U.S. Census Bureau’s U.S. Census Bureau’s American Community Survey: American Community Survey: Implications for Rural Implications for Rural Alaska Alaska Nicole Grewe, Ph.D. Nicole Grewe, Ph.D. Division of Community & Regional Division of Community & Regional Affairs Affairs Alaska Municipal League Annual Conference November 10 – 14, 2008 Angoon Kotzebue
18
Embed
U.S. Census Bureau’s American Community Survey: Implications for Rural Alaska Nicole Grewe, Ph.D. Division of Community & Regional Affairs Alaska Municipal.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
U.S. Census Bureau’s U.S. Census Bureau’s American Community Survey: American Community Survey: Implications for Rural AlaskaImplications for Rural Alaska
Nicole Grewe, Ph.D. Nicole Grewe, Ph.D. Division of Community & Regional AffairsDivision of Community & Regional Affairs
Alaska Municipal League
Annual Conference
November 10 – 14, 2008
Angoon Kotzebue
Decennial Census Short Form - ContinuedDecennial Census Short Form - Continued Counts populationCounts population
Decennial Census Long Form - DiscontinuedDecennial Census Long Form - Discontinued Describes population Describes population
American Community Survey (ACS) - ImplementedAmerican Community Survey (ACS) - Implemented Describes populationDescribes population
Reengineering the Decennial Census:Reengineering the Decennial Census:
The Sample:The Sample: ACS survey data is collected from every census ACS survey data is collected from every census
area and borough in Alaska from independent area and borough in Alaska from independent monthly samples of randomly selected addresses. monthly samples of randomly selected addresses.
A Statistics Moment - SamplingA Statistics Moment - Sampling
Good Survey Data:Good Survey Data: Sound samplingSound sampling Quality addresses for full sample distributionQuality addresses for full sample distribution High response rateHigh response rate
2000 Census vs. ACS Sample2000 Census vs. ACS Sample
2000 Census Sample:2000 Census Sample: 17% US households sampled17% US households sampled 50% rural AK households sampled50% rural AK households sampled
ACS Sample:ACS Sample: 19% - 50% rural AK households sampled19% - 50% rural AK households sampled Smaller sample size requires rolling averagesSmaller sample size requires rolling averages
2000 Census vs. ACS Sample:2000 Census vs. ACS Sample: Moving or “rolling averages”Moving or “rolling averages” Different reporting periodsDifferent reporting periods Different size samplesDifferent size samples Smaller sample size increases sampling errorSmaller sample size increases sampling error
ACS ChallengesACS Challenges
Response Rate:Response Rate: Goal: mirror 2000 CensusGoal: mirror 2000 Census Places > 65,000 population = 1 year tabulationsPlaces > 65,000 population = 1 year tabulations Places < 65,000 population = 3 or 5 year tabulationsPlaces < 65,000 population = 3 or 5 year tabulations Statewide data requires good response rates from both Statewide data requires good response rates from both
2.2. Survey removals = Survey removals = smaller sample size and smaller sample size and reduced confidence.reduced confidence.
3.3. Marginal sample size Marginal sample size may be so severely may be so severely impacted the data is not impacted the data is not released.released.
Kasaan
Remote Alaska ExceptionRemote Alaska Exception
Classification of Borough/Census Area Quantity Includes
Totally Remote Alaska 9Aleutians E, Aleutians W, Bristol Bay, Denali, Lake and Penn, NW Arctic, SE FAI, Wade Hampton,
Yukon Koyukuk
Partially Remote Alaska 5 Bethel, Dillingham, Nome, North Slope, Valdez-Cordova
Totally Non-Remote Alaska 13
Comparability: Comparability: ACS and 2000 CensusACS and 2000 Census
Target Population:Target Population: ACS = householdACS = household 2000 Census = population2000 Census = population
Time PeriodTime Period ACS = continuous rolling natureACS = continuous rolling nature 2000 Census = point in time estimate2000 Census = point in time estimate
Reference Time Periods:Reference Time Periods: ACS = 12 months prior, shifting monthsACS = 12 months prior, shifting months 2000 Census = April 1, current decennial year2000 Census = April 1, current decennial year
ACS and Rural AlaskaACS and Rural Alaska
27 Alaska Census 27 Alaska Census Areas/Boroughs:Areas/Boroughs:
5-Year Rolling Average: designed for 5-Year Rolling Average: designed for ≤≤ 20,000 population. 20,000 population.
Ninilchik
Research ConsequencesResearch Consequences
More frequent data, but perhaps unreliable data.More frequent data, but perhaps unreliable data.
Characteristic Comparability Challenges - Characteristic Comparability Challenges - describing the population differently.describing the population differently.
Longitudinal Challenges – 2000 Census and ACS Longitudinal Challenges – 2000 Census and ACS data often not comparable. data often not comparable.
Potentially higher margins of errors.Potentially higher margins of errors.
Limited Test Data - 2010 ACS is test for rural AK.Limited Test Data - 2010 ACS is test for rural AK.
Community Development and Community Development and ConsequencesConsequences
Potentially unreliable community demographics.Potentially unreliable community demographics.
Challenges in tracking demographic community Challenges in tracking demographic community change over time. change over time.
Consequences for planning process.Consequences for planning process.
Consequences for programmatic funding.Consequences for programmatic funding.
Be an informed user - critically evaluate. Be an informed user - critically evaluate.