What's Wrong with Understanding Variation Using a Single-geographic Scale? A Multilevel Geographic Assessment of Life Expectancy in the United States

There has been limited effort to consider multiple areal units or scales in understanding spatial and geographic processes. Treating observed differences in the results by choice of geographic unit of analysis simply as a nuisance is conceptually problematic and can be empirically misleading. We consider the existing research on geographic variations in life expectancy in the United States to demonstrate that prior county-level studies have overestimated the importance of the county level by omitting states. Future investigations should critically assess the relative importance of multiple geographic, spatial, and non-geographic contexts, including an assessment of what units/scales have been omitted.