that's not the idea. The standardised National Assessment Program for Literacy and Numeracy introduced by Julia Gillard as education minister is meant to be impossible to prepare for.

But teachers do. The next one is in May, two weeks into term two. The reputation of their students, the reputation of the school and their own reputations depend on it. Time that would have been spent encouraging year 3, 5, 7 and 9 students to ask questions and gain knowledge will instead be spent drilling them in how to pass the test.

That the two aren't the same ought to be apparent from the writing test. It's a hangover from an earlier era, before Donald Trump stood for president.

The persuasive writing test requires the student to introduce a topic, seek to persuade the reader by "for example, listing and describing parts, comparing and contrasting and showing cause and effect" and then to "synthesise" the ideas in order to draw a conclusion".

It might have been an effective technique in the 1960s, but since then decades of research have established what Trump instinctively knew – that clearly set-out arguments don't persuade. If they did, his greatest successes wouldn't have been in town hall rallies and on Twitter.

Fact checkers say 70 per cent of his campaign claims were false, but the method of argument required for students to succeed at NAPLAN couldn't defeat them. It's as if Clinton had on her team Australians who'd aced NAPLAN, and Trump had Americans who knew about people.

They key to convincing people appears to be repetition, something frowned on by those who mark NAPLAN. Study after study has found that the more often something is said the more likely it is to be regarded as true, all the more so if it is said by your side of politics.

In one, voters who were told about a policy assessed it in accordance with its merits and their ideological leanings. Another identically matched group, given the same information but also told that their side of politics supported it, backed it overwhelmingly. The political party effect "overwhelmed the impact of both the policy's objective content and participants' ideological beliefs".

In another, voters were given false information, given a "correction" telling them that it was false and in large numbers continued believed it. It happens because we are lazy. Processing a correction takes effort. Continually evaluating things to determine whether they are false takes effort. Using whatever our party thinks as a rough guide saves time.

And corrections can backfire. By repeating the original claim (as a correction or a fact check has to) they cement it in memory.

We're going to need new techniques of telling the truth from here on. They needn't be manipulative, but they'll have to move beyond the straightforward, beyond what's taught by time-stressed teachers teaching for the test.

In The Age and Sydney Morning Herald

Read more >>

Reserve Bank flying blind as the Coalition starves the ABS

So contemptuous was the Coalition of the Australian Bureau of Statistics that a few years back it outsourced the provision of its economic statistics to Channel Nine.

In the midst of the 2010 election and just days after the ABS had released the official inflation figures it issued a press release headed "Groceries Rise by over 10 per cent in past year under Labor".

Its source was the Channel Nine.  Nine had "independently demonstrated" that grocery prices have risen by 10 per cent in just over a year.

Whereas the ABS found that the price of a two-litre bottle of milk had climbed 0.1 per cent, the Nine Today Show had found it had jumped 96 cents (a percentage comparison is not available). Whereas the ABS found that the price of food in general had climbed 1.4 per cent, Nine found it had climbed 9.3 per cent.

Never mind that the ABS had personally visited around 10,000 stores to enter prices into handheld computers, never mind that the ABS had surveyed the prices of 100,000 separate goods, never mind that it weighted those prices in accordance with actual spending patterns, Nine had surveyed one basket of goods in one supermarket 12 months apart. Why bother with the ABS?

It'd be nice to think the coalition's attitude changed on taking office.

Yet within months in January its employment minister Eric Abetz was warning of a "wages explosion". He didn't say where he got the data from. It can't have been the ABS. It had wages growing at 2.7 per cent, down from around 4 per cent a few years previously. The rate has since slipped to 2.6 per cent...

The truth ought to matter to the government, especially the truth about the economy. It certainly matters to the Reserve Bank. It meets once a month to make just about the most important economic decision of the lot. Its two most important inputs are the official employment figures and the official inflation figures. And it's increasingly flying blind.

A decade ago the ABS surveyed 30,000 households each month to determine who was working and who was not. It kept going back to them until it got a response rate of 97 per cent. Now it surveys 26,000 households and accepts a response rate of 92 per cent. At the same time it's been fiddling with the way it interacts with those households, switching from phone calls to emails to save money.

As recently as March it was insisting the changes had affected the quality of its figures little. Then came August and a literally incredible jump of 121,000 in the number of Australians officially employed, an all-time record. At face value it suggested Australia had created 27 jobs every 10 minutes right around the clock for an entire month. The September number was an even bigger humiliation. If the figure had been presented normally it would have shown an even bigger collapse in employment of 172,000 people, meaning 40 jobs were lost each 10 minutes.

Something has gone horribly wrong. The ABS has called in outside consultants to help it work out what. In the meantime, the Reserve Bank is increasingly operating in the dark. Had it been less cautious and jacked up interest rates this month on the back of the published all-time jump in employment it would have shifted Australia's entire interest rate structure on the basis of a delusion.

The inflation figures suggest the Reserve Bank shouldn't be pushing up rates, but there's a chance the figures are misleading as well. Just about the entire developed world calculates inflation monthly. The ABS does it only quarterly, even though it has begged the government for the funds to do it more often. When the figures are occasionally misleading (the Reserve Bank points to a "couple of instances of quarterly readings for inflation that subsequently proved not to be representative of the general trend") the mistakes have remained official for an entire three months.

