Liang,
Do you know when the ACS will start using these new MSA delineations in the data reported?
Thank you,
Nancy
Nancy Reger | nreger(a)morpc.org
Deputy Director, Transportation | Mid-Ohio Regional Planning Commission
T: 614.233.4154 | | F: 614.233.4254
111 Liberty Street, Suite 100 | Columbus, OH 43215
[cid:image001.png(a)01CE1FDC.24E7F2D0]<http://www.morpc.org/> [cid:image002.png(a)01CE1FDC.24E7F2D0] <http://www.facebook.com/morpc> [cid:image003.jpg(a)01CE1FDC.24E7F2D0] <http://www.twitter.com/morpc>
From: ctpp-news-bounces(a)chrispy.net [mailto:ctpp-news-bounces(a)chrispy.net] On Behalf Of Liang Long
Sent: Monday, March 04, 2013 11:09 AM
To: ctpp-news(a)chrispy.net
Subject: [CTPP] OMB Released tje New Metropolitan and micropolitan areas
Hi, All
Our Census Bureau friend, Brian McKenzie told us that OMB released the metro/micro areas late last week. The new definitions' link http://www.whitehouse.gov/omb/bulletins_default .
The new 2006-2010 county worker flows will come out tomorrow some time.
Enjoy!
Liang Long
Cambridge Systematics, Inc.
4800 Hampden Lane
Suite 800
Bethesda, MD 20814
tel 301 347 0100
fax 301 347 0101
FHWA 202-366-6971
e-mail llong(a)camsys.com<mailto:llong(a)camsys.com>
www.camsys.com
That's true. If we knew how the census bureau was aggregating smaller areas like blocks into counties and calculating MOE's, then that would give us an idea for how to aggregate counties into arbitrary groupings.
Steven Farber, Ph.D
Assistant Professor
Department of Geography
University or Utah
http://stevenfarber.wordpress.com
-----Original Message-----
From: ctpp-news-bounces(a)chrispy.net [mailto:ctpp-news-bounces(a)chrispy.net] On Behalf Of hprawiranata mitcrpc.org
Sent: March-12-13 10:38 AM
To: ctpp-news(a)chrispy.net
Subject: Re: [CTPP] Working with County flow data
My logic:
Measurement from single entities (one county with it sample) has its own error. Combining multiple single measurements will create sum or error from each measurements.
IF
On Tue, Mar 12, 2013 at 12:18 PM, <liang.long(a)dot.gov> wrote:
> I can see why Census doesn't recommend do more than three variables at a time. When you add 17 counties together, you get a much bigger area with more households sampled. In theory, you should get a smaller MOEs compared each individual county. But if you derive MOEs from those 17 counties, you will get a much bigger MOEs, which is contradictory to the theory.
>
>
> ________________________________________
> From: ctpp-news-bounces(a)chrispy.net [ctpp-news-bounces(a)chrispy.net] on
> behalf of Ed Christopher [edc(a)berwyned.com]
> Sent: Tuesday, March 12, 2013 11:15 AM
> To: ctpp-news(a)chrispy.net
> Subject: Re: [CTPP] Working with County flow data
>
> Thanks--I know the spread sheet allows you to recalculate MOEs for more than three variables but I remember doing more than 3 a while back and I was getting some wild MOEs. When I dug into it I found something in the Census compass reports that said not to do more than three variables at a time. I was hoping that someone figured out a way around this.
>
> Ed C
>
> On Mar 12, 2013, at 9:59 AM, "Hoctor Mulmat, Darlanne" <Darlanne.Mulmat(a)sandag.org<mailto:Darlanne.Mulmat(a)sandag.org>> wrote:
>
> The New York State Data Center developed a Statistical Calculations Menu that includes an option for computing the margin of error for the sum of three or more estimates. See attached.
