1 / 10

2005 Census Survey of Maricopa Household Survey

2005 Census Survey of Maricopa Household Survey. Sample Design Overview Sample Size Assumptions. Sample Design Overview. 1. Calculate sample size 2. Select a sample of blocks in each jurisdiction - assign blocks to each jurisdiction

Télécharger la présentation

2005 Census Survey of Maricopa Household Survey

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. 2005 Census Survey of MaricopaHousehold Survey Sample Design Overview Sample Size Assumptions

  2. Sample Design Overview 1. Calculate sample size 2. Select a sample of blocks in each jurisdiction - assign blocks to each jurisdiction - virtually all blocks in sample in small jurisdictions - larger jurisdictions: all large blocks in sample & a subsample of small blocks

  3. 3. Create list of HU addresses in sample blocks - lister given current address list - add HU addresses not on current list - delete HU addresses on current list that are no longer in block - initial listing in May-July 2005 - supplemental list updating in Aug. 2005 4. Select a sample of HU addresses from listing

  4. Sample Size Assumptions 1. Desired accuracy - 95% confidence interval= +/- 2% - one standard error = 1% - see handout for an illustration of the 95% confidence interval expected for your jurisdiction

  5. 2. Vacancy rate • the higher the vacancy rate the larger the sample size needs to be • - 2000 census vacancy rates used • if 2005 rate smaller, smaller sample needed • if 2005 rate higher, larger sample needed • let Heidi know if you expect the 2005 vacancy rate will be different than the • 2000 vacancy rate shown in table 3 of the minutes of the 9/21/2004 POPTAC meeting

  6. the effect of this change in vacancy rates should not have a substantial impact on the sample size • for example, if a jurisdiction needed a sample size of about 3000 using a 10% vacancy rate from 2000 and the 2005 vacancy rate is expected to be 5% then the • sample size could be lowered to 2850. • Likewise, a 10 % vacancy rate raised to 20% in 2005 would require the sample size to be increased to 3375.

  7. 3. Variability in household size • the more the variation in household size the larger the sample size needs to be • 2000 census variability used • if 2005 less variable, smaller sample size • if 2005 more variable, larger sample size • let Heidi know if you expect the 2005 persons per household will be different than the 2000 persons per household shown in Table 3 of the minutes of the 9/21/2004 POPTAC meeting

  8. the effect of this change in persons per HH will be most substantial in jurisdictions with: • a large relative increase since 2000 • the size of the new HHs expected to be a lot different than the current HH size • for example a jurisdiction with • an average HH size of 3.0 • an expected 50% growth in HHs • new HHs mostly senior HHs with 1 or 2 This jurisdiction would probably require a significant increase in sample size due to the increase in HH size variability.

  9. 4. Proportion of total HUs in sample • the higher the % in sample the smaller the sample size needs to be • for example, consider a jurisdiction of 100,000 HUs that requires a sample size of 4750 to achieve the 2% accuracy. • if that jurisdiction was only 5,000 HUs, it would require a sample size of only 2500 • if it was 10,000 HUs, sample size = 3300 • if it was 50,000 HUs, sample size = 4500

  10. 5. Household survey response rate - 100% response rate assumed - 95 % confidence interval error will increase by about the nonresponse rate - 5% nonresponse rate means 95% CI error = +/- 2.05% rather than 2.0% - 10% nonresponse rate means 95% CI error = +/- 2.1% rather than 2.0%

More Related