Methodology
The methodology underlying America’s Health Rankings® reflects the evolving expectations and role of health in our society and our ability to measure various aspects of health. For each measure the raw data, as obtained from the stated sources and adjusted for age as appropriate, is presented and referred to as “value.” For several measures, such as Infant Mortality and Infectious Disease, data from multiple years are combined to provide sufficient sample size to be meaningful.
All age-adjusted data utilizes the population profile for the middle year of data. For example, if the data is from 2006 to 2008, the standard population is set at 2007.
The score for each state is based on the following formula. The score is stated as a decimal.
SCORE = (STATE VALUE – NATIONAL MEAN SCORE)/(STANDARD DEVIATION OF ALL STATE VALUES)
Often referred to as a “Z-score”, this score indicates the number of standard deviations a state is above or below the national mean. This results in a score of 0.00 for a state with the same value as the national mean. States that have a higher value than the national average will have a positive score while those with a lower value will have a negative score. Scores are calculated to three decimal places and, to prevent an extreme value from excessively influencing a final score, the maximum score any state could receive for a measure is plus or minus 2.
Where a value for the United States overall is not available, the national mean is set at the average value of the states and the District of Columbia.
The overall score was calculated by adding the scores of each measure multiplied by its weight or the percent of total overall ranking. (Note: Scores reported for individual measures may not add up to the overall scores due to the rounding of numbers.)
The ranking is the ordering of each state according to value. Ties in scores are assigned equal rankings.
Overall comparisons to prior years are based upon the relative change in the values of a measure compared to the national average for each measure. The overall result is the weighted sum of these variations. The change between years is the summation of all changes between those years for the components included in the models used for the years of interest.
To calculate the overall change in health, the rate of change in years 1990-1999 was averaged and compared to the average rate of change in 2000- 2011. The overall change is based on the slopes of the linear regression fit of these two periods of time.
The 2011 Edition uses the improved methodology introduced in the 2009 Edition to calculate state ranks. Rankings presented in this edition are comparable to rankings published in the 2010 and 2009 Editions, but they are not comparable to the rankings published in the earlier, printed editions. However, all prior rankings, including 1990 through 2008, have been recalculated using the improved method; they are available at www. americashealthrankings.org, and can be compared to the rankings in this print edition. All historical comparisons discussed in this report are to rankings calculated using the improved method.