Measuring Total Household Spending in a Monthly Internet Survey: Evidence from the American Life Panel
Beginning in May 2009 we fielded a monthly Internet survey designed to measure total household spending as the aggregate of about 40 spending components. We developed a unique summary table which permits a respondent to correct erroneous entries. This paper reports on a number of outcomes from 30 waves of data collection in panel. These outcomes include sample attrition, indicators of data quality such as item nonresponse and the variance in total spending, and how the variance is affected by the summary table. Total spending in our sample aggregates closely to total spending in the Consumer Expenditure Survey as reported in published estimates. We conclude that high-frequency surveying for total spending is feasible and that the resulting data show expected patterns of levels and change.
We are grateful to the National Institute on Aging for research support and funding for data collection under grants P01 AG008291, P01 AG022481, P30 AG012815, and R01 AG020717. We are grateful to the Social Security Administration for funding of data collection and research support