Volume 10 — May 02, 2013
A Multilevel Approach to Estimating Small Area Childhood Obesity Prevalence at the Census Block-Group Level
|Minimum County Sample Size||Correlation Coefficient|
Figure 1. The relationship between correlation coefficients of model-based and direct survey estimates and minimum county sample size. The figure shows an increasing positive relationship between the correlation coefficients of model-based and direct survey estimates and the minimum county sample sizes. As the minimum county sample size increases, the correlation between model-based and direct survey estimates also increases. The correlation reaches approximately 0.90 and levels off once the minimum county sample size reaches 150.
The figure is a map showing the block-group level estimates for childhood obesity prevalence, divided into 5 data ranges. The first range, the lowest (3.3%–12.0%), includes portions of states in New England, the Mid-Atlantic, the Midwest, the Rocky Mountains, and the Pacific Northwest. The second (12.1%–16.8%), third (16.9%–22.0%), and fourth (22.1%–28.1%) ranges are distributed throughout the United States. The highest range (28.2%–42.7%) includes portions of states along the southern Atlantic coast, the Deep South, the Southwest, California, and Alaska. In some states, block-level childhood obesity rates vary greatly.
Figure 2. Model-Based Childhood Obesity Prevalence Estimates by Block Group in the United States, 2010.
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors' affiliated institutions.