We have invested substantial sums in independent evaluations of each of our five strategic initiatives. In each case, the evaluation grantees were selected by competitive RFPs, and they represented a variety of organizations from universities to private, for-profit corporations. They were charged with providing both the Foundation and our grantees with regular feedback on program development and effectiveness, summarized in an annual progress report. Each evaluation grantee was also expected to deliver a final report at the conclusion of the initiative to address not only the outcomes achieved by individual grantees but also the overall impact of the initiative itself. 

Parenthetically, as a result of our experience with initiative evaluations, we are now clearer on what we expect in the way of a final report. It should be no more than 50-60 pages and written in “plain English” in a user-friendly graphic format. Such a document could then potentially be disseminated to a variety of audiences and stakeholders outside the Foundation itself.

As for grants made outside of our initiatives, we no longer require that each have an independent evaluation (as we did in earlier days), but we do expect them to have explicit objectives and a way of measuring progress toward their attainment. On a couple of occasions, we have also commissioned an outside investigator to conduct a post-hoc “cluster analysis” of a group of grants made for a similar purpose to get a better sense of their overall impact. Such evaluations are low-cost and can be completed in a relatively short period of time. One of the most recent examples of such an analysis is Jennie Schacht’s report on grants made to regional clinic associations and consortia, which is posted in this section of the website. 

“Cluster” evaluations can’t provide “definitive” data, but they can offer reasoned judgments about the effectiveness of our funding. They can also begin to capture more of the “story” of what happens as a result of our grants. Which raises a third question: