Harvard School of Public Health/Robert Wood Johnson Foundation -- Subethnicities Survey, United States, 2007 (ICPSR 38367)

Time Period(s) View help for Time Period(s)

Date of Collection View help for Date of Collection

Data Collection Notes View help for Data Collection Notes

Study Design View help for Study Design

Respondents were eligible if they were 18 years of age and older, with varying quota by White (and "others"), Black, Asian, Hispanic, or Native American.

Sample View help for Sample

The study was conducted for Harvard School of Public Health/Robert Wood Johnson Foundation via telephone by ICR-International Communications Research, an independent research company. Interviews were conducted from May 17 to August 10, 2007 among a representative sample of 4,334 respondents age 18 and older. Of those a total of 1,067 respondents were Hispanic including 100 Cubans and 100 Puerto Ricans. A total of 1,001 were non-Hispanic Whites. A total of 1,100 were non-Hispanic Blacks including 100 African respondents and 103 Caribbean respondents. A total of 102 were Native American and a total of 1,064 were non-Hispanic Asian, including 101 Japanese respondents. The margin of error for total respondents is +/- 1.99 percent at the 95 percent confidence level.

A disproportionate stratified sampling design was utilized for this study. This design, in short, breaks telephone exchanges into strata and then disproportionately dials areas of high ethnic incidence as compared to areas of low incidence. The net result is a sample design that attains 100 percent coverage of the target population while significantly increasing incidence over national population estimates.

For this study, exchanges were divided into the following 16 strata: Native American High, Native American Medium, African American High, African American Medium, Africans High, Caribbean High, Asian Medium, Asian Low, Japanese High, Vietnamese High, Hispanic Very High, Hispanic High, Hispanic Medium, Cuban High, Puerto Rican High, and Remainder. The design of these strata ensured that not only would the research team would meet their overall goal per each ethnicity, but that subethnic goals (such as Africans and Japanese) would be met as well.

The sample was an open design, meaning that all ethnicities were obtained from any strata when they were encountered (for example, a Native American interview was not terminated just because that person was found in the Japanese strata). This ensured accurate national representation for each subethnic group. Targets were set to ensure that groups did not go disproportionately over in each strata as compared to Census data of the exchanges within each strata.

Universe View help for Universe

National adult

Unit(s) of Observation View help for Unit(s) of Observation

Data Type(s) View help for Data Type(s)

Mode of Data Collection View help for Mode of Data Collection

Response Rates View help for Response Rates

The overall response rate for this study was calculated to be 53.8 percent.

Original Release Date View help for Original Release Date

2022-03-10

Weight View help for Weight

To ensure an accurate reflection of current population estimates for every subethnic group, the data was weighted separately for each of the following: Whites, African Americans, Native Americans, Chinese, Filipinos, Japanese, Koreans, Vietnamese, Asian Indians, remaining Asians, Mexicans, Puerto Ricans, Cubans, Central and South Americans, and other Asians. Within each of the groups, a pre-weight was developed to correct for the disproportionality of the sample design by strata. Then a post-stratification scheme was used to weight by gender, age, region, and education. Data was based on United States Census 2006 Current Population Survey (CPS) data. Each subethnic group's run of weights was balanced to the correct proportion of that group's representation in the overall U.S. population. Thus, when put all together, the data weights to overall 18+ general population parameters.

The weight variables in this study are POPWGHT, WEIGHT, and SUBWEIGH.