Measurement error in recall surveys and the relationship between household size and food demand.

However, even in the empirical results reported here there is a significant negative effect of household size on per capita food demand in most of the samples. This negative effect conflicts with the theoretical predictions of the Barren (1964) two-good model of scale economies. Thus it is likely that measurement error is only one part of the explanation for the puzzle about the effect of household size on food demand. Moreover, the theoretical results indicate that the puzzle raised by Deaton and Paxson cannot be accounted for by uncorrelated measurement error, so to the extent that measurement error does contribute a partial explanation, it can only be in its correlated variety.

The authors are grateful to Andrew Horowitz, Scott Rozelle, the editor and three anonymous referees for helpful discussions and to Trinh Le and Susan Olivia for outstanding research assistance. Any remaining errors are the authors'. The financial support from the AARES-AAEA Young Professionals Exchange Travel Award-Heading North is gratefully acknowledged.

[Received May 2004; accepted July 2006.]

References

Abdulai, A. 2003. "Economies of Scale and the Demand for Food in Switzerland: Parametric and Non-Parametric Analysis." Journal of Agricultural Economics 54(2):247-67.

Ahmed, N., M. Brzozowski, and T. Crossley. 2005. "Measurement Errors in Recall Food Expenditure Data." Mimeo. Canada: Department of Economics, McMaster University.

Attanasio, O., E. Battistin, and H. Ichimura. 2004. "What Really Happened to Consumption Inequality in the U.S.?" Mimeo. London: Institute for Fiscal Studies.

Barten, A. 1964. "Family Composition, Prices, and Expenditure Patterns." In R Hart, G. Mills, and J. Whitaker, eds. Economic Analysis for National Economic Planning. London: Butterworth.

Blair, E., and S. Burton. 1987. "Cognitive Processes Used by Survey Respondents to Answer Behavioural Frequency Questions." Journal of Consumer Research 15(3):280-88.

Bound, J., C. Brown, and N. Mathiowetz. 2001. "Measurement Error in Survey Data." In J. Heckman, and E. Learner, eds. Handbook of Econometrics, vol. 5. Amsterdam Elsevier, pp. 3705-843.

Browning, M., T. Crossley, and G. Weber. 2003. "Asking Consumption Questions in General Purpose Surveys." Economic Journal 113(491):540-67.

Chesher, A., and C. Schluter. 2002. "Welfare Measurement and Measurement Error." Review of Economic Studies 69(2):357-78.

Deaton, A. 1997. The Analysis of Household Surveys: A Microeconometric Approach to Development Policy. Baltimore: Johns Hopkins University Press.

Deaton, A., and C. Paxson. 1998. "Economies of Scale, Household Size, and the Demand for Food." Journal of Political Economy 106(5):897-930.

--. 2003. "Engel's What? A Response to Gan and Vernon." Journal of Political Economy 111(6):1378-81.

Deaton, A., and V. Kozel. 2005. "Data and Dogma: The Great Indian Poverty Debate." World Bank Research Observer 20(2):177-99.

Eisenhower, D., N. Mathiowetz, and D. Morganstein. 1991. "Recall Error: Sources and Bias Reduction Techniques." In P. Biemer, R. Groves, L. Lyberg, N. Mathiowetz, and S. Sudman, eds. Measurement Error in Surveys. New York: Wiley, pp. 127-44.

Gan, L., and V. Vernon. 2003. "Testing the Barten Model of Economies of Scale in Household Consumption: Toward Resolving a Paradox of Deaton and Paxson." Journal of Political Economy 111(6):1361-77.

Gardes, F., and C. Starzec. 2000. "Economies of Scale and Food Consumption: A Reappraisal of the Deaton-Paxson Paradox." Working paper no. 2000-08, TEAM, University of Paris I, Pantheon-Sorbonne.

Gibson, J. 2002. "Why Does the Engel Method Work? Food Demand, Economies of Size and Household Survey Methods." Oxford Bulletin of Economics and Statistics 64(4):341-60.

Grosh, M., and R Glewwe. 1996. "Household Survey Data from Developing Countries: Progress and Prospects." American Economic Review 86(2):15-19.

Horowitz, A. 2002. "Household Size and the Demand for Food: A Puzzle Resolved?" Paper Presented at the Northeast Universities Development Consortium Conference, Williams College, October, 2002.

Jolliffe, D. 2001. "Measuring Absolute and Relative Poverty: The Sensitivity of Estimated Household Consumption to Survey Design." Journal of Economic and Social Measurement 27 (1):1-23.

