I wonder how the median life expectancy has been changing in the past few years. Using the mean certainly allows for some shock and awe, and certainly serves t

I wonder how the median life expectancy has been changing in the past few years. Using the mean certainly allows for some shock and awe, and certainly serves to scare people (considering how much it is skewed by things like infant mortality, drug epidemics, wars, and most anything else that involves people dying at drastically different ages than the otherwise usual range), but it's always seemed to me that the median would be a more useful number. Indeed, the median has generally been a good few years above the mean, at least in America, though I can't say I've found tracking the data down nearly as easy as with the mean (okay, I tried for like, 7 seconds...but it wasn't in the top few results). Never mind the mode, which is even higher than either. It'd be nice to see data compiled showing the relation of mean, median, and mode, plotted over time, for life expectancy data. At the very least it'd likely help curb inflation, as people recognize that they're going to need a lot more retirement savings if they're suddenly saving for an extra decade than they were expecting using statistics improperly...