>
> Darlanne Hoctor Mulmat
> Applied Research Division - Criminal Justice/Public Policy San Diego
> Association of Governments
> 619-699-7326
>
> From:
> ctpp-news-bounces(a)chrispy.net<mailto:ctpp-news-bounces(a)chrispy.net>
> [mailto:ctpp-news-bounces(a)chrispy.net] On Behalf Of
> Ed.Christopher(a)dot.gov<mailto:Ed.Christopher(a)dot.gov>
> Sent: Tuesday, March 12, 2013 6:57 AM
> To: ctpp-news(a)chrispy.net<mailto:ctpp-news(a)chrispy.net>
> Subject: [CTPP] Working with County flow data
>
> Has anyone come up with some easy ways for collapsing and grouping counties together using last week's county flow data and recalculating new MOEs. I have so many counties that I want to group together that I am looking for a quick way that can handle "lots" of counties. Another issue I am struggling with is that we are always told not to group more than three variables at a time or the formulas for calculating the new MOE do not really work. This is particularly troublesome especially if I am trying to group 17 counties together. What it comes down to is 9 different calculations given that I can only group 3 counties at a time together. Anyone figure out any short cuts or ways around this short of disregarding the MOEs altogether? Given all the clustering that I am looking at using the "cheat" sheets I am used to, I will be recalculating MOEs for weeks.
>
>
> Ed Christopher
> <StatisticalCalculationsMenu.xls>
> _______________________________________________
> ctpp-news mailing list
> ctpp-news(a)ryoko.chrispy.net<mailto:ctpp-news(a)ryoko.chrispy.net>
> http://ryoko.chrispy.net/mailman/listinfo/ctpp-news
>
> _______________________________________________
> ctpp-news mailing list
> ctpp-news(a)ryoko.chrispy.net
> http://ryoko.chrispy.net/mailman/listinfo/ctpp-news
_______________________________________________
ctpp-news mailing list
ctpp-news(a)ryoko.chrispy.nethttp://ryoko.chrispy.net/mailman/listinfo/ctpp-news
I see one reason why aggregation is causing larger, rather than smaller, standard errors.
The formula used, known as Bienaymé's formula, assumes that each variable you are adding is independent of the others. In a map of the counties, this is clearly not the case, since we know the counties are autocorrelated. Perhaps you can estimate the covariance between counties and use that to more accurately estimate the variance using the formula Var(sum(X_i))=sum(sum(Cov(X_i,X_j)).
This is the more appropriate way to calculate variance of a sum of random variables that are not independent.
Isn't there a statistician in the room?
Steve
Steven Farber, Ph.D
Assistant Professor
Department of Geography
University or Utah
http://stevenfarber.wordpress.com
-----Original Message-----
From: ctpp-news-bounces(a)chrispy.net [mailto:ctpp-news-bounces(a)chrispy.net] On Behalf Of liang.long(a)dot.gov
Sent: March-12-13 10:18 AM
To: ctpp-news(a)chrispy.net
Subject: Re: [CTPP] Working with County flow data
I can see why Census doesn't recommend do more than three variables at a time. When you add 17 counties together, you get a much bigger area with more households sampled. In theory, you should get a smaller MOEs compared each individual county. But if you derive MOEs from those 17 counties, you will get a much bigger MOEs, which is contradictory to the theory.
________________________________________
From: ctpp-news-bounces(a)chrispy.net [ctpp-news-bounces(a)chrispy.net] on behalf of Ed Christopher [edc(a)berwyned.com]
Sent: Tuesday, March 12, 2013 11:15 AM
To: ctpp-news(a)chrispy.net
Subject: Re: [CTPP] Working with County flow data
Thanks--I know the spread sheet allows you to recalculate MOEs for more than three variables but I remember doing more than 3 a while back and I was getting some wild MOEs. When I dug into it I found something in the Census compass reports that said not to do more than three variables at a time. I was hoping that someone figured out a way around this.
Ed C
On Mar 12, 2013, at 9:59 AM, "Hoctor Mulmat, Darlanne" <Darlanne.Mulmat(a)sandag.org<mailto:Darlanne.Mulmat(a)sandag.org>> wrote:
The New York State Data Center developed a Statistical Calculations Menu that includes an option for computing the margin of error for the sum of three or more estimates. See attached.
