Thursday, June 25, 2009

Technical - Statistical

Statistical Links

SAS code for using Replicate Weights
Statistical packages such as stata can handle replicate weights to produce standard errors and tests of significance. However, for the ACS and CPS, the replicate weights are based on a successive difference model rather than a jack-knife or generalized linear model. No statistical package handles successive difference models as an option.
Thus, one has to go through the actual algorithm to produce standard errors based on the successive difference replicate weights. The following program provides the code for this:

SAS code for using replicate weights
Theoretical basis for successive difference model is:
Fay, R. and Train, G. 1995 "Aspects of Survey and Model-Based Postcensal Estimation of Income and Poverty Characteristics for States and Counties," Proceedings of the Section on Government Statistics, American Statistical Association, pp. 154-159.

Technical - Census Geography

Geography Extras

Creating County level statistics from microdata files
The following document illustrates how one can use microdata files to produce statistics at the county level.
One uses a PUMA equivalency crosswalk to determine which PUMAs to combine for a county and/or which PUMA to associate with multiple counties.
The example uses Illinois as an example, illustrating the percent boomers by PUMA and the percent boomers by county.
Contact me for the programming behind this. The goal is to create a widget that will take PUMA-based statistics as input and output county-based statistics as input. However, all that exists at the moment is a SAS-based program.
PUMA Equivalency File [based on 2000 Census]
The following PUMA equivalency file describes how each PUMA corresponds to counties in the US. The ACS uses the same mapping scheme.

Tuesday, June 23, 2009

GIS/Thematic Maps

Shape Files

Posted by lisan at 09:20 AM | Comments (0)

Mapping Resources

ACS Map Making Tutorial
Jennifer Ailshire
GETTING CENSUS DATA for CENSUS 2000: With Emphasis on Mapping Themes from Census 2000
This is a tutorial from the University of Virginia on how to get data into ArcGIS. This is one of the more frustrating steps in making maps. Pay special attention to the bottom of page 2 and top of page 3.
Using ArcGIS
This is a good generic guide to using ArcGIS. It has some specific links to the Yale University library, which developed the resource. Ignore these and use it as a general reference.

Data file for class exercise

Child Poverty [PUMA-based]
Mapping from the Census Bureau website
This is a short description of how to make maps using the Census Bureau website. It requires no GIS skills or software. All of the tools reside on the Census Bureau site. The drawback is the limited flexibility on being able to find a table and/or geography that meets the users' needs.

Sunday, June 21, 2009

Using PDQ Explore

PDQ Explore software

The following is a link to the PDQ-Explore software that you used during the workshop. You will need to download the zip file to the C:\ drive on your PC.
PDQ-Explore
As a reminder of what you need to do to get started, here is the introduction to Using PDQ-Explore, which includes some technical notes.
Introduction to PDQ-Explore


Guides to PDQ Explore

There are about 10 guides that describe in quite a bit of detail how to do many of the procedures covered in the class. If you find you need a reminder, check these out. You will need to scroll down the page to see the guides.

IPUMS

IPUMS-USA

Guide to Variables
Selected variables from IPUMS with PUMS, ACS, and CPS equivalent names
Register to use IPUMS
(required for making an extract)
Extract
(zipped)
Extract
(not zipped)

SAS program

Class Powerpoints

Powerpoint Presentations and Guest Speakers

Below are links to the lectures presented at the workshop. On occasion, there were handouts associated with a presentation. These are listed under the presentation.

An Introduction to Summary Products (and American FactFinder)
[shortened version of a Census Bureau presentation]
Presentation Notes
[shortened version of a Census Bureau presentation]
[See class exercises for the American Factfinder Exercises]
[See class exercises for the rest of the PDQ examples]

Technical Issues (mostly statistical) Associated with the ACS
Detroit: A History of Race Relations through Maps
[To be posted later]
Jason Booza
Introduction to Mapping
Jeff Strickland
The Economy, Public Policy and Poverty in the U.S.: What Changes Can President Obama Make?
[To be posted later]
Sheldon Danziger

Workshop Participants

Workshop Participants

Participants and Instructors/Speakers

Class Exercises

American FactFinder Exercises

[Summary data]
American FactFinder Exercises
Pages 1-4: Unusual Poverty HotSpots
Pages 5-9: Not ready for prime time; census tracts
Pages 10-12: Not ready for prime time; bulk downloads of zip code data
Handouts
Looking at poverty with ACS Summary data

"College town 'poverty' exposed: Low-income students help win grant money"
Stephen Koff and Bill Sloat | Cleveland Plain Dealer
September 17, 2006

Data files
Household Poverty (ACS 2007) [Excel]
Vacancy data for Cleveland census tracts (Census 2000) [Excel]
Median Income for zip codes (Census 2000) [Excel]

PDQ-Explore Exercises
[Microdata]
Using the ACS, 2007
Using several years of the ACS
Using data from Census2000
Using ACS and census data
Looking at change over time
Complex Queries Using "Pointer" items
Using the Current Population Survey

Additional Resources for PDQ-Explore

Indexing for Cost of Living
Variable Names across the Census, CPS, and IPUMS