Effect of Computer-Assisted Personal Interviews in the U.S. Consumer Expenditure Interview Survey

摘要

The U.S. Consumer Expenditure (CE) Interview Survey began utilizing the CAPI instrument for data collection in April 2003. In this paper, the CAPI method is compared to the conventional PAPI method. We compare mean expenditures and reporting rates across groups defined by expenditure categories, respondent demographic variables, and interviewer variables. We will provide the relevant literature review, with special attention to the mode effect. We also will present an overview of the CE Interview Survey and CAPI application. Finally, we will describe the comparison study and present the statistical tools used. The proposed methods are applied to selected subsets of items from the CE Interview Survey. Specifically, we distinguish between parts that were exact translations of the PAPI version and parts with additional changes, such as wording and order changes.