Focus Group Findings on Prepaid Debit Cards
This fact sheet focuses on the lessons learned from consumers who purchase and use prepaid debit cards.
More infoMethodology
This longitudinal study was designed to track the financial behaviors of low-income households over time. The first phase of the survey was conducted from July 8, 2009, to September 18, 2009, and the second phase nearly one year later from may 17, 2010, to September 29, 2010. With the data that emerged from this two-phase survey, we were able to analyze the financial evolution of the 2,000 tracked households across the economic turmoil of 2009 and 2010.
The initial panel of 2,000 respondents was drawn from eight neighborhoods located throughout the City of Los Angeles. In Phase II, 1,455 respondents from Phase I were successfully contacted again, representing a 28 percent attrition rate. Respondents were recruited to represent two specific subgroups: (a) 1,000 unbanked households, defined as those with no current bank account; (b) 1,000 banked households, defined as those with at least one bank account. In the first phase, the split between subgroups was essentially equal with 1,021 unbanked households and 1,000 banked households. Attrition was slightly greater among banked respondents; in Phase II, we surveyed 784 unbanked households and 671 banked households.
The neighborhoods selected for this study included four previously identified as targets for the Bank on LA campaign and four additional neighborhoods with similar socio¬economic traits that were not targets of Bank on LA. While all eight neighborhoods are low-income, they are geographically and ethnically diverse, representing a variety of economic segments of Los Angeles. The study included areas with a history of redlining and divestment, dynamic immigrant enclaves, underutilized urban spaces under pressure from gentrification and stable working-class communities. We attempted to select communities that could provide not only a sample of low-income Los Angeles, but also a sample of low-income populations across the country.
Respondents were screened and recruited through a door-to-door, interviewer-administered survey with the households selected by random location sampling. Proportional quotas were determined for each of the eight target neighborhood geographies based on the broader sample quota of 1,000 banked households and 1,000 unbanked households. Each neighborhood was further subdivided according to census block group quotas using weighted census data. individual quotas per block group were assigned via a Geographic Information Systems process based on the percentage of each block group within the total population of the census tracts in the neighborhood.
Three households per block were randomly selected as possible starting points. If qualified respondents were not available at the first household, the second household was approached, followed by the third household, if necessary. Upon the successful screening and recruitment of a household, interviewers approached other households on the block, skipping a specific number of homes, based on the block’s population size and target quota. Local city officials assisted field researchers to obtain safe access to large public housing projects and arranged a secure location within the building for interviews.
Data collection for Phase II of the survey was administered by both telephone and in-person interviews.
Field researchers were instructed to request the participation of the head of the household or individuals responsible for household financial decisions. If the relevant individuals were not available, the interviewers returned at a different time. Once deemed eligible, respondents were recruited into the panel and immediately administered the Phase I questionnaire. Interviews were attempted at various times throughout the day, including during both peak and off-peak employment hours, to ensure comprehensive representation. Field staff reflected the ethnic and racial backgrounds of the target areas’ demographics. Because the population of these neighborhoods is heavily Latino, most interviewers were fluent in both Spanish and English; all interviews were conducted in the respondents’ language of choice. A total incentive of $75 was offered to each participating respondent, distributed through the two phases: $30 was offered immediately following Phase I interviews and the remaining $45 upon the successful completion of Phase II interviews.
The survey involved between 70 and 150 questions, depending on the respondent’s answers and financial behaviors. Almost all questions were closed-ended, multiple-choice questions. The interviews lasted between 30 and 60 minutes. For Phase II, the survey instrument was revised slightly to refine questions with ambiguous results. The survey instrument for both phases is available upon request.
