FROM ELAINE MURAKAMI (FHWA), and NANDA SRINIVASAN
Attached is some analysis and data based on the 2000 Census data for 49 large MSAs (Population > 1 million). The national average for travel time was 25.5 minutes, up about 3 minutes from 22.4 minutes in 1990.
Nationally, drove alone increased from 73.2 to 75.7. Transit declined from 5.3% to 4.7%. Carpooling declined from 13.4 (1990) to 12.2 (2000).
Thank you!
Nanda Srinivasan
Yes, carpooling went down drastically from 1980-90 and obviously not as
dramatic from 1990-2000. We had over 23% JTW trips by carpool in 1980
Nationwide! 1990 it was down to just over 13% (Nationwide).
I would argue that given the economic condition in 1990 (beaten down)
vs. 2000 (just when the .com bubble was to burst but still upbeat) -
losses in carpooling and transit are not as significant.
Comparing 1995 NPTS and Census 2000 I have the following to offer:
* Trip chaining (the part that goes with JTW) is up during 1990s
(NPTS)
* Huge buy-in in favor of 'flex schedules' during 1990s has
essentially marginalized the significance of JTW (there was a drop of
more than 5% from 1990 to 1995 in peak hour trip starts in our case)
* Each trip taken by transit would have a front end and a back
end trip - (park & ride or ride & walk) - Census asks for only one mode
that covered most of the distance (is comparable to 1990) thereby
undercounting all other trips;
* Vehicle occupancy rate for HBW is a little bit different from
what JTW indicates for the same reason as above - our HBW VOR is less
than what census shows (Tulsa)
Another factor - I have noticed with NPTS is - Women as a percent of
peak hour commuting public is higher than for men (13.6% men vs. 19.7%
women in 1995 for Tulsa). It could be because women tend to keep more
regular hours than men. May be 'Rideshare' programs should focus on
Women-only carpools as a potential market share.
Whatever it may be, we might notice with CTPP an increase in share of
women in commuting during peak hours - to somewhat contributing to the
erosion in transit patronage, decline in carpooling and increase in
commute times.
Viplav Putta
INCOG
I fully agree that the largest MPOs due to their large and heterogeneous
populations and the distribution of this population over a large area have
often the greatest difficulties in collecting good data. The data
collection process is by far the most difficult and challenging aspect of
the transportation modeling process and certainly not the most attractive
part to work on for modelers.
Tony Van Haagen
Caltrans,
Los Angeles
Tony Van Haagen
Sam.Granato(a)dot.state.oh.us(a)chrispy.net on 06/04/2002 06:29:04 AM
Sent by: owner-ctpp-news(a)chrispy.net
To: Ed Herlihy <ed.herlihy(a)comcast.net>
cc: ctpp-news(a)chrispy.net, owner-ctpp-news(a)chrispy.net
Subject: Re: [CTPP] Census JTW
Mr. Herlihy,
Why pick on the smaller MPOs? From what I've seen around the country, the
most "invalid" models are predominantly in the largest metro areas (with
the worst one - in terms of documented traffic assignment error - from the
agency in YOUR metro area). Where's the proof that the extensive data
collection done in the largest MPOs actually helps?
Sam Granato
Ohio DOT, Office of Technical Services
1980 W. Broad Street, Columbus, OH 43223
Phone: 614-644-6796, Fax: 614-752-8646
"We don't see things as they are, we see them as we are." Anais Nin
Ed Herlihy
<ed.herlihy(a)comcast.net To:
> ctpp-news(a)chrispy.net
Sent by: cc:
owner-ctpp-news(a)chrispy Subject:
.net Re: [CTPP] Census JTW
06/04/02 06:35 AM
Folks:
As a transportation modeler and one who is constantly looking for good data
that will help us better calibrate and validate our forecasting procedures,
the early review of the CTPP macro data and the trends are of course
interesting and an exciting preview of what is about to come.
