Abstract
Recall food expenditure data, which is the basis of a great deal of empirical work, is believed to suffer from considerable measurement error. Diary records are believed to be more accurate. We study an unusual data set that collects recall and diary data from the same households and so allows a direct comparison of the two methods of data collection. The diary data imply measurement errors in recall food expenditure data that are substantial, and which do not have the properties of classical measurement error. However, we also present evidence that the diary measures are themselves imperfect.
Highlights
Information on household food expenditure is crucial for a broad range of economic and policy research, including research on consumption and demand behaviour, and on living standards, poverty and inequality
Under the assumption that true food expenditure can be constructed from the diary records, measurement errors in recall food expenditure data appear to be substantial, and they do not have the properties of classical measurement error
The Canadian Food Expenditure Survey (FoodEx) survey allows for the comparison of simple recall food expenditure questions with diary methods of collecting food expenditure data using a within-subject design
Summary
Information on household food expenditure is crucial for a broad range of economic and policy research, including research on consumption and demand behaviour, and on living standards, poverty and inequality. Because diary measures are widely considered the gold standard for collecting expenditure information, and because of the within-subject design, it is possible to use treat the FoodEx as an approximation to a validation study of the recall data, and to carry out analyses similar to those discussed by Bound et al At the same time, how well the FoodEx approximates a genuine validation study depends on how well the diary measures capture true expenditure, and we investigate this question. Under the assumption that true food expenditure can be constructed from the diary records, measurement errors in recall food expenditure data appear to be substantial, and they do not have the properties of classical measurement error They are neither mean independent of true expenditure nor homoscedastic.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