Darlanne Hoctor Mulmat
Applied Research Division - Criminal Justice/Public Policy San Diego Association of Governments
619-699-7326
From: ctpp-news-bounces(a)chrispy.net<mailto:ctpp-news-bounces(a)chrispy.net> [mailto:ctpp-news-bounces(a)chrispy.net] On Behalf Of Ed.Christopher(a)dot.gov<mailto:Ed.Christopher(a)dot.gov>
Sent: Tuesday, March 12, 2013 6:57 AM
To: ctpp-news(a)chrispy.net<mailto:ctpp-news(a)chrispy.net>
Subject: [CTPP] Working with County flow data
Has anyone come up with some easy ways for collapsing and grouping counties together using last week's county flow data and recalculating new MOEs. I have so many counties that I want to group together that I am looking for a quick way that can handle "lots" of counties. Another issue I am struggling with is that we are always told not to group more than three variables at a time or the formulas for calculating the new MOE do not really work. This is particularly troublesome especially if I am trying to group 17 counties together. What it comes down to is 9 different calculations given that I can only group 3 counties at a time together. Anyone figure out any short cuts or ways around this short of disregarding the MOEs altogether? Given all the clustering that I am looking at using the "cheat" sheets I am used to, I will be recalculating MOEs for weeks.
Ed Christopher
<StatisticalCalculationsMenu.xls>
_______________________________________________
ctpp-news mailing list
ctpp-news(a)ryoko.chrispy.net<mailto:ctpp-news(a)ryoko.chrispy.net>
http://ryoko.chrispy.net/mailman/listinfo/ctpp-news
_______________________________________________
ctpp-news mailing list
ctpp-news(a)ryoko.chrispy.nethttp://ryoko.chrispy.net/mailman/listinfo/ctpp-news
Sorry, unfinished email draft got sent out !
I still have to think about it and see the data..
Hary
On Tue, Mar 12, 2013 at 12:37 PM, hprawiranata mitcrpc.org
<hprawiranata(a)mitcrpc.org> wrote:
> My logic:
> Measurement from single entities (one county with it sample) has its
> own error. Combining multiple single measurements will create sum or
> error from each measurements.
>
> IF
>
>
> On Tue, Mar 12, 2013 at 12:18 PM, <liang.long(a)dot.gov> wrote:
>> I can see why Census doesn't recommend do more than three variables at a time. When you add 17 counties together, you get a much bigger area with more households sampled. In theory, you should get a smaller MOEs compared each individual county. But if you derive MOEs from those 17 counties, you will get a much bigger MOEs, which is contradictory to the theory.
>>
>>
>> ________________________________________
>> From: ctpp-news-bounces(a)chrispy.net [ctpp-news-bounces(a)chrispy.net] on behalf of Ed Christopher [edc(a)berwyned.com]
>> Sent: Tuesday, March 12, 2013 11:15 AM
>> To: ctpp-news(a)chrispy.net
>> Subject: Re: [CTPP] Working with County flow data
>>
>> Thanks--I know the spread sheet allows you to recalculate MOEs for more than three variables but I remember doing more than 3 a while back and I was getting some wild MOEs. When I dug into it I found something in the Census compass reports that said not to do more than three variables at a time. I was hoping that someone figured out a way around this.
>>
>> Ed C
>>
>> On Mar 12, 2013, at 9:59 AM, "Hoctor Mulmat, Darlanne" <Darlanne.Mulmat(a)sandag.org<mailto:Darlanne.Mulmat(a)sandag.org>> wrote:
>>
>> The New York State Data Center developed a Statistical Calculations Menu that includes an option for computing the margin of error for the sum of three or more estimates. See attached.
>>
>> Darlanne Hoctor Mulmat
>> Applied Research Division – Criminal Justice/Public Policy
>> San Diego Association of Governments
>> 619-699-7326
>>
>> From: ctpp-news-bounces(a)chrispy.net<mailto:ctpp-news-bounces(a)chrispy.net> [mailto:ctpp-news-bounces(a)chrispy.net] On Behalf Of Ed.Christopher(a)dot.gov<mailto:Ed.Christopher(a)dot.gov>
>> Sent: Tuesday, March 12, 2013 6:57 AM
>> To: ctpp-news(a)chrispy.net<mailto:ctpp-news(a)chrispy.net>
>> Subject: [CTPP] Working with County flow data
>>
>> Has anyone come up with some easy ways for collapsing and grouping counties together using last week’s county flow data and recalculating new MOEs. I have so many counties that I want to group together that I am looking for a quick way that can handle “lots” of counties. Another issue I am struggling with is that we are always told not to group more than three variables at a time or the formulas for calculating the new MOE do not really work. This is particularly troublesome especially if I am trying to group 17 counties together. What it comes down to is 9 different calculations given that I can only group 3 counties at a time together. Anyone figure out any short cuts or ways around this short of disregarding the MOEs altogether? Given all the clustering that I am looking at using the “cheat” sheets I am used to, I will be recalculating MOEs for weeks.
