I've been a quiet subscriber of the list for a few months now and had some Chicago data that might be useful for comparison purposes. While I have been primarily engaged in CTA market research for the last 14+ years, my job here is now changing so that Census related issues are much more important and so I'll probably be contributing from time to time - more often with questions than with data I'm afraid.

1) It is not necessarily the case that the share of Transit Dependant Riders has grown in every market - in Chicago, our best source of data indicates that CTA's share of Transit Dependent Customers has fallen from 49% in 1997, to 32% in 2000, that is, 68% of CTA customers can now be considered choice customers (Source: CTA Technical Report MR01-09, Traveler Behavior and Attitudes Survey, CTA Riders and Non-Riders, July, 2001. Note: this was an RDD telephone travel survey of 2,768 residents (Riders and non-riders) in the CTA Service Area the fourth in a series conducted in 1988, 1990, 1993 and 2000).

2) We have seen a change in the mix of frequent (5+days per week) vs. infrequent (1-4 days per week) customers among our regular customers. From 1997 - 2002 the share of Infrequent Customers grew from 42% to 49% of those who rode at least once per week (Source: Technical Report MR-02-05, 2001 CTA Customer Satisfaction Survey, April 2002. Note: those who rode CTA less than once a week were not included in this sample. N= 2,505, Stratified RDD biennial telephone survey.)

3) I believe that a mix of factors including an improved automated fare system which allows many more fare options(including better counting and tracking of fare types), customer satisfaction focused service improvements, service improvements (rail line improvements, limited stop express bus services and employer based bus services) and transit marketing programs (New Residents Program, Transit Benefits Program, UPASS) designed to replace customers lost to turnover and develop new riding habits, are responsible for reversing decades of ridership loss and causing substantial ridership growth between 1997 and 2001 (8.5%). A "good chunk" of this growth has been in off-peak discretionary riding and/or riding by occasional customers.

a) Note that CTA's adoption of its Transit Card automated fare system resulted in many beneficial impacts too many to insert here - the share of uncounted customers has certainly improved as have revenues - many more types of fares can be offered and validated by the system (relieving the operator of the need to do so in many cases - much better tracking of fare card usage by type - especially for daily, weekly or monthly passes is now available - customers accept the accuracy of electronically time stamped transfers and argue less with operators - and most importantly - pass users of many kinds now make more discretionary off-peak trips for other purposes resulting in an increase in infrequent choice customers on our system.

a) By being taken only once every 10 years (as prescribed by law) interim trends of growth and loss are not observed. How tragic it would be if CTA (and other agencies like it), after finally reversing the trend of decades of ridership loss would lose some % of capital funding eligibility over the next decade that would have allow it to continue to grow and rebuild its market share.

c) It seems to me that these weaknesses, and other weaknesses that those more expert on Census Data than I could provide, suggest that better, more frequent national sampling of travel data and household auto/transit mode shares are needed than can be provided by the census. Whether the cost of doing so nationwide, with sample sizes large enough that data that would be useful on regional and local levels is worth the benefits is arguable. Clearly I would favor such a project. In saying this I am not attempting to put down the Census - just point out that it may be that there may be a need to push for other more frequent data sources using sampling methods rather than a full count and the long/short form.

5) Responding to Ed Christopher's point, our TBAS surveys referenced above indicate that while the share of trips to or from work in the CTA service area made by non-CTA Customers has been relatively stable from 1990 (40%) to 2000 (39%), work trips grew significantly among Frequent CTA Customers (44% in 1990 to 55% in 2001) and Infrequent CTA Customers (29% in 1990 to 39% in 2000, 2000 Nw=5,769 trips). (Note that a PDF file of this report is available on request.) This suggests growth among work Trip use over the decade by Transit Customers but no change for the general population. Note also that results for CTA are likely to be very different than for non-urban systems, only 8% of households in our service area say they have never used CTA.)

****************************************************************************
****************************************************************************
*******************************************
Unless otherwise stated, the views expressed are those of the author and not
those of the Chicago Transit Authority.
****************************************************************************
****************************************************************************
*******************************************
http://www.TransitChicago.com

RTA Travel Information
836-7000

----- Original Message -----
From: "Gardner, John F" <GardnerJF@dot.state.sc.us>
To: <ctpp-news@chrispy.net>
Sent: Monday, July 08, 2002 1:58 PM
Subject: RE: [CTPP] Reconciling Census Transit Commuters with Ridership Statistics

> Additional possible reasons for the APTA/FTA reported growth in total
> ridership vs. unchanged census count of transit commuters:
>
> 1) Because transit is losing mode share, one assumes the remaining users
> are more transit dependent than in 1990 and therefore more likely to use
> transit for ALL trips, not just work trips.
>
> 2) Aging of the population results in long-term riders continuing to ride,
> but no longer making work trips (retired).
>
> 3) ADA accessibility requirements have made transit more accessible to
> disabled persons, whose riding pattern may include a smaller than average
> proportion of work trips.
>
> 4) Most transit systems now use electronic fareboxes that improve passenger
> count reliability; fewer systems had these in 1990. Some undercounting of
> passengers may have occurred previously -- drivers had to punch mechanical
> counters, or even less reliable methods were used. Some of the reported
> growth in total ridership may result from more accurate passenger counts.
>
>
> -----Original Message-----
> From: ed christopher [mailto:edc@berwyned.com]
> Sent: Monday, July 08, 2002 1:33 PM
> Cc: ctpp-news@chrispy.net
> Subject: Re: [CTPP] Reconciling Census Transit Commuters with Ridership
> Statistics
>
>
> As for Chucks comments I believe that he was actually searching to gain a
> better understanding into the trip purpose side of mode
> split issue. Is the the work trip declining in its mode share? The answer
> of course in not in any census data.
>