Data for CHIS 2009 were collected between September 2009 and April 2010 (47,614 adult; 3,379 teen; and, 8,945 child interviews). Beginning in 2011, CHIS began to produce in both one- and two-year estimates. The 2011 Health Profile estimates were based on data collected between June 2011 and December 2011 (22,590 adult; 1,335 teen; and 3,488 child interviews). Estimates for the larger 2011–2012 sample were collected between June 2011 and January 2013 (42,935 adult; 2,799 teen; and 7,334 child interviews). Approximately half of the interviews were conducted during the 2011 calendar year and half during the 2012 calendar year.
CHIS uses a multi-frame probability sample designed to allow estimation of key health indicators for:
- Counties and groups of small counties
- Key ethnic/racial groups and subgroups for each two-year cycle.
Residents of large counties are roughly sampled based on probability proportional to size, while medium and smaller counties have fixed sample sizes to ensure stable county-level (or small county group) estimates for adults. Due to sample size limitations, only regional profiles were provided for 2011 (“one-year”).
The three sampling frames used to collect CHIS data include:
- A random digit dial sample of landline telephone numbers
- A random digit dial sample of cell phone numbers
- Listed sample of surnames associated with Korean and Vietnamese households to oversample these ethnic groups (Korean and Vietnamese households have been oversampled every CHIS cycle).
In CHIS 2011–2012, an American Indian/Alaska Native (AIAN) oversample was conducted to increase the number of AIAN participants and improve statistical stability and precision of estimates for this group. In each CHIS cycle, the sampling frames are combined using statistical weighting adjustments to obtain the overall estimates provided in the Health Profiles.
More detailed information about CHIS sampling, data collection, and weighting methods can be found on the CHIS Design and Methods site.
