New subject: [CTPP] putting forth a cohesive argument for Census

19 May 2013

I really feel we do bear some of the burden of proof here.  We do a terrible
job of making the case for data and for the big stuff in particular - so
called general purpose data are soft in their clear applications - being
implicit doesn't do it.  

Part of the issues is just grandstanding on the part of politicians, but a
big part of it is we don't tell them what we do with the stuff.  Have you
ever had a cocktail party conversation about, say,  "why do you need to know
what time I leave for work?"  

We are in a societal situation today where:

a.)    There is great fear of govt snooping and a real lack of trust

b.)    A massive amount of big data observation

c.)     AND, the contrary crazy part, of some who will divulge just about
everything on their facebook etc. site.

d.)    And don't forget the Bureau's unwillingness to publish much of the
data they collect because of disclosure.

Trust me with your money and your personal data just won't work anymore.  We
have to do our homework and do a far better job of making the case for the
growing govt need for more and more data growing out of new mandates.   Alan

PS  the Bureau is good but not close to being best or leader in World.
Stats Canada,  Australian Bureau of Statistics,  Netherlands Bureau of
Statistics are the real leaders and a bunch more.  

From: ctpp-news-bounces(a)chrispy.net [mailto:ctpp-news-bounces(a)chrispy.net]
On Behalf Of Ken.Cervenka(a)dot.gov
Sent: Friday, May 17, 2013 5:37 PM
To: ctpp-news(a)chrispy.net
Subject: Re: [CTPP] The Case for ACS, the Economic Census, sampling and
Federal data-gathering

Hi Jonathan,

Some excellent observations, thanks for sharing.

An example of one of the underlying strengths of CTPP is that even though
CTPP 2000 flow data is now 13 years old, the journey-to-work data still
represents the "best available" data across the country, for use in
home-based work regional travel model validation checks.  I am not referring
to simple HBW trip length frequency checks (for which much smaller samples,
e.g., the samples available from typical household surveys), I am talking
about district-to-district checks of predicted versus observed HBW person
trips by (let's say) 0-vehicle, 1-vehicle and 2+ vehicle socioeconomic
groups.  Yes, the "observed" data is getting old, but can be
"fratared" with
the growth rates in population and employment to be a reasonably good
representation of "current year" HBW flows.  Many people think travel
forecasts are way off from reality because of problems with the mode choice
model and/or traffic assignment procedures, but I wonder how often the real
"problem" can be traced back to trip (or activity) generation and the
prediction of the zone-to-zone person trip (or activity flow) tables!

The next "best available" dataset of home-to-work commuter trips is going to
be the next available "five-year" CTPP, that is now very close to a public
release.  There will of course be concerns that the overall number of raw
records in the new "five-year" CTPP is a lot less than what was obtained in
CTPP 2000, but all that really means, I think, is that one must be very
careful to never rely on the actual TAZ-to-TAZ numbers:  but still feel very
confident with the prudent aggregation of TAZ-to-TAZ numbers into locally
meaningful district-to-district totals.  In regards to the national sample,
the new CTPP will still be a pretty huge number of records, compared to just
about anything else that is not some type of "private" purchasable database.
It would be a shame if the about-to-be-released set of CTPP-based TAZ-to-TAZ
flows is the last one ever produced.  If the NHTS sample size could come
anywhere close to the CTPP sample size, then maybe one wouldn't need the
CTPP, but that is not going to ever happen.

Ken Cervenka

FTA Office of Planning and Environment

From: ctpp-news-bounces(a)chrispy.net [mailto:ctpp-news-bounces(a)chrispy.net]
On Behalf Of Jonathan Lupton
Sent: Friday, May 17, 2013 3:36 PM
To: ctpp-news(a)chrispy.net
Subject: [CTPP] The Case for ACS, the Economic Census, sampling and Federal
data-gathering

The loss of Census sampling (today the ACS), the Census of Agriculture, the
Economic Census, etc., would be disastrous in so many ways I find it hard to
believe anyone - even deeply partisan politicians - would advocate their
demise. The small gain to the Federal government's bottom line would be
undercut by huge (albeit hidden) losses to the private and public sectors.
The loss would be especially acute for small businesses. 

1.       There can be no serious doubt that Federal sample products leverage
their cost many times over in benefits to the economy. The benefits are so
widespread, and so implicit, that the burden of proof must lie on anyone
attempting to undo Federal data-gathering. And they will find no such proof.

2.       The U.S. government has, in the past, set the world standard for
data-gathering. The widespread availability of free, accurate data runs
hand-in-hand with upholding the standard as the world's foremost democratic
society. To surrender the ACS and related products is not just a bad idea,
it is a retreat from leadership.

3.       Answering ACS forms, or any other Federal questionnaire, is a
matter of personal responsibility. To survive, democracy depends not just on
the protection of personal rights; it also demands a sense of responsibility
by its citizens.

4.       I have never heard of anyone going to prison, or even being fined,
for failing to provide data to Census takers. Everyone knows that there are
people and businesses which refuse to cooperate; the practice of
non-compliance is already tolerated. But compliance is the law, and this
sets a tone of legality which allows the ACS and other projects to gather
the necessary data.

5.       If the data business becomes mostly private in nature, the cost of
obtaining data will largely limit its availability to large corporations
that could afford to purchase it, creating another disadvantage to small
businesses and business start-ups.

6.       Here in Little Rock we host one of the country's largest
data-gathering agencies, the Acxiom Corporation.  It's an open secret that
Acxiom, and other companies like it, hold vast amounts of data about just
about everybody. While Census data is protected by confidentiality laws,
disturbingly intimate corporate data can be sold to the highest bidder. 

7.       While the anti-census anti-government lobby argues unconvincingly
about government as "Big Brother," there is therefore another, less
accountable version of "Big Brother," existing in secret corporate
data-gathering. Such data could become the only basis for information about
our society. Without Federal laws, and Executive and Congressional
oversight, who could prevent this private data from being falsified? Without
the credibility of ACS and related programs as a "cross-check," false
information could be fed into the system, and could be manipulated by
private power brokers.

8.       Here in Little Rock we have a small spinoff company which has used
Acxiom data to attempt census-like products. Around 2009, they privately
gave me a total for the state's largest county (Pulaski) that disagreed with
my careful estimates. They ended up being high by about 7 percent, compared
with the Census 2010 count that appeared a few months later. My own
estimate, based on housing records, was within 1 or 2 percent. A corporate
representative thought their figures were inarguably correct; I thought
their methods for counting people were flawed. Guess who was right.

9.       Data-gathering by the Census Bureau and related agencies isn't
perfect, but it has oversight through the democratic process. I'll trust a
process that's been around since 1790 before I trust a private company that
answers first to shareholders.

It is my earnest hope that the effort to kill the ACS is so blatantly
foolish that it will never make it to a vote by the U.S. Congress or Senate.
I ask those who keep their ear to Congress to please keep the data community
well-informed about this disturbing development.

Jonathan Lupton AICP

Research Planner

Metroplan

Little Rock, Arkansas

501-372-3300 

__________ Information from ESET Smart Security, version of virus signature
database 8345 (20130517) __________

The message was checked by ESET Smart Security.

http://www.eset.com

__________ Information from ESET Smart Security, version of virus signature
database 8350 (20130519) __________

The message was checked by ESET Smart Security.

http://www.eset.com

Re: [CTPP] The Case for ACS, the Economic Census, sampling and Federal data-gathering