Optimal Sampling Design with Random Size Clusters for a Mixed Model with Measurement Errors
Our investigation concerns sampling in epidemiological studies, in the presence of both strata and clusters; the problem is to choose the number of clusters to sample in each stratum given that the size of the clusters in general is a random variable. The issue of unplanned randomness in the design seems to have been scarcely addressed in the survey sampling literature. We were motivated by a sample survey — carried out in 1990–1995 by the Italian National Institute of Nutrition (INN-CA) — on the food habits of the Italian population, divided into four geographical areas: the household came in both as random factor which influenced the individual response and — due to the varying number of its members — as a random component of the design which affected the sample size. In this paper we assume various mixed models under different hypothesis on measurement errors (typically correlated) in the response and for each of them find the optimal designs under several optimality criteria, namely the determinant, the trace, the maximum eigenvalue of the unconditional Fisher information of the fixed effect parameters. In all the models we deal with in the present paper, the optimal design depends on just one unknown parameter τ, a given function of the variance components and correlation coefficients. The dependence of the design on τ is investigated through some simulations. The solutions given for the special cases motivated by the INN-CA study should be applicable to a wider variety of situations.
Keywordscorrelated errors epidemiological studies measurement errors mixed models random designs sample surveys
Unable to display preview. Download preview PDF.
- Bingham, S.A. (1987). The dietary assessment of individuals: methods, accuracy, new techniques and recommendations. Nutr. Abstr. Rev. A 57, 705–42.Google Scholar
- Blumer, P.P., Groves, R.M., Lyberg, L.E., Mathiowetz, N.A. and Sudman, S. (Eds) (1991). Measurement Errors in Surveys . Wiley: New York.Google Scholar
- Guseo, R. (1997). Split-plot design: extension and robustness aspects. Statist. Appl. 9, 61–79.Google Scholar
- ISTAT, (1998). Indagine Statistica Multiscopo sulle Famiglie: Aspetti della Vita Quotidiana. Anno 1997. Roma: ISTAT.Google Scholar
- Kaaks, R., Riboli, E. and van Staveren, W. (1995). Sample size requirements for calibration studies of dietary intake measurements in prospective cohort investigations. Am. J. Epidemiol. 142, 557–565.Google Scholar
- Martino, L. (1999a). Problema delia Stima e Aspetti del Disegno di Campionamento in Modelli a Effetti Misti con Errori di Misura. Ph.D. Thesis, Dipartimento di Scienze Statistiche, Università di Bologna.Google Scholar
- Martino, L. (1999b). A multilevel mixed linear model with measurement errors. In Proceedings of the XIV International Workshop on Statistical Modelling, Graz, 19–23 July 1999, pp. 580–83. Heidelberg: Physica-Verlag.Google Scholar
- Rao, J.N.K. (1979). Optimization in the design of sample surveys. In: Optimizing Methods in Statistics: Proceedings of an International Conference Ed. J.S. Rustagi, pp. 419–434. New York: Academic Press.Google Scholar
- Ross, S. (1989). Stochastic Processes. New York: Wiley.Google Scholar
- Stram, D.O., Longnecker, M.P. and Shames, L. (1995). Cost-efficient design of a diet validation study. Am. J. Epidemiol. 142, 353–362.Google Scholar
- Turrini, A. (1996). Survey form applied in a national food consumption study. In Proceedings of the III Meeting dell’Azione Concertata AIR-CAT “Methodologies in Product Development and Market Research”. Barcelona, 3–5 October 1996.Google Scholar
- Wynn, H.P. (1977b). Optimum designs for finite population sampling. In Statistical Decision Theory and Related Topics II Eds S.S. Gupta and J. Berger, pp. 471–478. New York: Academic Press.Google Scholar