Appendix B: Detailed Methology AARP Multicultural Survey
Detailed Methodology and Statistical Methods
A. Detailed Methodology
This section contains descriptions of the basic research and sample designs for the study. It also presents brief summaries of the research procedures (questionnaire construction, pretests, and interviewing procedures) used to implement the research design. Each of the four sample frames is described. Finally, the weighting scheme devised for the study is explained.
To realize AARP’s desire to better understand the attitudes and experiences of the Sandwich Generation (people between the ages of 45 and 55), the research team from Belden Russonello & Stewart (BRS), Research/Strategy/Management (RSM), and AARP’s Research Office designed a multi-faceted attitudinal research project. This design called for drafting and pre-testing a questionnaire for a nationwide survey, devising a series of sample frames to include a nationally representative sample and over samples for three racial/ethnic groups (African-Americans; Hispanics/Latinos; and Asian Americans), and conducting a series of elite, in-depth interviews with members of the general public. A description of each of these components of the research design follows.
A questionnaire was drafted for the nationwide survey and was subsequently reviewed by the research team and other interested parties within AARP, undergoing several iterative revisions. When a final draft was completed, it was programmed into a CATI system and pre-tested in an iterative procedure on March 2 and March 6. Eight interviews, mostly with minority respondents, were initially conducted by trained interviewers and monitored by members of the research team. Based on the results of this first phase of the pretest and a debriefing of the interviews, the questionnaire was revised in several areas, shortened, and reprogrammed. Another pretest of 13 interviews was then conducted, which was also monitored by members of the research team. More revisions were made in the questionnaire until the questionnaire was ready to be fielded. Some minor wording and response category changes were made in the initial days of interviewing to meet feedback from AARP and the respondents. The questionnaire was translated into Spanish and this version was used for Spanish-speaking respondents who preferred to be interviewed in their native language.
This study contains the results of 2,352 telephone interviews with residents of the 50 United States, who are between the ages of 45 and 55. 1,500 of these interviews were carried out among a nationally representative probability sample of the U.S. population for the indicated age group. In addition, we over sampled 251 African-Americans (for a total of 404), 316 Hispanics (for a