The University of Arkansas issued a study claiming that charter schools get a higher “return on investment” than public schools, yet are underfunded especially given their great “productivity” and “ROI.” (I admit I stumble over the idea of applying ROI when we are talking about education and children, but that’s just me.)

Bruce Baker of Rutgers University analyzes the University of Arkansas study and takes it apart.

Baker shows that the Arkansas study “shamelessly” and “knowingly” uses bogus data. The Arkansas study is meant to refute an earlier critique of their work by Baker.

Here is what Baker concludes:

“The acknowledgement of my critique, highly selective misrepresentation of my critique, and complete failure to respond to the major substantive points of that critique display a baffling degree of arrogance and complete disregard for legitimate research.

“Yes – that’s right – either this is an egregious display of complete ignorance and methodological ineptitude, or this new report is a blatant and intentional misrepresentation of data. So which is it? I’m inclined to believe the latter, but I guess either is possible.

“Oh… and separately, in this earlier report, Kevin Welner and I discuss appropriate methods for evaluating relative efficiency (the appropriate framework for such comparisons)…. And to no surprise the methods in this new UARK report regarding relative efficiency are also complete junk. Put simply, and perhaps I’ll get to more detail at a later point, a simple “dollars per NAEP score” comparison, or the silly ROI method used in their report are entirely insufficient (especially as some state aggregate endeavor???).

“And it doesn’t take too much of a literature search to turn up the rather large body of literature on relative efficiency analysis in education – and the methodological difficulties in estimating relative efficiency. So, even setting aside the fact that the spending measures in this study are complete junk, the cost effectiveness and ROI approaches used are intellectually flaccid and methodologically ham-fisted.

“But if the measures of inputs suck to begin with, then the methods applied to those measures really don’t matter so much.

“To say this new UARK charter productivity study is built on a foundation of sand would be offensive… to sand.

“And I like sand.”