Optimal Recall Period Length in Consumer Payment Surveys

• Activity Number: 590
• Type: Topic Contributed
• Date/Time: Wednesday, August 7, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Marcin Hitczenko*+
• Companies: Federal Reserve Bank of Boston
• Keywords: modeling survey recall ; payment use ; mean-square error

Abstract

In surveys that ask respondents to recall behavior over a specific period of time, there is a well-known tradeoff in increasing the length of the recall period. On the one hand, collecting data over longer periods of time provides the researcher a view of a wider variety and larger sample of behavior patterns. On the other hand, there is worry that respondents tend to forget events or begin relying on cognitive techniques other than pure memory recall to provide estimates, resulting in biased responses. One can thus ask what the optimal length of the recall period should be for a particular study. In this work, we study the effect of the recall period on the accuracy of reporting past usage for four different payment instruments. We develop a model mapping the actual number of uses to the reported numbers, and present a methodology for estimating this model using data from a three day diary study of consumer payments. We then consider the implications of this model for mean-square errors based on reported values for different recall periods.