Estimating Population Means in the 2012 Survey of Consumer Payment Choice
2015 • No. 15–2
By Marcin Hitczenko
This report examines the effect of adding to a longitudinal panel on estimates of population parameters in the 2012 Survey of Consumer Payment Choice (SCPC) more than 1,000 newly recruited respondents specifically targeted to fill segments of the U.S. population that tend to be underbanked and underrepresented in the longitudinal panel. In many ways, the new respondents have fundamentally different characteristics from the ongoing respondents. To minimize confounding sources of change to annual estimates when making comparisons across years, the official 2012 SCPC publication was based on the longitudinal panel only. This research data report presents estimates based on all available responses and evaluates the effect of the additional panel on these estimates. A complete set of 2012 SCPC tables with estimates based on the full sample accompanies this report.
- Although the new SCPC respondents tend to be younger, more likely to come from households with lower incomes, and more likely to represent minority groups, especially Hispanic ones, for most variables there is little evidence of substantial differences in many estimates based on the two samples. However, inclusion of the newly added group leads to lower estimates of certain variables closely related to the ownership and use of traditional bank accounts.
- Estimates based on the full sample show lower adoption rates of bank accounts and, most notably, lower adoption rates of payments associated with such accounts, such as checks, debit cards, online banking, and bank account number payments.
- Even when considering the mean number of monthly uses by payment instrument adopters only, the estimates based on the full sample show the payment instruments most closely tied with bank accounts being used less frequently, and the results with the full sample provide weak evidence of lower cash holdings on person, lower value of prepaid card reloadings, and higher use of mobile banking.
In many respects the panel relied on for estimates in the 2008–2012 SCPC provides adequate estimates for the U.S. population as a whole, at least as compared with the full sample. However, in terms of determining parameter estimates for 2012, the targeted recruiting strategy and sampling strategy seem to have improved the quality of the population parameters by improving the coverage of important subpopulations, and there seems little reason to avoid using the full sample. The results of this work illustrate that it is important to have good coverage across the entire socio-economic spectrum to generate accurate population estimates.
This paper examines the extent to which the addition of newly recruited respondents to a longitudinal panel in the 2012 Survey of Consumer Payment Choice affects population estimates. The new respondents were specifically targeted to fill segments of the U.S. population that tend to be underbanked and underrepresented in the longitudinal panel. We discuss key differences between the longitudinal panelists and the new members and how those differences could affect certain population estimates. We look at a collection of 54 variables and find that changes in estimates of most variables are not significant. As expected, there is evidence that the improved coverage of individuals who are young, low income, or members of a minority population leads to lower estimates of certain variables most closely related to ownership and use of traditional banking accounts.