Page 13 of 18

Methodology

Great Expectations is based on a telephone survey of 1,015 adults aged 18 years or older. Interviews were also conducted with 202 African American, 202 Hispanic and 201 white parents of children in high school. The survey was complemented by eight focus groups conducted in sites across the country, as well as by consultation with several experts in the field of higher education policy.

The Telephone Survey

A total of 1,015 telephone interviews with adults aged 18 or older were conducted between December 2 and December 14, 1999. The interviews averaged 28 minutes in length. The interviews were conducted using a random sample of households in the continental United States and a standard, random-digit-dialing technology whereby every household in the region covered had an equal chance of being contacted, including those with unlisted numbers. The margin of error for the 1,015 randomly selected adults is +/ - 3 percentage points; the margin of error is higher in comparison of percentages across subgroups.

Telephone interviews also were conducted in December 1999 with 202 African American, 202 Hispanic and 201 white parents of children in high school. These groups were derived the following way: Respondents matching the demographic criteria (race/ethnicity and parental status) were culled from the national random sample and pooled with additional respondents, obtained through a targeted sampling method for African American and Hispanic parents and a random sample for white parents. The targeted samples included only those telephone exchanges in census tracts with at least a 30% density of households meeting the demographic criteria for African American and Hispanic. Of the 202 African American parents, 34 came from the national sample and 168 from the targeted sample; of the 202 Hispanic parents, 16 came from the national sample and 186 from the targeted sample; of the 201 white parents, 104 came from the national sample and 97 came from additional interviews. The margin of error for each of these groups of parents is +/ -- 7 percentage points.

The survey also refers to "high school parents" as a subgroup within the general population. This group consists of 200 parents of high school students: 163 were culled from the national random sample and pooled with 37 additional respondents obtained from the sample of parents of white high school students. These 37 parents were added in order to make the "high school parents" sample racially representative of the nation.

The survey instrument was translated into Spanish and households identified as Spanish-speaking were interviewed by bilingual interviewers as needed. A total of 14 interviews were conducted in Spanish.

The survey was fielded by Robinson and Muenster Associates, Inc., of Sioux Falls, South Dakota.

The Questionnaire

The questionnaire was designed by Public Agenda, and all interpretation of the data reflected in this report was done by Public Agenda. As in all surveys, question order effects and other non-sampling sources of error can sometimes affect results. Steps were taken to minimize these, including extensively pre-testing the survey instrument and randomizing the order in which some questions were asked.

The Focus Groups

Focus groups allow for an in-depth, qualitative exploration of the dynamics underlying the public's attitudes toward complex issues. Insights from these groups were important to the survey design, and quotes were drawn from them to give voice to attitudes captured statistically through the survey interviews. Additional follow-up interviews were also conducted with some people who responded to the telephone survey.

A total of eight focus groups were conducted in six cities: Old Bridge, New Jersey; Santa Clara, California; Chicago, Illinois; El Paso, Texas; Charlotte, North Carolina; and Bala Cynwyd, Pennsylvania.

DOWNLOAD | PREVIOUS | NEXT

© 2000 The National Center for Public Policy and Higher Education