What I would hope, however, is that before jumping to lots of conclusions
about what has happened and what is most likely to happen in the future, we
would continue to organize and plan for the release and use of the data at
the TAZ level in our MSA's and provide guidance (and support) to those who
maintain the models in these areas on how to use the data to re-validate
(and improve) the local forecasting models, especially in smaller MSA's.
I know that there are several superb analysts on this list (and the TMIP
list too) and that for many of the larger MSA's, that such a detailed
validation plan and process is in place and ready to go. I wonder, in some
cases, for some of the smaller or new UA.s and even some of these newer
UC's, that when the CTPP detailed data is finally released, whether the
time and resources will be spent to look at the data at the community and
also the corridor/TAZ level and then to see if the existing forecast models
that are up and running need fine tuning, and whether they are
demonstrating consistency (validity) in the forecast mode for the Year
2000.
Also, as we all are seeing very clearly, the non-work travel purpose
continues to have an increasing impact on our system needs.
So, I am also curious to know if there is has been a defined process in the
smaller jurisdictions to improve data collection efforts for these non work
travel modes, as well. It is my hope that with the raft of local
governmental fiscal constraints that such important and vital data
collection efforts were not cut back too much, if at all possible.
Ed Herlihy
Transportation Consultant
Reston VA.
----- Original Message -----
From: Alan E. Pisarski
To: Putta, Viplava ; ctpp-news(a)chrispy.net
Sent: Monday, June 03, 2002 4:58 PM
Subject: Re: [CTPP] Census JTW
There is a chart in CIA II that shows that women lag about a 1/2 later than
men nationwide on average. It will be interesting to see what has happened
to that distrib since. AEP
----- Original Message -----
From: Putta, Viplava
To: ctpp-news(a)chrispy.net
Sent: Monday, June 03, 2002 4:26 PM
Subject: [CTPP] Census JTW
Yes, carpooling went down drastically from 1980-90 and obviously not as
dramatic from 1990-2000. We had over 23% JTW trips by carpool in 1980
Nationwide! 1990 it was down to just over 13% (Nationwide).
I would argue that given the economic condition in 1990 (beaten down) vs.
2000 (just when the .com bubble was to burst but still upbeat) - losses in
carpooling and transit are not as significant.
Comparing 1995 NPTS and Census 2000 I have the following to offer:
· Trip chaining (the part that goes with JTW) is up during 1990s
(NPTS)
· Huge buy-in in favor of 'flex schedules' during 1990s has
essentially marginalized the significance of JTW (there was a drop of more
than 5% from 1990 to 1995 in peak hour trip starts in our case)
· Each trip taken by transit would have a front end and a back end
trip ? (park & ride or ride & walk) ? Census asks for only one mode that
covered most of the distance (is comparable to 1990) thereby undercounting
all other trips;
· Vehicle occupancy rate for HBW is a little bit different from what
JTW indicates for the same reason as above ? our HBW VOR is less than what
census shows (Tulsa)
Another factor ? I have noticed with NPTS is ? Women as a percent of peak
hour commuting public is higher than for men (13.6% men vs. 19.7% women in
1995 for Tulsa). It could be because women tend to keep more regular hours
than men. May be 'Rideshare' programs should focus on Women-only carpools
as a potential market share.
Whatever it may be, we might notice with CTPP an increase in share of women
in commuting during peak hours ? to somewhat contributing to the erosion in
transit patronage, decline in carpooling and increase in commute times.
Viplav Putta
INCOG
Mr. Herlihy,
Why pick on the smaller MPOs? From what I've seen around the country, the
most "invalid" models are predominantly in the largest metro areas (with
the worst one - in terms of documented traffic assignment error - from the
agency in YOUR metro area). Where's the proof that the extensive data
collection done in the largest MPOs actually helps?
Sam Granato
Ohio DOT, Office of Technical Services
1980 W. Broad Street, Columbus, OH 43223
Phone: 614-644-6796, Fax: 614-752-8646
"We don't see things as they are, we see them as we are." Anais Nin
Ed Herlihy <ed.herlihy(a)comcast.net>
Sent by: owner-ctpp-news(a)chrispy.net
06/04/02 06:35 AM
To: ctpp-news(a)chrispy.net
cc:
Subject: Re: [CTPP] Census JTW
Folks:
As a transportation modeler and one who is constantly looking for good
data that will help us better calibrate and validate our forecasting
procedures, the early review of the CTPP macro data and the trends are of
course interesting and an exciting preview of what is about to come.
