Earlier this week I made my Modest Proposal for B.B. (Broader, Bolder or is it Buying Bananas?). I noted that Randi Weingarten denounced vouchers as a waste of time despite considerable evidence supporting it, while she embraced the B.B. idea of community schools despite there being absolutely no evidence to support the claim that public schools could improve achievement by expanding their mission to include a host of social services.
Given the lack of evidence for B.B. I generously : ) offered to support a series of large pilot studies of the community schools approach, if Weingarten, Leo Casey, and the B.B. crowd would agree to a similar series of large pilot voucher programs as a way of learning more about both reform strategies. No word yet but perhaps their internet is broken (just try unplugging it and plugging it back in).
Shital Shah from the Coalition for Community Schools, however, sent me a nice note with a link to a report claiming to contain the evidence supporting their approach. After reviewing the report I still see virtually no evidence to give us confidence that public schools can increase student achievement by offering everything from legal assistance to health care.
In Appendix B the report lists 21 studies of the community school approach. Seven of them have no student achievement outcomes. Seven examine student test scores but only make pre/post comparisons without any control group. And another seven have comparison groups but none employ random assignment, regression discontinuity, or another rigorous research design. Four of those seven just compare achievement at schools using the B.B. approach to city or statewide averages. And of the seven studies with some kind of control group, two find null effects, another finds null effects in math but not reading and even then only among schools with “high implementation” of the approach. The quality (and quantity) of the evidence supporting community schools is no greater than what we could find to support the healing power of crystals.
I understand why Randi Weingarten or Leo Casey would be pushing the educational equivalent of crystal healing. Their job is to advocate for the interests of their union, not to make fair and reasonable assessments of research claims. If schools expand their mission to include providing health care and other social services just think of all of the dues-paying nurses and social workers they could add to their rolls.
The greater mystery is why normally tough-minded and rigorous researchers, like Jim Heckman and Diane Ravitch, would sign on to this approach entirely lacking empirical support. Heckman won the Nobel Prize for Economics for crying out loud. But then again Linus Pauling won the Nobel Prize for Chemistry and later became a public advocate for mega doses of vitamin C to cure cancer, another intervention completely unsupported by rigorous evidence.
I’ll repeat that I am not against trying the B.B. community school approach with large pilot programs that are carefully studied. I just can’t see why normally smart people would fully endorse untested approaches while ignoring other interventions, like expanding choice and competition in education, which have considerably more supporting evidence.
(edited for typos)
Posted by Jay P. Greene 
