|Sponsor:||UCLA Center for Health Policy Research|
|Description:||The California Health Interview Survey (CHIS) is a biennial telephone survey that began in 2001. CHIS collects information on California children (0-11 yrs), adolescents (12-17 yrs), and adults (18 and older) about their health and health care access. Specific topics addressed by the survey include health status, health conditions, health-related behaviors, health insurance coverage, access to and use of health care services, and the health and development of children and adolescents. CHIS data are used to produce population-based estimates for most California counties, all major ethnic groups, and several ethnic subgroups within California. The overall sample is representative of the states non-institutionalized population.|
|Relevant Policy Issues:||Measurement of Health Status, Disease-specific Measurements, Key Health Disparities, Factors Contributing to Measured Health Disparities, Measures of Well-being for Children, and Measures of Well-being for Elders.|
|Unit of Analysis:||Individual|
|Identification of AI/AN/NA:||The CHIS interview asks several questions about race. Below are those relevant to the identification of AI/AN/NA in this data source.
|AI/AN/NA Population in Data Set:||The CHIS 2003 Random Digit Dial (RDD) sample is representative of Californias non-institutionalized population. The numbers below are based on any mention of a specific race or ethnicity (rather than the single-race categories):
Adolescents (ages 12-17)
Children (ages 0-11)
|AI/AN/NA Subpopulations:||Data source allows identification of federally recognized tribes: Apache, Blackfeet, Cherokee, Choctaw, Mexican American Indian, Navajo, Pomo, Pueblo, Sioux, Yaqui.
Additional subpopulation categories include Native Hawaiian alone, Pacific Islander alone, and a breakdown by Pacific Island (i.e., Samoan/American, Samoan, Guamanian, Tongan, Fijian, Polynesian, and Other Pacific Islander). Sample size for some Tribes or Islands may be too small for analysis.
|Geographic Scope:||The geographic scope of the study is the state of California. Additional geographic identifiers include counties, zip codes, and exact longitudes and latitudes of the residence for about 80 percent of the sample.|
|Date or Frequency:||The CHIS survey began in 2001 and is fielded every 2 years. Data from 2003 are the most recent data that are publicly available.|
|Data Collection Methodology:||Data are collected using a computer assisted telephone interviewing (CATI) system, with respondents selected through a geographically stratified random digit dialing approach. One adult in a household is selected and responds for him/herself and one sampled child (0-11 yrs) if there are any in the household. Adolescents respond for themselves, after approval is given from their guardian. Up to three individuals in any given household may be sampled (adult, adolescent, child). In addition to English, the survey is fielded in five additional languages (Spanish, Mandarin, Cantonese, Korean and Vietnamese).|
|Participation:||Optional, without incentives|
|Response Rate:||Most recent response rate was 33.5 percent (composite of screener and interview completion rates).|
|Sampling Methodology:||CHIS employs a multi-stage sample design in which the state of California is divided into 41 geographic sampling strata (primarily counties) and within each geographic stratum, households were selected through random-digit dialing. Within each household, an adult (age 18 and over) respondent was randomly selected, and for those households with adolescents (ages 12-17) and children (under age 12), one of each was randomly selected for interview.|
|Analysis:||The complex survey design of the CHIS requires that adjustments to weighting and standard error calculations be made in order to produce robust estimates. Failure to make these adjustments can yield estimates where the standard error appears smaller than it actually should be, suggesting the accuracy of the estimate is better than it actually is. The technique used to address this adjustment process in the CHIS Public Use Files is replication, and the data set includes the set of replicate weights for users to apply in the calculation of standard errors. Special software is required to conduct statistical analyses using replicate weights.|
|Strengths:||The CHIS data sets contain a large number of AI/AN/NA respondents. The data are collected on key policy issues, including health and child welfare. There are multiple years of data available. The sample size of the overall survey is of sufficient size to allow for analyses of racial/ethnic subgroups of interest, which includes a breakout of Native Hawaiian from Pacific Islander as well as specific tribal affiliations. While sample sizes for certain tribes are small, some tribes do have sufficient sample size to permit at least basic descriptive statistical analyses. Furthermore, many of the questions fielded in the CHIS are taken from the National Health Interview Survey, which can provide a national benchmark for many variables.|
|Limitations:||While a very large survey, results are only generalizable to the state of California. The sample does not include an institutionalized population or people without telephones. Some tribes and other subpopulations have very small sample size that will allow only minimal analyses. The response rate is low.|
|Access Requirements and Use Restrictions:||CHIS Public Use Files are available through a data use agreement at no cost.|
|Contact Information:||The main website for the survey is located at http://www.chis.ucla.edu/. Data can be downloaded at this site. There is also an on-line analysis query system named AskCHIS. In addition, potential users can link to a wide variety of other CHIS-related information at this site.
California Health Interview Survey
"report.pdf" (pdf, 1.56Mb)
"apa.pdf" (pdf, 107.76Kb)
"apb.pdf" (pdf, 88.97Kb)
"apc.pdf" (pdf, 233.07Kb)