Comments on howard hogan s building new products from analysis of existing data
1 / 14

- PowerPoint PPT Presentation

  • Uploaded on

Comments on Howard Hogan’s Building New Products From Analysis of Existing Data. Dudley L. Poston, Jr. Texas A&M University April 7, 2011. Poston’s Comments.

I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
Download Presentation

PowerPoint Slideshow about '' - feo

An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.

- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
Comments on howard hogan s building new products from analysis of existing data

Comments onHoward Hogan’s Building New Products From Analysis of Existing Data

Dudley L. Poston, Jr.

Texas A&M University

April 7, 2011

Poston s comments
Poston’s Comments

The Census Bureau’s “Improving Operational Efficiency Program” (IOE) is certainly a grand idea and one very deserving of the respect and congratulations of those of us from outside the Census Bureau.

It goes without saying that the 2011 IOE Program and the 2011 Development Program will reduce costs and generate savings, while at the same time focusing on priorities of high value and worth to the Census Bureau.

Poston s comments1
Poston’s Comments

I will focus my commentson Theme 3of the CB’s 2011 Development Program. The three themes of this program are:

Theme 1. Expand the user base and utility of our statistics

Theme 2.  Exceed the expectations of our external and internal customers

Theme 3.  Create new products from existing data

Great i mportance of theme 3
Great Importance of Theme 3

By addressing Theme 3 (Creating new products from existing data), Themes 1 and 2 will fall in line (i.e., expanding the user base of CB statistics; and exceeding customer’s expectations).

At the PAA meetings last week here in Washington, DC, I attended a session where one of the presentations by CB staff directly addressed Theme 3.

Comments on howard hogan s building new products from analysis of existing data

Estimating Domestic Migration by

Demographic Characteristics in the United States:

A Rate-Based Model Using Administrative Records

Caleb Miller, Esther Miller, Rachel Cortes, Rodger Johnson, Charles Coleman, and Steve Smith

Population Division

U.S. Census Bureau

For presentation at The 2011 Annual Meeting of the

Population Association of America

Washington, DC

April 2, 2011

This paper is released to inform interested parties of research and to encourage

discussion. Any views expressed on methodological issues are those of the

author and not necessarily those of the U.S. Census Bureau.

Data methods

  • The IRS produces an annual data extract for the Census Bureau that contains administrative data collected for every 1040 tax form processed.

  • These data contain:

    • Information on filer, plus the spouse of the filer and all exemptions listed.

    • The filer’s address including the ZIP+4

  • The filer’s nine-digit zip code associated with a tax return is geocoded using a ZIP+4-to-county correspondence file.

  • The file contains all the ZIP+4’s for a given state and/or county and/or statistical equivalent area.

  • Geocoding the tax data allows one to determine the geography of migration.

Merging irs and census data
Merging IRS and Census data

  • Consecutive year tax data are matched for each person.

    • Residence in Year X = Migration Origin

    • Residence in Year X+1 = Migration Destination

    • If Residence in Year X = Residence in Year X+1 then no migration has taken place (nonmigrants).

    • If State/County in Year X ≠ State/County in Year X+1 then migration has occurred (migrants).


  • CIMP simplifies the programming steps and mathematical logic used to model domestic migration by utilizing single cell data.

  • CIMP uses an identical approach for both county- and state-level migration, does not combine state and county data, allows data to be calculated at the single cell level, and treats counties independently of states.


  • Figure 1 displays the in- and out-migrant age distributions for Arlington County, Virginia computed using the CIMP and the vintage 2008 migration method, along with the in- and out-migrant age distributions for the state of Virginia.


  • Unlike California’s in-migration distribution which is dominated by cohorts in their twenties, the CIMP method shows that Marin County has a large distribution of in-migrants in their mid to late thirties.

A few other merging possibilities
A Few Other Merging Possibilities

Merging Social Security files with CB files on characteristics to develop new data on the migration of the elderly

Merging Death Certificate files with CB characteristics to develop new data on the hazard of dying

Using additional characteristics from the CB files (say, dealing with household relationships) in the eventual merge with the IRS files or with the Social Security files.

Some possible issues for consideration
Some Possible Issues for Consideration

  • Don’t only work with complete matches of files.

  • Re. the CB’s file of characteristics data, don’t automatically fill in the missing cases prior to the merging.

  • Since CB is undertaking specific statistical analyses of these data, consider using missing data approaches that are more appropriate to the specific analyses.


This objective of Theme 3 of creating new products from existing data is a tremendously important objective.

I have only addressed a few of the issues pertaining to a specific task currently underway (merging IRS data with characteristics data to create new data on migration).

We need to know more about parallel endeavors and tasks at the CB.

In my opinion, the Theme 3 objective is exciting, and an especially relevant one, particularly in these times of reduced budgets.

I have nothing but high praise for the general objective, as well as for the specific CB activity (developing new migration data with IRS and characteristics files) I have just discussed.

Good Work!