Intra-class Correlation Estimates for Assessment of Vitamin A Intake in Children

In many community-based surveys, multi-level sampling is inherent inthe design. In the design of these studies, especially to calculate theappropriate sample size, investigators need good estimates ofintra-class correlation coefficient (ICC), along with the cluster size,to adjust for variation inflation due to clustering at each level. Thepresent study used data on the assessment of clinical vitamin Adeficiency and intake of vitamin A-rich food in children in a districtin India. For the survey, 16 households were sampled from 200 villagesnested within eight randomly-selected blocks of the district. ICCs andcomponents of variances were estimated from a three-level hierarchicalrandom effects analysis of variance model. Estimates of ICCs andvariance components were obtained at village and block levels.Between-cluster variation was evident at each level of clustering. Inthese estimates, ICCs were inversely related to cluster size, but thedesign effect could be substantial for large clusters. At the blocklevel, most ICC estimates were below 0.07. At the village level, manyICC estimates ranged from 0.014 to 0.45. These estimates may provideuseful information for the design of epidemiological studies in whichthe sampled (or allocated) units range in size from households to largeadministrative zones.