What I would hope, however, is that before jumping to lots of conclusions
about what has happened and what is most likely to happen in the future,
we would continue to organize and plan for the release and use of the data
at the TAZ level in our MSA's and provide guidance (and support) to those
who maintain the models in these areas on how to use the data to
re-validate (and improve) the local forecasting models, especially in
smaller MSA's.
I know that there are several superb analysts on this list (and the TMIP
list too) and that for many of the larger MSA's, that such a detailed
validation plan and process is in place and ready to go. I wonder, in
some cases, for some of the smaller or new UA.s and even some of these
newer UC's, that when the CTPP detailed data is finally released, whether
the time and resources will be spent to look at the data at the community
and also the corridor/TAZ level and then to see if the existing forecast
models that are up and running need fine tuning, and whether they are
demonstrating consistency (validity) in the forecast mode for the Year
2000.
Also, as we all are seeing very clearly, the non-work travel purpose
continues to have an increasing impact on our system needs.
So, I am also curious to know if there is has been a defined process in
the smaller jurisdictions to improve data collection efforts for these non
work travel modes, as well. It is my hope that with the raft of local
governmental fiscal constraints that such important and vital data
collection efforts were not cut back too much, if at all possible.
Ed Herlihy
Transportation Consultant
Reston VA.
----- Original Message -----
From: Alan E. Pisarski
To: Putta, Viplava ; ctpp-news(a)chrispy.net
Sent: Monday, June 03, 2002 4:58 PM
Subject: Re: [CTPP] Census JTW
There is a chart in CIA II that shows that women lag about a 1/2 later
than men nationwide on average. It will be interesting to see what has
happened to that distrib since. AEP
----- Original Message -----
From: Putta, Viplava
To: ctpp-news(a)chrispy.net
Sent: Monday, June 03, 2002 4:26 PM
Subject: [CTPP] Census JTW
Yes, carpooling went down drastically from 1980-90 and obviously not as
dramatic from 1990-2000. We had over 23% JTW trips by carpool in 1980
Nationwide! 1990 it was down to just over 13% (Nationwide).
I would argue that given the economic condition in 1990 (beaten down) vs.
2000 (just when the .com bubble was to burst but still upbeat) - losses in
carpooling and transit are not as significant.
Comparing 1995 NPTS and Census 2000 I have the following to offer:
· Trip chaining (the part that goes with JTW) is up during 1990s (NPTS)
· Huge buy-in in favor of 'flex schedules' during 1990s has essentially
marginalized the significance of JTW (there was a drop of more than 5%
from 1990 to 1995 in peak hour trip starts in our case)
· Each trip taken by transit would have a front end and a back end trip ?
(park & ride or ride & walk) ? Census asks for only one mode that covered
most of the distance (is comparable to 1990) thereby undercounting all
other trips;
· Vehicle occupancy rate for HBW is a little bit different from what JTW
indicates for the same reason as above ? our HBW VOR is less than what
census shows (Tulsa)
Another factor ? I have noticed with NPTS is ? Women as a percent of peak
hour commuting public is higher than for men (13.6% men vs. 19.7% women in
1995 for Tulsa). It could be because women tend to keep more regular
hours than men. May be 'Rideshare' programs should focus on Women-only
carpools as a potential market share.
Whatever it may be, we might notice with CTPP an increase in share of
women in commuting during peak hours ? to somewhat contributing to the
erosion in transit patronage, decline in carpooling and increase in
commute times.
Viplav Putta
INCOG
One item that we can't see right now from the Census profiles is
differences by gender in the JTW. Historically women have had shorter
distance and time commutes than men. Women were more likely to take
jobs closer to home and took care of children and other household
responsiblities. I am wondering if some of the increase in overall
travel times might be an indication that women are making employment
location choices more similarly to men. We'll have to wait for CTPP
to do this analysis!
