A wise man proportions his belief to the evidence. - David Hume,
Scottish Historian and Empiricist
by Nick Randell, Program Officer
On September 16, 2014, I completed the last phone survey of an evaluation of a five-year grant initiative at the Tower Foundation. If I haven't lost you already, I'd like to take this blog space to summarize what we learned. I'm keeping this overview at a high level. We had an article published about this evaluation, so if you crave details, email me ([email protected]) and I will send you a copy.
In 2004 The Tower Foundation issued an RFP for non-profits to implement programs with a strong evidence base. There was a pre-selected list of programs to chose from, mostly behavioral health interventions targeting young people and their families. That first year there were eight programs on the list. The RFP was issued in each of the next four years, and that list of eligible programs grew to 38 by the third year, but was "right-sized" to 16 in the fifth and final year of the initiative.
Here's the list of programs that applicants actually chose (and that were funded):
In five years, we funded 25 program implementations for 22 agencies, generally three-year grants. The value of the grants totaled $2.1 million, with an average award of $84,050. We encouraged applicants to develop robust implementation plans that recognized the realities of staff turnover, program certification demands, internal cultural resistance to change, etc.
To assess the effectiveness of all this grantmaking, we developed an evaluation methodology, heavy on phone surveys, that would get at two key issues:
The evaluation included phone interviews at one and two year intervals, post grant period. For each intervention, we developed a metric to reflect fidelity to the model. A higher fidelity score demonstrated attention to things like certification requirements for clinicians, adherence to program scripts/manuals, the correct length and number of sessions, appropriate caseloads, and clinical supervision. We also asked questions to try to identify factors that most contributed to the success or failure of an implementation. We called these "success drivers" and "failure drivers."
Some would question the wisdom of conducting this evaluation with foundation staff, rather than engaging a third-party evaluator. We actually did solicit evaluation proposals for this work. Evaluators actually recommended that we do the evaluation internally, provided we keep the focus on sustained program delivery that maintained some minimum level of adherence to the protocols of the specific intervention. In fact, since staff had done a lot of work to understand just what these protocols entailed, we were in the best position to do this evaluation. To maintain some level of objectivity, I conducted all the phone interviews and metric scoring personally, and was not involved in grant reviews or monitoring activities for any of these grants.
So what did we learn? Two
years post-grant, 15 of 25 programs were still running. Fidelity scores for these continuing programs
placed eight in the "exemplary" range (score 9.0 or higher), five "good"
(7.0-8.9), and three "fair" (5.0-6.9).
Two programs actually improved their ratings across the two post-grant
interviews. Nine held their ground; five
slipped slightly. The table below
captures some of the factors that distinguished successful implementations from
less successful ones.
The initiative as a whole had some pretty clear wins. In several cases, state agencies agreed to reimburse for new intervention models. One intervention in particular, Trauma-Focused Cognitive Behavioral Therapy, achieved significant penetration in Tower regions. 50 clinicians were trained to offer it in eastern Massachusetts, 80 in western New York. Subsequent grantmaking has built on efforts to introduce research-tested social/emotional curricula in pre-k programs. To date, we have supported these implementations in over 150 classrooms.
Photo by Skara Kommun
Flickr: David Hume | 6133856373
Creative Commons 2.0 Licensed