Excessive air pollution is thought to have a negative long-term effect on health. In this study, the “pollution potential” (the product of the tons emitted per day per square kilometer of a pollutant and a dispersion factor which accounts for mixing height, wind speed, number of episode days, and dimension of the area over which measurements were taken) of three pollutants (hydrocarbons, nitrogen oxides, and sulfur dioxide) was determined for 60 metropolitan statistical areas in the United States. The outcome for this analysis is the total age-adjusted mortality rate in the metropolitan area. To account for potentially confounding effects, a number of demographic variables were included in the analysis as well.
y: Total age-adjusted mortality from all causes (Annual deaths per 100,000 people)X: Rows of X are labeled with the name of the metropolitan area; columns are as follows:
Precip: Mean annual precipitation (inches)Humidity: Percent relative humidity (annual average at 1:00pm)JanTemp: Mean January temperature (degrees F)JulyTemp: Mean July temperature (degrees F)Over65: Percentage of the population aged 65 years or overHouse: Population per householdEduc: Median number of school years completed for persons 25 years or olderSound: Percentage of the housing that is sound with all facilitiesDensity: Population density (in persons per square mile of urbanized area)NonWhite: Percentage of population that is nonwhiteWhiteCol: Percentage of employment in white collar occupationsPoor: Percentage of households with annual income under $3,000 in 1960HC: Pollution potential of hydrocarbonsNOX: Pollution potential of oxides of nitrogenSO2: Pollution potential of sulfur dioxideI obtained the data from the Sleuth3 package. The original reference is:
McDonald GC and Schwing RC (1973). Instabilities of Regression Estimates Relating Air Pollution to Mortality. Technometrics, 15: 463-481.