The Age is running a series this week on the liveability of Melbourne. Its league table of suburbs ranked by liveability might be of doubtful utility (as I wrote last time) but the space it’s devoting to discussing urban issues is valuable e.g. see Liveability and a tale of two Melbournes by Daniel Terrill and Adam Terrill.

One of the offerings is a community forum titled Fixing Melbourne’s problems; what would you do? Readers are invited to “share and debate ideas about making Melbourne a better place in which to live and work” via the comments section.

It’s a great idea and as of this morning there are 216 comments. But it got off to a bad start when the host of the forum, journalist Michael Short, posted a graph comparing  median property prices against liveability for each of Melbourne’s 321 suburbs (see exhibit). (1)

If you click through the exhibit and hover your mouse over any point, it will show you the liveability ranking and median selling price for a house in that suburb.

The x axis shows price; it indicates the highest ranking suburb for liveability, inner city East Melbourne, has a median price of \$1.72 million.

That’s around four times higher than the \$0.42 million median price in the least liveable suburb, Skye, on Melbourne’s outer south eastern fringe.

So far so good; but the y axis, which shows liveability, indicates that East Melbourne is 321 times as liveable as Skye. 321 times! Obviously that’s an absurd result; if it were true the median price in Skye should be around \$10,000.

This silliness is because the graph measures liveability by the cardinal rank order of each suburb i.e. from 1 to 321.  The necessary consequence is that the “distance” between any two suburbs in terms of their liveability is the same i.e. one.

The appropriate method would’ve been to graph a summary liveability score (derived from the 15 criteria used by The Age) for each suburb against price. That would inevitably show the differences in liveability between most suburbs, like differences in price, are extremely small.

Consider the index The Economist Intelligence Unit uses for its ranking of 140 world cities according to their liveability. The difference between the leading cities is miniscule; top-ranked Melbourne scored 97.5 out of 100 and tenth-ranked Auckland scored 95.7.

Melbourne’s score is only two and a half times that of bottom-ranked Dhaka (38.7), even though the latter is riven with crime and grinding poverty. The difference in real living conditions between these two cities is a lot more dramatic than that between East Melbourne and Skye.

The graph reinforces my doubts about the usefulness of The Age’s liveability league table. It strengthens my contention that the paper should make the report it relies on available to its readers.

And it emphasises that while it’s a great thing for the media to promote discussion around key public policy issues, it has a responsibility to promote informed debate.

It should ensure the information put out there under its name is well-founded, salient and useful.

The Age might care to take a look at a study released this month by The Centre for London, Urban demographics: why people live where they do, for ideas on how to define and analyse suburban liveability.

In this commentary on the Centre’s study, renowned urbanist Richard Florida says:

Not surprisingly, the key things that matter to people about the neighborhoods they live in include a mix of housing costs, being close to family, and proximity to where they work. More than a quarter (28 percent) of respondents cited housing costs and proximity to friends as key factors in the neighborhoods where they live, followed by the size and type of available housing (22 percent), and proximity to their workplace or their partner’s workplace (21 percent).

As I noted last time, The Age’s definition of suburban liveability doesn’t take account of proximity to work or the size of housing. Nor does it measure closeness to family and friends.

Update 12/11/15: This is about as close as Fairfax is likely to get to acknowledging it got its league table wrong, Melbourne’s best suburb? Predictably nobody can agree.

______________