1 / 17

Prepared for the 2010 SDC-CIC Joint Steering Committee Meetings February 23, 2010

American Community Survey: Controlling Estimates of Group Quarters Population at the County Level. Prepared for the 2010 SDC-CIC Joint Steering Committee Meetings February 23, 2010. Presented by: Michael Beaghen Department of Commerce U.S. Census Bureau Washington, DC 20233. 1.

moana
Download Presentation

Prepared for the 2010 SDC-CIC Joint Steering Committee Meetings February 23, 2010

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. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. American Community Survey: Controlling Estimates of Group Quarters Population at the County Level Prepared for the 2010 SDC-CIC Joint Steering Committee Meetings February 23, 2010 Presented by: Michael Beaghen Department of Commerce U.S. Census Bureau Washington, DC 20233 1

  2. Overview • How we control ACS estimates of GQ population • Why we don’t control ACS estimates of GQ population at the county-level

  3. Why Control ACS Estimates? • Reduce coverage bias • Make ACS estimates consistent with the Population Estimates Program (PEP) estimates

  4. How are Estimates of GQ Population Controlled? • Adjust total weight of sampled GQ persons to equal PEP estimates • At the state level • By seven major types of GQ

  5. Seven Major GQ Types Used in Controlling • Correctional institutions • Juvenile facilities • Nursing homes • Other long-term care facilities • College dorms • Military facilities • Other non-institutional GQs

  6. Example of Controlling • PEP estimate of the population in college dorms in Big State is 11,000 • ACS pre-controlled estimate of population in college dorms is 10,000 • Adjust the weight of each ACS sample person in a college dorm by a factor of 1.1

  7. Question • Why not control ACS estimates of GQ population at the county-level? • Achieve consistency between ACS and PEP estimates down to county-level

  8. Two Possible Alternatives under Consideration • Adjust ACS to PEP county estimates by major type of GQ • Adjust ACS to PEP by total county GQ population

  9. Major Limitation for these Methods of GQ County Controls • Some counties don’t have GQ sample for one or more major types of GQ which exist in the county.

  10. Counties With and Without ACS Sample of Major Types of GQ in 2006-2009 ACS

  11. Consequences of not having ACS Sample for Specific Types of GQ • Controlling by major type of GQ requires combining types of GQ • If we don’t have sample for a type of GQ, have to combine this type with a GQ type with sample • Controlling by total GQ population • Combines all seven types of GQ

  12. Consequences Continued • Would lead to distortions in ACS county estimates of GQ population • Because weights of people in types of GQ in sample would be adjusted up to account for people in types of GQ not in sample

  13. Example of Distortion • A county has two GQ facilities with total PEP GQ population of 220 • Nursing home with 200 • Juvenile facility with 20 • Juvenile facility is in ACS sample • Nursing home not in ACS sample • County-controlled ACS estimate • 220 people in a juvenile facility

  14. Distortion of Estimates • Distorts estimates of population by type of GQ • Distorts estimates of characteristics, because people in different types of GQ have different characteristics

  15. Another Consideration • PEP state estimates of GQ population are more robust than county estimates • County-level errors tend to cancel at the state-level

  16. Next Steps • We are considering county-level controls • Research is required to evaluate the severity of distortions on ACS county and state estimates • We are also considering very new estimation methodologies which would make controlling at the county-level less problematic

  17. Contact • Michael.a.beaghen@census.gov

More Related