RAND > L.A.FANS > Survey Design

Return to L.A.FANS Homepage

Overview of L.A.FANS Survey Design

Study Population

L.A.FANS is a panel study of a representative sample of all neighborhoods and all households in Los Angeles County. Poor neighborhoods and families with children are oversampled.

Project Design

Data were collected in two waves, in 2000-2001 and 2006-2008, with the possibility of additional waves in later years. Wave 1 was completed in January 2002. Wave 2 was completed in November 2008. The L.A.FANS follows neighborhoods over time, as well as children and families. Specifically, in the second wave of the survey, L.A.FANS: (1) reinterviewed all sampled children and adults, even if they moved out of the neighborhood, (2) reinterviewed all sampled respondents who remained in the neighborhood, and (3) interviewed a sample of new entrants into the neighborhood. Thus, the L.A.FANS combines the advantages of a panel study of children and families with the advantages of a repeated cross-sectional sample of each neighborhood. This feature allows researchers to examine neighborhood selection, i.e., the process by which families select themselves into and out of neighborhoods.

Sample Design and Characteristics

L.A.FANS is based on a stratified random sample of 65 neighborhoods (census tracts) in Los Angeles County. Poor neighborhoods were oversampled. In Wave 1, an average of 41 households were randomly selected and interviewed within each neighborhood, including an oversample of households with children under 18. Within each household, both adults and children were sampled and interviewed.

Each sampled person was interviewed in the first wave and will be interviewed in each subsequent wave, whether they remain in the neighborhood or move elsewhere. L.A.FANS-2 attempted to reinterview each Wave 1 respondent, whether they remained in the neighborhood or moved elsewhere. In the second wave, a “new entrant” sample of households that moved into the neighborhood in the years since first wave (six years on average) was also selected and interviewed. These households become part of the sample and would be followed in subsequent waves. Sample sizes for each wave can be found in the documentation.

Household Survey Content

In Wave 1, sampled adults were asked to complete: (a) a roster of all household residents, (b) questions about household economic status, health insurance, assets, current and past participation in welfare programs (e.g., AFDC, TANF, GR, Medicaid, SSI, WIC, housing assistance), use of private social service programs, and their neighborhood, (c) questions about education, employment, income, migration and immigration, marital history, neighborhoods of residence, reasons for moving, and social ties and social support, and (d) questions about health status and health care utilization.

For sampled children, their primary caregivers (usually the mother) were asked to complete questions about themselves and the home environment. They were asked about the child's behavior problems, school-related performance and disciplinary problems, and current health and disability status, as well as child care history (including names and addresses of current providers), immigrant status, school enrollment, health and disability history, child support payments, contact with the absent parent, residence history, and use of public and private social service programs (e.g., SSI, school lunch, after-school and recreational programs, Medicaid). Primary caregivers also completed cognitive assessments.

Cognitive assessments were administered to sampled children ages 3 and older. Children ages 9 and older were also asked about expectations for educational attainment, their friends, and social interaction. Children ages 12 and older were also asked about employment, school, and their families and neighborhoods.

In Wave 2, all respondents (including panel respondents and new entrants) who lived within Los Angeles County were interviewed in person. The survey content for these respondents was very similar to that in Wave 1, except that in many cases panel respondents were asked to update their event history information (employment, marriages, schools attended, etc.) for the period between Waves 1 and 2. Wave 2 also included health assessments (anthropometry, spirometry, blood pressure, dried blood spots, and salivary cortisol (for children only) for respondents interviewed in Los Angeles County.

Wave 2 panel respondents living outside of Los Angeles County were interviewed by telephone with a much briefer survey questionnaire. No cognitive assessments or health assessments were collected for these respondents.

Further Information

For details on the project design see:

Sastry, Narayan, Bonnie Ghosh-Dastidar, John Adams, and Anne R. Pebley

The Design of a Multilevel Survey of Children, Families, and Communities: The Los Angeles Family and Neighborhood Survey — 2006

Social Science Research, Volume 35, Number 4, Pages 1000–1024