Sometimes Kerry Makes Me Angry Options

[Hartmann]

http://www.newsvine.com/_news/2006/10/31/4...ms-kerry-remark

He's generalizing and it's just to dig at Republicans with.

Not everyone know that's in Iraq has a college education but some of them didn't want to go to college and others went to college but wanted to go into the military.

[Hartmann]

Interesting research into our armed-forces: http://www.heritage.org/Research/NationalS...ty/cda06-09.cfm

[pebkac]

Interesting research into our armed-forces: http://www.heritage.org/Research/NationalS...ty/cda06-09.cfm

Sigh.... yet another cornball study from the heritage foundation. Seriously, ignore think tank research. Whether liberal or conservative, they all have ideologies and will manufacture studies that reinforce those ideologies.

Case in point...

The Department of Defense (DOD) does not track family income data for recruits, and there are no individual income data for enlistees. Military service is the first full-time job for most of them. We approx­imate each recruit’s household income by using the median household income of his or her hometown ZIP code.

By assigning each recruit the median 1999 household income for his hometown ZIP code as deter­mined from Census 2000, the mean income for 2004 recruits was $43,122 (in 1999 dollars). For 2005 recruits, it was $43,238 (in 1999 dol­lars). These are increases over the mean incomes for the 1999 cohort ($41,141) and 2003 cohort ($42,822). The national median published in Cen­sus 2000 was $41,994. This indicates that, on aver­age, the 2004 and 2005 recruit populations come from even wealthier areas than their peers who enlisted in 1999 and 2003.

When comparing numbers (especially income), it's important to understand why economists and sociologists rarely use the mean. The problem with averages is that they're far too easily skewed by extremely large figures. Imagine you're comparing three people, a person who makes 10k a year, a person who makes 30k a year, and someone who makes 10 million a year. If we take the average, this means that the average person in that group makes about $3.3 million. If we take the median, that means the average person makes 30k a year. While that figure is still a ways off (mainly due to the fact that the sample size is so small), it's a lot more realistic than the mean.

Now, the study does use the median income of the zip code when determining the soldiers' income, but then it averages those incomes together, leading to the effect described above. Now comes something even trickier. Re-read this line:

By assigning each recruit the median 1999 household income for his hometown ZIP code as deter­mined from Census 2000, the mean income for 2004 recruits was $43,122 (in 1999 dollars). For 2005 recruits, it was $43,238 (in 1999 dol­lars).... The national median published in Cen­sus 2000 was $41,994.

That's right... they just compared a mean to a median, thus making the soldiers' income more than that other Americans'.