I'm surprised at you Alan. I'd certainly never bet against Michigan.
--Phil
"Alan E.
Pisarski" To: "Chuck Purvis" <CPurvis(a)mtc.ca.gov>,
<PISARSKI(a)ix.netc <ctpp-news(a)chrispy.net>
om.com> cc:
Sent by: Subject: Re: [CTPP] Michigan #1 in Drive Alone
owner-ctpp-news(a)c Share
hrispy.net
06/03/2002 02:22
PM
THE SMART MONIES ON WYOMING. AEP
----- Original Message -----
From: Chuck Purvis <CPurvis(a)mtc.ca.gov>
To: <ctpp-news(a)chrispy.net>
Sent: Monday, June 03, 2002 1:06 PM
Subject: [CTPP] Michigan #1 in Drive Alone Share
> TO: CTPP-News
>
> It looks like Michigan is still the number one state in the union in
terms
of share of commuters driving alone to work. Data released this morning
(6/3/02) shows that the Michigan drive alone share is 83.2 percent
(compares
to 81.5 percent in 1990.) Today's data release also shows Ohio with a very
high drive alone share, at 82.8 percent (comparing to 80.3 percent in
1990).
I don't think any states have drive alone shares higher than Michigan or
Ohio, but we're still missing data for five states (Arkansas, Colorado,
Utah, Idaho, Wyoming). That data will be out about 1:00 PM Eastern time on
Tuesday, along with US totals.
>
> Very good article on increasing commute times in this past Friday's
Christian Science Monitor:
>
> http://www.csmonitor.com/2002/0531/p01s01-ussc.html
>
> Alan Pisarski, Tim Lomax and the usual suspects are interviewed.
Interesting to note that the Monitor didn't fall into the trap of reporting
on anecdotal commutes. I'll be interested in how the Wall Street Journal
and
the Monitor and other national papers handle this information once the
balance of data is released Tuesday....
>
> Alan: what's this about a 12 percent drop in Virginia carpooling?
>
> Re-cap of other relevant Census sites:
>
> Press release site for demographic profiles:
> http://www.census.gov/Press-Release/www/2002/demoprofiles.html
>
> Best gateway to acquire demographic profiles in PDF format:
> http://censtats.census.gov/pub/Profiles.shtml
>
> FTP site for the Demographic Profile datasets:
> http://www2.census.gov/census_2000/datasets/100_and_sample_profile/
>
> (I'll be looking at the "All States" sub-directory to have all national,
state, metro area and place-level data perhaps by Tuesday?)
>
> Metropolitan Data Shack (site with selected 1990 Census data for all 284
metro areas and 50 states)
> http://home.earthlink.net/~clpurvis/metrodat/
>
> Chuck Purvis, MTC
> 27-28.
>
The Christian Science Monitor article (http://www.csmonitor.com/2002/0531/p01s01-ussc.html) included:
"The change [increase in commute time], while it may stem in part from new data-gathering methods, is significant."
Is the change in the range for long commutes the only change that was made in "data-gathering methods"?
(What was the change in the range for long commutes (someone said the 1990 high-end range was 99+ minutes)?
Since the 1990 long form simply asked how long, in minutes, is the commute (ie no ranges were included on the form),
why didn't the Bureau just add up the total travel time and divide by the number of respondents?)
Considering the fact that the total population of North Dakota did not change in the 1990's, I doubt that the actual rise in average commuting time was any where near the reported 22%.
Rob
Robert B. Case, PE, PTOE
Principal Transportation Engineer
Hampton Roads Planning District Commission
723 Woodlake Dr., Chesapeake, Va. 23320
voice:757-420-8300; fax:757-523-4881
rcase(a)hrpdc.org
MSA data: 1990-2000 comparison : Geographic differences in MSA definition. Elaine Murakami, FHWA and Nandu Srinivasan, CSI
June 3, 2002
MSAs and PMSAs and CMSAs are defined by the Census Bureau for statistical reporting purposes. For information on definitions, please see http://www.census.gov/population/www/estimates/metroarea.html
MSAs used in this analysis use June 1999 definition. All the Census 2000 data will use this geographic definition. For a comparison between the 1990 and 2000 definition, please see the attached excel file (geography_comparison.xls).
