For nearly a decade now, “research” has been the buzz word in education reform.  It comes in many flavors, and it usually comes with a number of adjectives — scientifically based, high quality, effective, squishy, and such.  And by now we all know that “scientifically based research” is in the NCLB law more than 100 times.

With all of the talk about research, we know there is good research and there is not so good research.  We have action research passed off as longitudinal.  We have customer satisfaction studies passed off as randomized trials. We have people mis-using, mis-appropriating, and downright abusing the word “research.”

Through it all (at least for the past seven years or so), the U.S. Department of Education was supposed to be the arbiter between good and bad research.  IES was founded to serve as the final, most official word on what constitutes good education research.  Dollars have been realigned.  Programs have been thoroughly examined.  Priorities have been shaken up.

So where does it all leave us?  In this morning’s Washington Post, EdSec Margaret Spellings launches a passionate defense of the DC voucher program.  http://www.washingtonpost.com/wp-dyn/content/article/2008/07/07/AR2008070702216.html  (Personally, I’m still waiting for such a defense of Reading First, a program helping millions upon millions of more students in schools beyond our nation’s capital, but what can you do?)

It should come as no surprise that Spellings sought to use research to demonstrate the effectiveness and the need for the DC voucher program.  Without doubt, vouchers have had a real impact on the District of Columbia.  It has reinforced the importance of education with many families.  It has opened doors of schools previously closed off to DC residents.  It has forced DC public schools and charters to do a better job, as they seek to keep DC students (and the dollars associated with their enrollment) in the DCPS coffers.  And, of course, we are starting to see the impact vouchers are having on student achievement among students who previously attended the most struggling of struggling schools.

Spellings points out all of this in her detailing of the research validating the voucher program.  But there is one “research” point Spellings uses that just has Eduflack scratching his head.  From the EdSec’s piece — “The Institute of Education Sciences (IES) found that parents of scholarship children express confidence that they will be better educated and even safer in their new schools.”
 
Such a statement is downright funny, and quite a bit concerning.  In all of the discussions about scientifically based research, high-quality research, the medical model, double-blind studies, control groups, and the like, I don’t remember public opinion surveys meeting the IES standard for high-quality research.  Parents feel better about their children because of vouchers?  That’s a reason to direct millions in federal funding to the program? 

Don’t get me wrong.  I’m all for public opinion polling and the value of such surveys (along with the focus groups and other qualitative research that helps educate them).  But it is one of the last things that should be used to validate a program or drive government spending on educational priorities.

If DC is to keep vouchers, it should keep them because it is driving improvement in student performance and is giving a real chance to kids previously in hopeless situations.  It should be saved with real data that bears a resemblance to the scientifically based research we demand of the our programs and that we expect our SEAs and LEAs to use in decisionmaking.  It should be actionable research, with a clear methodology that can be replicated.
 
Otherwise, we’re just wrapping up opinion in a research wrapper.  That may be good enough for some for-profit education companies and others trying to turn a quick buck on available federal resources, but it shouldn’t make the cut for the government — particularly the branch of ED that is in charge of high-quality research.  Ed reform should be more than a finger-in-the-wind experiment.  And Spellings and IES should know that by now.