OMB req to increase NHPI screeners

DEPARTMENT OF HEALTH & HUMAN SERVICES

Centers for Disease Control and Prevention

National Center for Health Statistics

3311 Toledo Road

Hyattsville, Maryland 20782

July 22, 2014

Margo Schwab, Ph.D.

Office of Management and Budget

725 17th Street, N.W.

Washington, DC 20503

Dear Dr. Schwab:

The National Center for Health Statistics (NCHS), Division of Health Interview Statistics is conducting the National Health Interview Survey (OMB #0920-0214 exp. 12/31/16) on a sample of addresses from the 2012 American Community Survey (ACS) where a household member was identified as Native Hawaiian or Pacific Islander (NHPI) alone or in combination with one or more other races (the NHPI NHIS). This request is to increase the number of ACS addresses used in the NHPI NHIS sample.

For many years, NCHS has used oversampling and conventional screening methods to increase the proportion of some minority population groups but has not been able to adequately increase the number of smaller groups. With NHPI persons comprising only 0.4% of the US population, few population-based surveys can include samples of sufficient size to yield statistically relevant detailed estimates. NCHS sought an alternative sample source that would yield a more robust sample of NHPI households. The ACS, with its sample of 3 million households annually, was the most cost-efficient and operationally compatible sample source.

Beginning in February 2014, the NHPI NHIS was conducted on one-half of the 2012 ACS addresses where at least one household member was reported to be NHPI. Of the 4,000 contacted, about 41% no longer had a resident who identified as NHPI. Although 82.5% of addresses with an NHPI resident completed the interview, this reduction in the eligible sample resulted in much fewer completed NHIS interviews than was anticipated which seriously reduces the stability of estimates generated from the survey. The sample size originally selected was chosen so that national estimates for NHPIs calculated using this survey’s data could be reliably compared with NHIS estimates for each of the other four major race categories. The assumptions underlying the decision to use half of the NHPI ACS sample were not met.

To ameliorate the reduction in sample size, NCHS asked the Census Bureau for permission to conduct the NHIS with the remaining approximately 4,000 ACS addresses. This request was granted. Due to the large proportion of “screener-only” cases (1,443) anticipated, the increase in the sample will not exceed the burden requested in the information collection budget table.

cc:

V. Buie

T. Richardson

DHHS RCO