Finding fault in Australia’s latest data dump

Higher education data have been flying thick and fast in Australia, where a report on 2018 undergraduate applications, offers and acceptances – unusually released on a weekend, Sunday 28 October – was rapidly followed by comprehensive 2017 student statistics on the Monday and 2018 graduate outcomes survey results on the Wednesday.

It’s a welcome change in a country where up-to-date statistics on lucrative international students are published regularly, but information about their domestic counterparts is sporadic at best and well past its use-by date when it finally materialises.

Why so much rich data so close together? The full-year 2017 student data was released just a fortnight or so after the half-year data – and a day after the first meaningful tranche of 2018 data, for that matter.

As Times Higher Education understands it, 2017 student information was collated months ago so that the department could ensure that universities did not exceed teaching grant caps imposed last December. So why wait until so late in the year to release the data?

Is it just that several opportunities for low-level PR announcements suddenly became available in the minister’s calendar a few days apart? Or is someone trying to hide something by burying it in reams of information?

Times Higher Education asked the department and received a fairly predictable answer: “The timing for the release of departmental reports such as these is at the discretion of the minister for education and training.”

If it’s a choice between a conspiracy and a cock-up, every journalist knows that you opt for the latter every time. But when data in the education department’s normally scrupulously accurate tables are riddled with errors, you wonder if someone has been pressed to get something out in just too much of a hurry.

For example, the applications, offers and acceptances summary report says that applications were down by 3.3 per cent compared with last year, when the appendix figures suggest that they were down 4.1 per cent. The summary document says that the offer rate has increased by 0.7 percentage points, while the appendix suggests that it rose by 1.5 percentage points.

In other inconsistencies, different overall application figures and offer rates are cited in different appendix tables. And in one table, the offers made by two universities were mixed up.

This explains why Charles Darwin University’s offers shot up from 2,487 one year to 13,398 the next, while the University of Tasmania’s offers crashed from 15,062 to 2,196 – at least until the inversion was corrected after Times Higher Education pointed out the improbability of all this.

Why so many mistakes? One explanation offered to Times Higher Education was that the University of Notre Dame Australia this year switched its medical degree from an undergraduate to a postgraduate program. Its medical applications had to be excluded this year, because the report only tallies undergraduate applications.

This somehow led to inconsistencies in this year’s report, because Notre Dame medical studies application figures were also excluded from the 2017 data so that people didn’t end up comparing apples with cumquats.

Times Higher Education isn’t sure this passes the sniff test. Do so many people really apply to study medicine at Notre Dame? Anyway, when THE asked Notre Dame why its applications had crashed from 4,600 one year to 1,300 the next, it said student numbers had risen and it mentioned nothing about medical studies.

The lesson, if any can be drawn, is not to read too much into this year’s figures without triple-checking them first – as the department apparently has neglected to do this time around.

John Ross is the Asia Pacific editor at Times Higher Education.