Today, most k-12 program evaluation is, at best. a single snapshot of what worked - or not - where and when. At worst it’s a political weapon employed in the charter, voucher, privatization, math or reading wars. Typically it’s a less than completely informative marketing tool. Rarely is it employed for the kind practical evaluation described earlier in this series of blog posts.As it happens, post-NAS, RAND’s high profile k-12 evaluation work, much of which has been led by Brian Gill (whose move to Mathematica prompted my thinking about practical evaluation), has offered examples of each. In only one case have I seen any real evidence of the programmatic course corrections “en route” that typified the RAND experience with NAS .
That case is the Pittsburgh study noted last week.
• Edison’s decision to hire RAND in 2000 was surely in part an effort to bolster the firms business credibility and educational legitimacy at a time when both were in doubt. The study itself showed Edison offered school districts roughly the same odds of turning around failing schools as the district itself, at a marginally higher cost. I have seen nothing to suggest how or if Edison used the RAND study to improve its school management services. In the absence of any serious explanation by the firm, I succumb to the temptation of reminding readers of Korean automotive manufacturer Hyndai’s new ad: “Are car companies committed to quality, or to the phrase ‘committed to quality.’”
• RAND’s work on Philadelphia school turn around efforts - also showing no huge difference between schools managed by outsider and those reorganized by the districts. That report became fodder for a methodology debate between RAND and Paul Peterson that simply obscured the bottom line - school management is no silver bullet, it’s just another tool we’re still honing. Whatever role the study may have played in the City of Brotherly Love’s decisions on school management contract, it surely was not the centerpiece of a dispassionate discussion among stakeholders on how to improve their joint enterprise.
Evaluation is a scorecard. Buyers need some means of sorting through their choices, and imperfect means are better than none. But it’s not merely a scorecard, and it’s not just the program that’s being scored - it’s also all the other stakeholders whose efforts make for success or failure. The “scorecard function is not even the most important. Practical evaluation is a means of continuous improvement for the entire system where the school improvement program is introduced. It’s basic purpose is to provide the fact base for ongoing conversations between buyers and sellers and every other group engaged in implementation.
I’m not singling out RAND or Gill for special treatment. They are just examples of a widespread unhelpful social phenomena in the use of evaluation - one the (not so) new (but slow to get started) Society for Research on Education Effectiveness should address as its first priority (but doesn’t).
For more on the related subject of adaptive management.
