Friday, June 17, 2011

Belshaw's World - regional university analysis ‘pretty meaningless’

Note to readers: This post appeared as a column in the Armidale Express on 8 June 2011. I am repeating the columns here with a lag because the Express columns are not on line. You can see all the columns by clicking here for 2009, here for 2010, here for 2011.

In my last column I spoke of the Grattan Institute report into regional development in Australia. I was very critical and said that in this column I would look at their comments on regional universities.

Why is this important?

Well, Grattan recommended that Governments re-consider whether additional funding to regional universities was justified by social and cultural benefits given limited economic impact. Instead, the report proposed that the Government should consider providing additional support to regional students to attend capital city universities.

So how did the Grattan authors reach this conclusion?

The first plank in their methodology was to compare university cities with non-university cities that had roughly comparable populations. So Armidale was compared with Nowra-Bomaderry, Wollongong with the Sunshine Coast.

The Sunshine Coast actually has a university with over 8,000 full time student equivalents. However, it was counted as a non-university city on the grounds that the university was a relatively small proportion of the local economy.

Note that this comparison has very little to do with regional economic performance or contribution. We are talking cities, not regions. In Armidale’s case, the analysis was actually restricted just to the old city of Armidale.

Having defined cities as a comparative substitute for regions, just what data was used?

Some data was derived from the 2006 census – prime labour force, higher education level and the unemployment rate. Other data looked at changes between the 2002 and 2006 censuses – private sector and population growth. Then patents per 100,000 people was used as a proxy for innovation.
I cannot comment on the patents numbers because I don’t know how the number was derived.

Looking at the other numbers, they strike me as pretty meaningless.

The jobs lost in Armidale during the 1990s from changes to education led to actual population losses. The city has just begun to recover.

By contrast, Nowra-Bomaderry’s growth has been affected by the sea change phenomenon. You would expect both population and private sector growth to be faster.

In similar vein, the sun and sea change phenomenon makes the Sunshine Coast one of Queensland’s fastest growing areas. By contrast, while Wollongong is fast changing, it remains an industrial city.
I could go on to examine all the city pairs. I don’t think that I need to.

My point is that the analysis is pretty meaningless in a general sense, more so if it is to be used as a measure for economic contribution from regional universities.

The report also purports to challenge the idea that regional universities increase either the regional participation rate or the retention of university graduates in regional areas. Again we have city based comparisons.

I really struggled to understand the logic in the analysis.

The raw data actually does suggest a positive effect. Leaving that aside, the small area comparisons fatally affects the analysis. Again, I will take UNE as an example.

UNE is a national university (I have said before why I hate the word regional) with a majority of students now external coming from many places. If you want to measure local or regional effects on participation or retention, you have to net out most of the external students.

If you focus just on UNE’s local and regional students, then you have to factor in transport and time costs.

I don’t think that there can be any doubt that in the immediate area around Armidale – perhaps a sixty-mile radius – participation is higher. Once you move outside that area, then people have to live in residence at whatever university they study. This is one reason why country participation rates are lower than the city. Study costs are higher.

Beyond the sixty-mile radius from Armidale, UNE still has an effect on participation rates because of lower residential costs and, to a diminishing degree, lower transport costs. Parents with cash with cash will send their kids where those kids want to go. Not all parents are so lucky.

So far as retention is concerned, it remains a hard fact of life that most people studying at UNE have to go where the jobs are. However, there is now plenty of evidence, including my own work, to indicate that country education does help retain people in areas like medicine and law. You can’t pick this up in the gross statistics.

Finally, the absurdity of suggesting that country kids should be redirected to metro universities can be seen simply in the accommodation problems now being experienced by kids attending Sydney universities. Reasonable rental accommodation is no longer available despite the huge student accommodation building programs now underway.


Mark said...

Grattan being paid off by big city univeristies perhaps?

Greg said...

Lies, damned lies and statistics.

Jim Belshaw said...

Probably not, Mark. Just metro myopia!

You are right, Greg. The problem with this type of analysis is that we all have to spend tme responding. It wouldn't matter if the analysis just vanished, but it gets quoted.

For both, I don't know whether or not you picked it up, but Newcastle was classified as slow growth, ie not worth supporting! Slow growth? I wish they would check facts, for the lower hunter is hardly slow growth.

Rod said...

Very good blog.

Jim Belshaw said...

Thank's Rod!