These changes in definitions are why some of the numbers for 1990 differ from those published in the "Journey to Work Trends" report published by FHWA in 1993.
Average Travel Time to Work
Historically, the New York City metropolitan area has had the longest average travel time to work. This remains true, with the average travel time of 34.0 minutes.
In 1990, the MSAs with the longest travel times were: New York City (30.0 minutes); Washington, DC (28.2 minutes), and Chicago (27.9 minutes).
In 2000, the MSAs with the longest travel time after NYC are: Washington, DC (31.7 minutes); Atlanta (31.2 minutes), and Chicago (31.0 minutes). Workers in Atlanta reported the largest increases since 1990, with an increase of 5.2 minutes.
Means of Transportation to Work
In most of these MSAs, there was a small increase in the proportion of workers driving alone, and a corresponding decrease in carpooling. Transit shares, in general, showed slight declines.
The largest increase in the proportion of workers using transit for their journey-to-work was in the Las Vegas, NV MSA, where the transit share increased from 2.0 percent to 4.1 percent. Considering the very large increase in total population, the number of workers using transit in the Las Vegas MSA increased from 8,150 to 28,526.
The number of workers using transit in MSAs with increases in the share using transit include: Boston, MA (from 237,463 to 261,862 workers); Portland, OR (from 41,023 to 63,126 workers); and Seattle, WA (from 91,391 to 119,919 workers).
Vehicle Availability
In general, the proportion of households without any vehicles in declining, and the proportion of households with 2 vehicles is increasing. However, in the Los Angeles MSA and the Las Vegas MSA, the proportion of households without vehicles has increased since 1990.
States not included:
The following States are not included in this analysis, even though the states had MSAs with population greater than one million:
Arkansas
Colorado
Minnesotta
Michigan
Ohio
Utah
We will re-do the spreadsheet as the data for the remaining states are released.
Enjoy!
TO: CTPP-News
It looks like Michigan is still the number one state in the union in terms of share of commuters driving alone to work. Data released this morning (6/3/02) shows that the Michigan drive alone share is 83.2 percent (compares to 81.5 percent in 1990.) Today's data release also shows Ohio with a very high drive alone share, at 82.8 percent (comparing to 80.3 percent in 1990). I don't think any states have drive alone shares higher than Michigan or Ohio, but we're still missing data for five states (Arkansas, Colorado, Utah, Idaho, Wyoming). That data will be out about 1:00 PM Eastern time on Tuesday, along with US totals.
Very good article on increasing commute times in this past Friday's Christian Science Monitor:
http://www.csmonitor.com/2002/0531/p01s01-ussc.html
Alan Pisarski, Tim Lomax and the usual suspects are interviewed. Interesting to note that the Monitor didn't fall into the trap of reporting on anecdotal commutes. I'll be interested in how the Wall Street Journal and the Monitor and other national papers handle this information once the balance of data is released Tuesday....
Alan: what's this about a 12 percent drop in Virginia carpooling?
Re-cap of other relevant Census sites:
Press release site for demographic profiles:
http://www.census.gov/Press-Release/www/2002/demoprofiles.html
Best gateway to acquire demographic profiles in PDF format:
http://censtats.census.gov/pub/Profiles.shtml
FTP site for the Demographic Profile datasets:
http://www2.census.gov/census_2000/datasets/100_and_sample_profile/
(I'll be looking at the "All States" sub-directory to have all national, state, metro area and place-level data perhaps by Tuesday?)
Metropolitan Data Shack (site with selected 1990 Census data for all 284 metro areas and 50 states)
http://home.earthlink.net/~clpurvis/metrodat/
Chuck Purvis, MTC
27-28.