Australia's next inflation figure is out on Wednesday. New Zealand's is out on Thursday. The New Zealand figure will be shiny and new. Every three years it adjusts the "weights" it gives to different categories of spending in accordance with changes in consumer behaviour. Australia's ABS does it only every six years, a deterioration from the previous five years in order to save money. The current weights reflect spending in 2009-10.

The Coalition has been missing in action. It cut another $68 million from the ABS in this year's budget on top of $10 million cut by Labor on the way out, apparently unconcerned about the effect on statistics and its own ability to manage the economy. The head of the ABS bowed out in January and the best part of a year later hasn't been replaced. His last public statement, recorded in his last annual report said he had barely enough money to "keep the lights on".

In The Age and Sydney Morning Herald

Read more >>

Don't fall for the election spin. Victoria's job creation record is appalling

Timing is everything when you're trying to stretch the truth.

Launching his six-point plan for jobs last week Premier Denis Napthine boasted there were 100,000 more Victorians employed than when his government came to office in 2010. The figure was correct at the time he quoted it, but it was an admission of failure. It meant the number of Victorians with jobs had climbed just 3.9 per cent at a time when Victoria's working age population climbed 6.8 per cent.

No state other than Tasmania performed as badly. Victoria's Coalition government inherited an unemployment rate of 4.9 per cent and will bequeath to its successor something close to the present 6.8 per cent. Australia's national unemployment rate was also 4.9 per cent when the Victorian Coalition assumed office. But it is now 6.1 per cent, well below Victoria's.

The beauty of Napthine's boast was the timing. He spoke on Monday. Four days later on Thursday the Bureau of Statistics released updated employment figures for September that marked down the Coalition's job creation record to 89,300.  Employment is now only 3.1 per cent higher than it was back in December 2010. Victoria's working age population has climbed 6.9 per cent.

The revised figures are a measure of both how much monthly job figures bounce around these days and of how weak Victoria's job creation record has really been. But it's all right. The premier has a six-point plan. It would be more impressive if much of it wasn't simply a (glossy) reprinting of the things he was doing anyway that have so far had little success.

Not quite the oldest trick in the book, the six-point plan dates back to at least 1976. The former ABC host Stuart Littlemore was working as a media advisor to the Tasmanian Labor Premier Bill Neilson at the time. As Littlemore tells it in his book The Media and Me, polling had established "with embarrassing clarity" the public perception that the Nielson government had achieved nothing...

To counter the perception Littlemore and Labor came up with the "Neilson Plan – a 15-point program to get Tasmania working". "Neilson launched the plan (little more than a repackaging of existing policies with addition of a few items such as a retirement option for state public servants at the age of 55) about three months before he intended to go to the polls," writes Littlemore. "Thereafter he worked 'the Neilson Plan' or 'my 15-point Plan' into most answers he gave to journalists, whatever their questions may have been."

Back then it worked. "The result was that by the time we go into the campaign proper the Neilson Plan was a fixed value. Nobody even questioned its content and the perception of a do-nothing government had been turned around ... because the journalists neglected their fundamental duty," writes Littlemore.

It's unlikely to happen this time, in part because Labor has produced its own jobs plan, released the day before the Coalition's. It has five points.

Labor's plan is the boldest. Daniel Andrews wants to establish a $100 million fund to provide payroll tax relief to companies who hire unemployed young people and long-term and retrenched workers. He'll also set aside $500 million for grants to "drive growth and create high-skill high-wage jobs", plus another $200 million for a Future Industries Fund and $200 million for a Regional Jobs Fund. He is silent on how he would be able to afford it.

Napthine is also talking of big sums, but is silent on how he will afford them and how many of them are rebadging of things he is doing anyway.

The sad truth for both is that there are limits to what state governments can do. More than anything else it's the condition of the national economy that creates jobs. Local factors such population growth, the mix of industries and housing and transport make a smaller difference at the state level.

The latest ANZ forecasts show the Victorian economy growing just 2.3 per cent this financial year. Only South Australia, Tasmania and the ACT will grow by less. The national economy will grow 2.8 per cent, the NSW economy 3 per cent.

Napthine and Mr Andrews can talk as big as they like in the leadup to the November election, but when it's over, jobs will be hostage to whatever Joe Hockey decides in his December economic statement to be delivered a few weeks later. The Treasurer needs to find billions to pay for billions to pay for the Iraq campaign, billions to pay for the measures that were blocked in the Senate, and billions to replace the tax revenue that's vanishing as the iron ore price slips. Andrews is unlikely to be able to put in a good word for Victoria.

And nor is Napthine. So little influence does Victoria's premier have over the Federal government  that in May Hockey cut grants to the states for schools and hospitals by $80 billion over ten years. Almost all of those states were governed by the Coalition.

It's fine to vote on the basis of who'll do the most for jobs, but it's not fine to believe that either will be able to do very much.

In The Age and Sydney Morning Herald

Read more >>