>>
>>
>> Ed Christopher
>> <StatisticalCalculationsMenu.xls>
>> _______________________________________________
>> ctpp-news mailing list
>> ctpp-news(a)ryoko.chrispy.net<mailto:ctpp-news(a)ryoko.chrispy.net>
>> http://ryoko.chrispy.net/mailman/listinfo/ctpp-news
>>
>> _______________________________________________
>> ctpp-news mailing list
>> ctpp-news(a)ryoko.chrispy.net
>> http://ryoko.chrispy.net/mailman/listinfo/ctpp-news
My logic:
Measurement from single entities (one county with it sample) has its
own error. Combining multiple single measurements will create sum or
error from each measurements.
IF
On Tue, Mar 12, 2013 at 12:18 PM, <liang.long(a)dot.gov> wrote:
> I can see why Census doesn't recommend do more than three variables at a time. When you add 17 counties together, you get a much bigger area with more households sampled. In theory, you should get a smaller MOEs compared each individual county. But if you derive MOEs from those 17 counties, you will get a much bigger MOEs, which is contradictory to the theory.
>
>
> ________________________________________
> From: ctpp-news-bounces(a)chrispy.net [ctpp-news-bounces(a)chrispy.net] on behalf of Ed Christopher [edc(a)berwyned.com]
> Sent: Tuesday, March 12, 2013 11:15 AM
> To: ctpp-news(a)chrispy.net
> Subject: Re: [CTPP] Working with County flow data
>
> Thanks--I know the spread sheet allows you to recalculate MOEs for more than three variables but I remember doing more than 3 a while back and I was getting some wild MOEs. When I dug into it I found something in the Census compass reports that said not to do more than three variables at a time. I was hoping that someone figured out a way around this.
>
> Ed C
>
> On Mar 12, 2013, at 9:59 AM, "Hoctor Mulmat, Darlanne" <Darlanne.Mulmat(a)sandag.org<mailto:Darlanne.Mulmat(a)sandag.org>> wrote:
>
> The New York State Data Center developed a Statistical Calculations Menu that includes an option for computing the margin of error for the sum of three or more estimates. See attached.
>
> Darlanne Hoctor Mulmat
> Applied Research Division – Criminal Justice/Public Policy
> San Diego Association of Governments
> 619-699-7326
>
> From: ctpp-news-bounces(a)chrispy.net<mailto:ctpp-news-bounces(a)chrispy.net> [mailto:ctpp-news-bounces(a)chrispy.net] On Behalf Of Ed.Christopher(a)dot.gov<mailto:Ed.Christopher(a)dot.gov>
> Sent: Tuesday, March 12, 2013 6:57 AM
> To: ctpp-news(a)chrispy.net<mailto:ctpp-news(a)chrispy.net>
> Subject: [CTPP] Working with County flow data
>
> Has anyone come up with some easy ways for collapsing and grouping counties together using last week’s county flow data and recalculating new MOEs. I have so many counties that I want to group together that I am looking for a quick way that can handle “lots” of counties. Another issue I am struggling with is that we are always told not to group more than three variables at a time or the formulas for calculating the new MOE do not really work. This is particularly troublesome especially if I am trying to group 17 counties together. What it comes down to is 9 different calculations given that I can only group 3 counties at a time together. Anyone figure out any short cuts or ways around this short of disregarding the MOEs altogether? Given all the clustering that I am looking at using the “cheat” sheets I am used to, I will be recalculating MOEs for weeks.
>
>
> Ed Christopher
> <StatisticalCalculationsMenu.xls>
> _______________________________________________
> ctpp-news mailing list
> ctpp-news(a)ryoko.chrispy.net<mailto:ctpp-news(a)ryoko.chrispy.net>
> http://ryoko.chrispy.net/mailman/listinfo/ctpp-news
>
> _______________________________________________
> ctpp-news mailing list
> ctpp-news(a)ryoko.chrispy.net
> http://ryoko.chrispy.net/mailman/listinfo/ctpp-news
The New York State Data Center developed a Statistical Calculations Menu that includes an option for computing the margin of error for the sum of three or more estimates. See attached.