Kemsley, W., and J. Nicholson. 1960. "Some Experiments in Methods of Conducting Family Expenditure Surveys." Journal of the Royal Statistical Society 123(3):307-28.

Kulshrestha, A., and A. Kar. 2005. "Consumer Expenditure from the National Accounts and National Sample Survey." In A. Deaton and V. Kozel, eds. The Great Indian Poverty Debate. New Delhi, India: MacMillan, Chapter 7.

Lancaster, G., R. Ray, and R. Valenzuela. 1999. "A Cross-Country Study of Equivalence Scales and Expenditure Inequality on Unit Record Household Budget Data." Review of Income and Wealth 45(4):455-82.

Lanjouw, P., and M. Ravallion. 1995. "Poverty and Household Size." Economic Journal 105:1415-34.

McWhinney, I., and H. Champion. 1974. "The Canadian Experience with Recall and Diary Methods in Consumer Expenditure Surveys." Annals of Economic and Social Measurement 3(2):411-35.

Minhas, B. 1988. "Validation of Large-Scale Sample Survey Data: Case of NSS Household Consumption Expenditure." Sankhya, Series B 50(3):S1-S63.

Nicholson, L. 1976. "Appraisal of Different Methods of Estimating Equivalence Scales and Their Results." Review of Income and Wealth 22(1):1-11.

NSSO Expert Group on Sampling Errors. 2003. "Suitability of Different Reference Periods for Measuring Household Consumption: Results of a Pilot Study." Economic and Political Weekly 37(4):307-21.

Perali, F. 2001. "The Second Engel Law and Economies of Scale: An Empirical Puzzle to be Resolved." Working paper no. 28, Centre for Household, Income, Labour, and Demographic Economics, University of Torino.

Pradhan, M. 2001. "Welfare Analysis with a Proxy Consumption Measure: Evidence from a Repeated Experiment in Indonesia." Mimeo.

Scott, C., and B. Amenuvegbe. 1991. "Recall Loss and Recall Duration: An Experimental Study in Ghana." Inter-Stat 4(1):31-55.

Scott, K., and W. Okrasa. 1998. "Analysis of Latvia Diary Experiment." Washington DC: World Bank Development Research Group.

Sudman, S., and N. Bradburn. 1973. "Effects of Time and Memory Factors on Response in Surveys." Journal of the American Statistical Association 68(344):805-15.

White, H., and E. Masset. 2002. "Child Poverty in Vietnam: Using Adult-Equivalence Scales to Estimate Income-Poverty for Different Age Groups." Working paper no. 6, Young Lives: An International Study of Childhood Poverty, Department for International Development, London.

(1) In 1995, 30% of articles in three leading development journals used household survey data, up from only 6% in 1975 (Grosh and Glewwe 1996).

(2) In Taiwan, a subsample of households in the Survey of Personal Income Distribution have their expenditures measured with both the diary and the recall methods. The results from these two methods are then compared to check and correct the results for all of the other sampled households, who are just given the recall method. In the U.S. Consumer Expenditure Survey (CEX), the interview sample comprises a panel of households making four, consecutive, three-monthly recalls of recurring expenses, including food and other groceries with over 600 detailed spending, income, and wealth items in the questionnaire. Another sample use diaries to record purchases for two weeks. The published results integrate the data from the two samples.

(3) This alternating design avoids a flaw in earlier studies that carried out similar comparisons (Kemsley and Nicholson 1960: McWhinney and Champion 1974). These flawed studies asked respondents to recall expenditures from the previous week and then gave the diaries to be completed in the week following. The data from the unbounded recall interview may thus be biased upwards by telescoping errors--the incorrect placement of earlier expenditures in the recall period by the respondent (Sudman and Bradburn 1973).

(4) In the Indonesian survey used below, there is only a 5% chance that a three-person household is composite, but a 23% chance of a ten-person household being composite.

(5) In the diary sample in Gibson (2002), increasing household size from two people to ten people causes the number of food purchases to triple (from 50 to 140 per fortnight) but nonfood purchases only doubled (from 25 to 50).

(6) If the diary is filled out by the enumerator on behalf of either a noncompliant or illiterate respondent this advantage is likely to be lost. There is limited research on the issue of who actually fills out expenditure diaries.

(7) Despite the lack of theoretical support for this method, as first pointed out by Nicholson (1976) and repeated by Deaton (1997), a number of economists continue to use the Engel method in studies of scale economies and the adult-equivalence of children in developing countries. Examples include Lanjouw and Ravallion (1995), Lancaster, Ray. and Valenzuela (1999), and White and Masset (2002).