Is a User-Friendly Diary More Effective? Findings from a Field Test

Sally Reyes-Morales is a mathematical statistician in the Division of Price Statistical Methods, Branch of Consumer Expenditure Surveys, Bureau of Labor Statistics. Diary surveys are often used to collect information on daily activities such as consumer spending. They are particularly useful for collecting daily records of small frequently purchased items, which are normally difficult to recall.1 The Consumer Expenditure (CE) survey, sponsored by the U.S. Bureau of Labor Statistics (BLS), with data collected by the U.S. Census Bureau, uses a diary survey to collect data on weekly household expenditures. Recent efforts to improve the performance of the CE diary survey have focused on designing a more userfriendly form. Such a form would have a simpler recording scheme and be more attractive in appearance than the form currently used in production. Several prototype diaries were developed and refined with the use of feedback from survey respondents, field interviewers, and program staff.2 On the basis of this feedback, CE management selected one of the designs (the Redesigned Diary) for field testing. This diary was intended to stem declining response rates and improve data quality by reducing respondent’s burden associated with the diary now used: the Production Diary. The Redesigned Diary is smaller and shorter than the Production Diary, has a simpler organization, and highlights important instructions and examples. The Redesigned Diary was tested in the field from October through December of 2002.3 The primary objective of this field test was to compare the response rates and data quality obtained from the Redesigned Diary with those obtained from the Production Diary. The results showed no statistically significant difference between diary forms in completion response rates and only a few significant differences in expenditure means and allocation rates. (The latter measure the proportion of expenditures requiring further processing because they are reported with insufficient detail.4 ) However, the Redesigned Diary performed statistically significantly better than the Production Diary in a majority of tests pertaining to the collection of item attribute information needed for