Darlanne Hoctor Mulmat
Applied Research Division - Criminal Justice/Public Policy
San Diego Association of Governments
619-699-7326
From: ctpp-news-bounces(a)chrispy.net [mailto:ctpp-news-bounces(a)chrispy.net] On Behalf Of Ed.Christopher(a)dot.gov
Sent: Tuesday, March 12, 2013 6:57 AM
To: ctpp-news(a)chrispy.net
Subject: [CTPP] Working with County flow data
Has anyone come up with some easy ways for collapsing and grouping counties together using last week's county flow data and recalculating new MOEs. I have so many counties that I want to group together that I am looking for a quick way that can handle "lots" of counties. Another issue I am struggling with is that we are always told not to group more than three variables at a time or the formulas for calculating the new MOE do not really work. This is particularly troublesome especially if I am trying to group 17 counties together. What it comes down to is 9 different calculations given that I can only group 3 counties at a time together. Anyone figure out any short cuts or ways around this short of disregarding the MOEs altogether? Given all the clustering that I am looking at using the "cheat" sheets I am used to, I will be recalculating MOEs for weeks.
Ed Christopher
Has anyone come up with some easy ways for collapsing and grouping counties together using last week's county flow data and recalculating new MOEs. I have so many counties that I want to group together that I am looking for a quick way that can handle "lots" of counties. Another issue I am struggling with is that we are always told not to group more than three variables at a time or the formulas for calculating the new MOE do not really work. This is particularly troublesome especially if I am trying to group 17 counties together. What it comes down to is 9 different calculations given that I can only group 3 counties at a time together. Anyone figure out any short cuts or ways around this short of disregarding the MOEs altogether? Given all the clustering that I am looking at using the "cheat" sheets I am used to, I will be recalculating MOEs for weeks.
Ed Christopher
Just to provide some BASIC items of differences on data sources for home/work FLOWS.
ACS and CTPP based on ACS: This is a sample SURVEY of households and household members. The ACS is an address-based sample. The response rate to ACS is lower than to Census 2000 "long form" and the field follow-up for non-respondents to the mail-back form is for a SAMPLE of non-respondents. All the responses are weighted to come up with a total estimate. Respondents are asked to provide their workplace location, which are then geocoded by the Census Bureau. About 25 % of workers do not provide sufficient information to code to a census block and their workplace is imputed using a routine using 4 variables: industry, occupation, travel mode and travel speed. See the recent CTPP Status Report article by Melanie Rapino on the details of the imputation process.
LEHD: is from ADMINISTRATIVE records, based on Employment Security (unemployment), and augmented with federal employment records. The base is from files submitted to the Census Bureau from State Departments of Employment Security who maintain the QCEW files (formerly called ES-202). LEHD does not currently include self-employed, or military employment. In this case, the workplace location is the address(es) provided by businesses. If the business has multiple locations in a county, many states require that each location be reported. Only in Minnesota is the employee record tied to a specific workplace location. That is, in all other states, for businesses with multiple sites, the link between an employee's home location and the workplace location is imputed. That is, an algorithm selects among the multiple worksites which is the most likely workplace for that worker. The home location may be moved to protect individual confidentiality, but largely the moves are within the same census tract. The link between the unemployment (QCEW) and home location is primarily from IRS tax records, using an encrypted version of SSN.
Sometimes people use a different home address for IRS than where they are actually living, for example young people or college students may use their parents' address. And similarly, sometime people report on the ACS form a workplace location where they may not actually be working, for example, a construction worker might report the business address of his employer. So, there is plenty of room for inconsistencies.
Private lists of businesses and employment totals: Typically these are compiled using web resources and telephone surveys. Often the list guarantees that each business has been contacted within 3 (or 5) years to get the most current employment figures. Because these firms have been compiling these lists and checking them for so many years, they have been able to refine their databases. However, these lists do not provide FLOW between work and home, just workplace location totals.
Please let me (and the listserv) know if I have made any mistakes above.
