How the estimates are made

The earlier parts report gradients, rankings and rates, and each depends on choices that a chart does not show. The size of the area that a number describes, the weight each place carries, the way a model or a grouping is built and the counts that a source withholds can each change what a comparison means. This part takes those choices one at a time and shows how each one moves a result. The construction of the deprivation index and the comparison of tracts with counties open the book, in Chapter 1.1, What the deprivation index measures and Chapter 1.2, What a county average leaves out. The methods page lists the rules that every chapter shares.

Chapter 10.1, Gradients, slopes and variance shows how the slope of life expectancy on the deprivation index is estimated across all tracts and within counties, and finds lower life expectancy in more deprived tracts in every metro it fits separately. It splits the variation in tract life expectancy among states, counties and tracts, and part of the share that lies between tracts in the same county is estimation error in the modeled tract estimates. It also shows that a correlation changes with the size of the area.

Chapter 10.2, Families of places sorts tracts into groups by their deprivation profile instead of ranking them, and checks the groups against outcomes that the sort never used. The groups are regions of a continuum and not separate clusters, and the chapter finds families that the index scores differently but that show no detectable difference in life expectancy.

Chapter 10.3, Suppressed counties in death data explains a rule that every chapter using CDC WONDER death counts follows. WONDER withholds small counts, and the withheld county-years cover many places and few residents. A county enters the published data in the years when its count is high, so a rate pooled over only the published years is biased in a direction that depends on the size of the county. The book uses complete counties, those published in every year, or a modeled source instead. The chapter also covers a second rule, which decides which counties WONDER names at all in its birth products.