[PRCo] Re: pat__service__cuts__2007.01.23-changed to 2/1/07

[Fred Schneider]

Remember too that the data from those long forms are subject to the  
biases that people insert.   I recall looking at the 1970 census data  
for commuting to work patterns in southern New Jersey ... for numbers  
by subway, by elevated, by train, by streetcar ... add them all  
together and you got PATCO's patronage which was close to 40,000 a  
day at that time.  I was surprisingly accurate ... much more so than  
income data because the people filling in the forms apparently  
thought they had no reason to skew the data.

Commuting to work data today are the primary reason people use public  
transportation in large cities.   In the United States, large cities  
are the places where people commonly use public transportation.   You  
don't normally use the subway or the bus to take your girl friend or  
wife out to dinner ... you use your car or, in New York, the taxi.    
Even in Holland, Frits van Dam was thought of as being rather unique  
(strange might have been a better word) in his neighborhood for  
driving the family to the end of the car line to go downtown in the  
evening to go shopping.

The census was only collecting the data which people wanted and that  
was data relating to the peak hours of transport and highway usage.

On Feb 5, 2007, at 6:32 PM, Joshua Dunfield wrote:

>
> John Swindler wrote:
>>
>> The national averages will be skewed towards auto assembly plants  
>> in rural
>> Ohio, where everyone must drive to get to work.  Transit is not an  
>> option.
>> That's why the term "peer group" might give a better picture.
>> Unfortunately, that will be  a lot more difficult to develope.   
>> National
>> average is meaningless to individual situations.
>
> Well, in this case the Pittsburgh-specific numbers aren't that  
> different.
> I have seen figures of 40%+ transit for Pittsburgh-city-resident-to- 
> Downtown,
> but make the set small enough and you can get whatever you want.
>
>> And besides, if you stop for a lottery ticket at the convenience  
>> store on
>> way home, is that one auto trip or two trips?  And how are the  
>> transit trips
>> being counted?  Person trips or boarding passengers?  Doesn't mean  
>> much in
>> Pittsburgh, but a significantly different number in grid pattern  
>> cities like
>> Chicago, Los Angeles, etc.  And do you count teachers driving to  
>> school as a
>> auto work trip, but what about the students on the school bus?   
>> Are they in
>> the transit count?  And if not, why not?  Could it be to skew the  
>> numbers
>> towards auto?
>
> The numbers I posted are from the census long form.  They ask only  
> how you get
> to work.  I don't remember the exact wording, but you only get to  
> choose one
> thing, so a bus trip with 0, 1, 2, 3 transfers is one person  
> checking "bus",
> and a car trip with five stops is still one person checking "car".   
> Kids on
> a school bus aren't going to "work", at least, no one calls it that.
>
> -j.
>