Elaine Murakami
________________________________________
From: ctpp-news-bounces(a)chrispy.net [ctpp-news-bounces(a)chrispy.net] on behalf of hprawiranata mitcrpc.org [hprawiranata(a)mitcrpc.org]
Sent: Friday, March 08, 2013 11:49 AM
To: ctpp-news(a)chrispy.net
Subject: Re: [CTPP] Commuter Flow Data Concerns
Thanks, Ed.
I have been working LEHD data for 2 weeks and comparing it with other
sources (REMI, Claritas, WP, etc). All sources has different number
for 2010, for example, which one is the good one ? LEHD has all jobs,
primary jobs, private jobs, etc. I found the use primary job data give
close number with other sources, am I on the right track ?
I have not compared LEHD data with 'just released' commuting data set,
I guess it would be the same and LEHD has better resolution. What do
you think ?
Thanks,
Hary
On Fri, Mar 8, 2013 at 10:22 AM, Ed Christopher <edc(a)berwyned.com> wrote:
> A few of us have been getting calls and emails from people who have been
> looking at the commuter follow data the Census Bureau just released, the
> On the Map (LEHD) and some of the earlier flow data. If you have, you
> probably noticed some interesting (or maybe wierd) things. Needless to
> say a good understanding of the data is needed to sort out some of the
> issues. Below are some references and points that may be of interest.
> What are people noticing with the data?
> ---------------
> In the January 20011 Status Report newsletter there is an article from
> the NY State DOT by the late Nathan Erlbaum which discusses the issues
> surrounding the various flow data sets. The article “Commutation Flow:
> CTPP 2000, ACS and CTPP, and LEHD-OTM” can be found at
> http://www.trbcensus.com/newsltr/sr0111.pdf (starting on page 6). In it
> is a link to a larger body of work that Nathan did on the topic. a read
> of this is must if you are having questions about the data.
>
> Regarding the LEHD and CTPP flow data, there is an NCHRP report that has
> been completed that is also worth reviewing. It gets into detail about
> some of the differences between the two data sets. It can be found at
> http://onlinepubs.trb.org/onlinepubs/nchrp/docs/NCHRP08-36%2898%29_FR.pdf
>
> Elaine Murakami reminds me that
> 1. LEHD does not include self-employed (approx. 10%)
> 2. LEHD only recently included federal workers, but does not include
> MILITARY
> 3. LEHD is jobs and ACS is "workers at work".
>
> Periodically issue like this pop up and there is a discussion on the
> list serve. there is an archive but off course it takes some time to
> mine and search through. If anyone is interested in looking through it,
> it is open to the public and available at
> http://www.chrispy.net/pipermail/ctpp-news/. There was a discussion
> around august of 2010 dealing with flow data but it centered more on
> availability.
>
> Hope this helps.
> --
> Ed Christopher
> 708-283-3534 (V)
> 708-574-8131 (cell)
>
> FHWA RC-TST-PLN
> 4749 Lincoln Mall Drive, Suite 600
> Matteson, IL 60443
> _______________________________________________
> ctpp-news mailing list
> ctpp-news(a)ryoko.chrispy.net
> http://ryoko.chrispy.net/mailman/listinfo/ctpp-news
_______________________________________________
ctpp-news mailing list
ctpp-news(a)ryoko.chrispy.nethttp://ryoko.chrispy.net/mailman/listinfo/ctpp-news
Hi, Hary
If you compare LEHD with commuting data, you may want to know that commuting data is only workers at work, which excludes workers on vacation, sick leave etc (approximately 1 or 2 pct).
The following article "CTPP Workers-at-Work Compared to Other Employment Estimates" maybe helpful if you try to compare ACS employment data with other sources. Although the article is about CTPP 2000 (Decennial long forms), but they can be applied to ACS as well.
http://www.fhwa.dot.gov/planning/census_issues/ctpp/status_report/sr0104.cfm
Liang
________________________________________
From: ctpp-news-bounces(a)chrispy.net [ctpp-news-bounces(a)chrispy.net] on behalf of hprawiranata mitcrpc.org [hprawiranata(a)mitcrpc.org]
Sent: Friday, March 08, 2013 2:49 PM
To: ctpp-news(a)chrispy.net
Subject: Re: [CTPP] Commuter Flow Data Concerns
Thanks, Ed.
