Effect of Recall Duration on Reporting of Household Expenditures: An Experimental Study in Ghana

摘要

An experimental study was carried out in Ghana with a sample of 135 households each of which was interviewed 11 times at varying time intervals. Recall of expenditures on 13 frequently purchased items was compared according to the length of the recall period. Clear evidence of recall loss was seen, with average daily expenditures falling consistently in association with increasing length of recall period from 1 day to 1 week. There was little or no further fall in going to 2 weeks. For a 1 year recall period data were erratic. Examining recall in the 1 day to 1 week range, weekly and monthly cyclic variations in expenditure were eliminated by re-weighting. Start-up bias was also ruled out, and in principle the recall periods were always bounded. In these circumstances the association of recall lapse with recall duration became even clearer with an average loss of 2.9% for every day added to the recall duration. Between different expenditure items the extent of the decline was associated positively with mean frequency of purchase of the item. The authors favor the interpretation that the 1-day recall data are the most accurate and argue the urgent need for further studies to explore the limits of the recall effect and possible means of palliating it. (Copyright (c) 1990 The World Bank.)