Education - some numbers

A couple of weeks ago I wrote questioning the conventional 'wisdom' of current practices.  And I was rightly questioned about unsubstantiated claims regarding private market based education vs public education.  In an effort to redress the lack of evidence or data I have been searching for the past week for hard facts on the subject.  I have to admit though, that I have neither the resources nor inclination to unearth more and so am left with an incomplete understanding.  But as zefrank pointed out, thats the way to start.  So here is what I found-

In Milwaukee (Hoxby, 2001) the gov spent 11,000k per public school student.  They offered an 8k voucher for children to go to private schools.  Test scores in public schools went up 8% in math, reading, and science.  And more money was left for public schools.  

Adam Smith spent a whole chapter in 'Wealth of Nations' giving examples of the costs of socializing education.  He argued that it cheapened the worth of the product (learning) and allowed inferior educators into the system causing a downward spiral.  

High school science scores decreased following the push for science education in public schools in the 1960's.  There is at least a small correlation.

Levitt of Freakonomics showed that academic achievement was correlated with the number of books in a household, when a kid's mom had her first child, the socioeconomic status of the parents, and the desire to change schools (notice not the actual attainment of that goal).  This would imply a more deterministic view.  No matter what, those destined for achievement will rise to the top.  Interesting side note - a recent longitudinal study in Wisconsin found each extra close friend by high-school friends correlated with 2% higher earnings later in life.  We'll be rich!

Lubienski 2008, adjusting for demographics found no difference in 4-8th grade test scores amongst private and public school kids.  The adjustment of data is questionable, but interesting to note

Lips and Feinberg of the Heritage Foundation 2006, studied the DC voucher program and found that parents of students involved became more attentive, public school scores improved, and african americans were helped the most.

Helen Ladd of Duke, 2003, argues that there is no data to show new school openings due to vouchers.  

Susan Choy 1997, found that private school teachers were paid less, had less credentials, taught more advanced classes, and were happier and more satisfied.  

Robert Costrell 2009 found that Milwaukee saved $32 million dollars because of the voucher program last year alone

Brandl 1998 found that per pupil government spending, inflation adjusted, has quintupled since 1950 while test scores have been stagnant.  

Wolf and Hoople 2006, tried to figure out what caused the improvements in African-American test scores in the DC voucher program.  Surprisingly the most important variable was that the schools where voucher kids went and excelled had fewer facilities and optional programs.  The authors did not anticipate this result and postulated several reasons - it created an 'all for one' atmosphere and focused on education rather than sports or other activities.

I apologize for the exceeding length of this post but I felt it was necessary to understand the current evidence before discussing improvements.  The articles and data were collected with the help of EBSCO.  Also, I tried to avoid any Cato commentary to avoid the semblance of bias.