I have been working LEHD data for 2 weeks and comparing it with other
sources (REMI, Claritas, WP, etc). All sources has different number
for 2010, for example, which one is the good one ? LEHD has all jobs,
primary jobs, private jobs, etc. I found the use primary job data give
close number with other sources, am I on the right track ?
I have not compared LEHD data with 'just released' commuting data set,
I guess it would be the same and LEHD has better resolution. What do
you think ?
Thanks,
Hary
On Fri, Mar 8, 2013 at 10:22 AM, Ed Christopher <edc(a)berwyned.com> wrote:
> A few of us have been getting calls and emails from people who have been
> looking at the commuter follow data the Census Bureau just released, the
> On the Map (LEHD) and some of the earlier flow data. If you have, you
> probably noticed some interesting (or maybe wierd) things. Needless to
> say a good understanding of the data is needed to sort out some of the
> issues. Below are some references and points that may be of interest.
> What are people noticing with the data?
> ---------------
> In the January 20011 Status Report newsletter there is an article from
> the NY State DOT by the late Nathan Erlbaum which discusses the issues
> surrounding the various flow data sets. The article “Commutation Flow:
> CTPP 2000, ACS and CTPP, and LEHD-OTM” can be found at
> http://www.trbcensus.com/newsltr/sr0111.pdf (starting on page 6). In it
> is a link to a larger body of work that Nathan did on the topic. a read
> of this is must if you are having questions about the data.
>
> Regarding the LEHD and CTPP flow data, there is an NCHRP report that has
> been completed that is also worth reviewing. It gets into detail about
> some of the differences between the two data sets. It can be found at
> http://onlinepubs.trb.org/onlinepubs/nchrp/docs/NCHRP08-36%2898%29_FR.pdf
>
> Elaine Murakami reminds me that
> 1. LEHD does not include self-employed (approx. 10%)
> 2. LEHD only recently included federal workers, but does not include
> MILITARY
> 3. LEHD is jobs and ACS is "workers at work".
>
> Periodically issue like this pop up and there is a discussion on the
> list serve. there is an archive but off course it takes some time to
> mine and search through. If anyone is interested in looking through it,
> it is open to the public and available at
> http://www.chrispy.net/pipermail/ctpp-news/. There was a discussion
> around august of 2010 dealing with flow data but it centered more on
> availability.
>
> Hope this helps.
> --
> Ed Christopher
> 708-283-3534 (V)
> 708-574-8131 (cell)
>
> FHWA RC-TST-PLN
> 4749 Lincoln Mall Drive, Suite 600
> Matteson, IL 60443
> _______________________________________________
> ctpp-news mailing list
> ctpp-news(a)ryoko.chrispy.net
> http://ryoko.chrispy.net/mailman/listinfo/ctpp-news
_______________________________________________
ctpp-news mailing list
ctpp-news(a)ryoko.chrispy.nethttp://ryoko.chrispy.net/mailman/listinfo/ctpp-news
Please keep your eyes out for this and let us all know what you
think...It looks to be based on migration flows but who knows someday it
might grow to consider commuter flows.
-------
American Community Survey
2006-2010 County-to-County Migration Flows Mapper Tool and 2006-2010
County-to-County Characteristics Migration Tables — Based on data
collected during the American Community Survey from 2006 to 2010, the
county-to-county migration tables give added information on the paths of
the 17.3 million people who moved to another county. The tables provide
the current county of residence, the county of residence one year ago
and the estimated number of movers between the counties. Additional
tables provide the same information broken down by selected
characteristics: age, sex, race or Hispanic origin.
The Census Bureau has developed its first beta version of an online
mapping tool called Census Flows Mapper. It is an interactive
application that allows users to select a county in the U.S. and view
the outbound, inbound and net migration flows for that county.
Additionally, users can choose flows based on characteristics such as
age, sex, race or Hispanic origin. The application also allows users to
download the data for the flow they have selected, zoom in and out on
the map to an area of interest, view additional statistics of the
selected county and print their map to a PDF file. (Scheduled for
release March 19.)
--
Ed Christopher
708-283-3534 (V)
708-574-8131 (cell)
FHWA RC-TST-PLN
4749 Lincoln Mall Drive, Suite 600
Matteson, IL 60443
