Tuesday, March 24, 2009

Problems with language and definition in public policy

I am still mildly annoyed with myself over the misuse of the word community that I referred to in Problems with words and measurements. Annoyed because I too have been guilty from time to time of using the word loosely.

In some ways life is too short to always subject things people say to semantic analysis; the working assumption that the writer or speaker means the same thing by a word that you do does make life easier. Yet there are a lot of cases around at present that show the way in which words can be misused.

I was just thinking of some of the examples that I have quoted over the last year or so.

The misuse of the word racist or racial prejudice to describe things that have nothing to do with race, but instead relate to other forms of perceived prejudice.

The use of the words efficiency dividend. This implies a return from increased efficiency - who could argue with that - when in fact it is a device to try to force increased efficiency.

The use of the phrase risk management when in fact the writer really means risk avoidance.

There is a special problem with official English - I am sure with official French or Chinese as well - because of the way that English builds up what is in fact jargon that rests on and is influenced by underlying administrative concepts and structures.

Consider, as an example, the current use of the words remote and very remote, something of great interest to the Commonwealth Government in an Aboriginal context. As often used in official quarters, these words actually mean remote and very remote as defined by ARIA.

ARIA is simply a mechanical device for measuring remoteness using simple criteria. At this level it can be useful if limited measure. However, difficulties arise if the measure is used to define boundaries for public policy purposes.

To illustrate with a simple example.

The NSW country town of Balranald has a population of around 1500 and lies more than 850k (over ten hours driving time) south west of Sydney. Under ARIA it is classified as outer regional.

Darwin, the capital of the Northern Territory, is also classified as outer regional. Do the people in Balranald have the same real access to services as those in Darwin? Clearly not. The comparison is silly. Yet the use of ARIA for policy purposes implies that they do.

Take Sydney. Australians all know what Sydney is. Or do we?

I, for one, have been caught out here because I failed to realise that the often quoted statistics for Sydney actually refer to the Sydney Statistical District. This includes the Blue Mountains to the west of the city, the Central Coast to the north.

The problem here is that statistics drive policy, and both statistics and policies are affected by constantly changing boundaries adopted for political, administrative and public policy purposes. There are now at least three very different definitions of Sydney.

To extend the argument, consider the NSW or New England Mid-North Coast. Historically, this term is used to describe the coastal strip between the Northern Rivers in the north and the Hunter Valley. More recently, the NSW Government has excised the Clarence Valley - the biggest of the Northern Rivers - from the Northern Rivers and added it to the Mid-North Coast.

This may sound a small change. Does it matter? Apart from creating confusion in historical terms, it actually invalidates - creates a break - in past statistical series.

It has become almost impossible without considerable local and technical knowledge, a fair bit of time and sometimes expense to undertake comparative analysis of performance over time at sub-state level.

I recently had cause to look at a simple statistical table. All I wanted to do was to add a column showing results from the 2006 census. To do this, I had to work out the geographic boundaries that had been used to ensure that I was comparing like with like.

I could not work out how the previous numbers had been derived. I kept on coming up with different numbers. I do not know whether it is a boundary problem, or simply one due to revisions of past statistics. Frustrated, I gave up.


Time to Shine said...

In English language, sometimes I personally think it's a crazy language.

for me, this link: http://www.verbivore.com/arc_ceng.htm, is a perfect demonstration of why people get so confused with this complex language.

Jim Belshaw said...

I roared with laughter, TtoS, at the link.