[sustran] Re: [NewMobilityCafe] Re: FWD:Europe: Commuting Faster in Suburbs than Cities

[Zegras, P. Christopher]

I actually also took a closer look at the urbanaudit.org data.  In fact, 
with enough digging it is possible to find the data which, apparently, 
underlie Cox's analysis.  Basically, he seems to take the city center 
and the LUZ (larger urban zone) data for travel times and then, using 
the population ratios between the two, derives an estimated suburban 
travel time (note, however, that I did not look for all of this data for 
all of the cities that Cox includes in his table; I only examined a 
select few city tables.  In fact, for the cases where the city center is 
higher than the Metropolitan area, this approach *might* underestimate 
the "city"-"suburb" difference, since we might expect number of workers 
(those commuting) per population to be lower in the suburbs than the 
city centers.

This approach might be okay to provide a broad-brush portrait of 
commuting time differences in the urban areas with data available. But, 
there are probably too many doubts to be sure, such as:
1. What kind of surveys are these data derived from? What are the 
margins of error?  Without this information we cannot know if the 
apparent differences between the two are statistically significant (this 
is not Cox' fault, per se, as the urbanaudit.org provides no readily 
accessible info on survey methodology, sample sizes, etc.).  Almost 1/2 
of the cities (27, by my count) have "city center"-"metropolitan area" 
differences of 2 minutes or less....
2. As is typically problematic to these kinds of data comparison 
efforts, there is very little information on the relevant physical 
characteristics (As Todd Litman pointed out in an earlier post, not all 
suburbs are the same) and related influences of, for example, the 
modifiable areal unit problem. Where you draw the boundaries around the 
spatial units, quite simply, affects the values of what is measured. Are 
all alleged "Suburbs (estimated)" the same? For example, what if Rome's 
"Central City" includes most of the urbanized area and the "Suburbs 
(estimated)" is basically farmland interspersed by the occasional 
bedroom community; most of the farmers commute to their field, while a 
few bedroom community dwellers drive long distances (and times to the 
city center)? This might be a highly stylized (and unrealistic) 
anecdote, but we cannot know from the available data.
3. What do we really care about?  Commuting time and center city v/s 
suburb comparisons is certainly one measure of importance, but we would 
also need to see other related factors and indicators of interest. For 
example, do the differences hold up once controlling for, say, income? 
Also, what about distances traveled or modes used, out-of-pocket costs, 
other costs, etc?  Also, while the journey to work is certainly an 
important indicator, what about all of the other trips (for these 
cities, I would imagine journey to work is at most 1/3 of total daily 
trip purposes)?

In my view, the data presented offer more in terms of raising further 
questions of interest than they do support for the kind of sweeping 
conclusions Cox would like us to believe. For example, why do almost of 
all the German "Central Cities" have travel times equivalent to (or less 
than) the Metropolitan Area Average?  Is this a result of metropolitan 
structure (or transport system or some other factor) or where the data 
boundaries were drawn (or both)?  What differences seem to characterize 
those cities with apparently very large Center City versus Metropolitan 
Area differences relative to those with very little difference (or 
indeed with lower center city travel times)?  Etc.

Cheers,

Chris Zegras

Roland Sapsford wrote:
> Hi
> 
> I have now undertaken a little research.  The EU Urban Audit project 
> home page can be found here: http://www.urbanaudit.org/   It covers a 
> wide range of indicators for 258 cities within the 27 members of the EU.
> 
> The table referenced here: 
> http://www.publicpurpose.com/ut-eu15-jtwtime2001.htm deals with a subset 
> of the Urban Audit cities. The table has three columns. The left-most 
> column is labelled "City Centre" in this table.  This is incorrect.  
> These figures are in fact the the data for cities as a whole from the 
> Urban Audit site.  They are the only travel time data that I can locate 
> on the urban audit site .
> 
> The Urban Audit site itself makes no reference to "suburban areas" or 
> "metropolitan areas" in its indicator list or elsewhere that I can 
> easily locate.  Others may correct me on this. 
> 
> One of the presentations given to the urban audit city meeting in March 
> of this year refers to Larger Urban Zones as a concept that is being 
> incorporated in the current Urban Audit based on 2004 data - this audit 
> is not yet available.  Data are also available for some indicators (not 
> travel time) for sub-city districts.
> 
> There is no published methodology on the the publicpurpose website for 
> how the two right hand columns of figures in their table were 
> calculated, or indeed to what they refer.  No definitions or data 
> sources are provided.
> 
> These figures have been circulated in the name of Wendell Cox; I would 
> invite him to provide some evidence to support his claims other that 
> simply circulating a table which has no obvious provenance in the Urban 
> Audit project itself.
> 
> Roland Sapsford
> Wellington, New Zealand
> 
> 
> 
> 
> 
> 
> 
> 
> 
>> Subject: Europe: Commuting Faster in Suburbs than Cities
>>
>> Commentary:
>> http://www.fromtheheartland.org/blog/2006/09/europe_commuting_faster_in_sub.
>> php#more
>> Data:http://www.publicpurpose.com/ut-eu15-jtwtime2001.htm
>>  
>> Europe: Commuting Faster in Suburbs than Cities
>>
>> <snip>
>>   
> 
> 
> 
> 
> 
> 
> Check in here via the homepage at http://www.newmobility.org   
> To post message to group: NewMobilityCafe at yahoogroups.com 
> Please think twice before posting to the group as a whole
> (It might be that your note is best sent to one person?)
> 
> 
> 
>  
> Yahoo! Groups Links
> 
> <*> To visit your group on the web, go to:
>     http://groups.yahoo.com/group/NewMobilityCafe/
> 
> <*> Your email settings:
>     Individual Email | Traditional
> 
> <*> To change settings online go to:
>     http://groups.yahoo.com/group/NewMobilityCafe/join
>     (Yahoo! ID required)
> 
> <*> To change settings via email:
>     mailto:NewMobilityCafe-digest at yahoogroups.com 
>     mailto:NewMobilityCafe-fullfeatured at yahoogroups.com
> 
> <*> To unsubscribe from this group, send an email to:
>     NewMobilityCafe-unsubscribe at yahoogroups.com
> 
> <*> Your use of Yahoo! Groups is subject to:
>     http://docs.yahoo.com/info/terms/
>  
> 
> 
>