The following is a guest post by Harold Pollack, the Helen Ross Professor of Social Service Administration at the University of Chicago.
Greg Mortenson’s sad predicament should chasten the world of philanthropy and nonprofit services. Some of the most basic lessons concern boring mechanical matters such as financial statements and accounting controls. These mechanical niceties are an incredible pain and frustration–right up to the moment they are screwed up. It shouldn’t have taken an investigative reporter to unmask this situation. Many nonprofit boards and nonprofit managers are untrained in this stuff, or they are simply too lax. This invites trouble.
I think there is also a more general market failure–or at least a general market challenge–in the philanthropic sector that is worth noting. If you want to attack a tough problem, it’s often easiest to catch the imagination of funders and the public when you have a visionary, charismatic leader proposing some disruptive innovation. Mortenson is one such leader, but anyone involved in education, environmental advocacy, microcredit, crime control, or global health could identify many others.
I was at a meeting not long ago at which where a foundation leaders challenged the assembled experts: “Find me the next Perry Preschool.” He’s not the only one asking such questions.
Perry Preschool certainly warrants the attention. In the early 1960s, a team in Ypsilanti, Michigan led by David Weikart identified 123 preschoolers at high risk of poor academic and social development. Some had very low IQs. Others had other difficulties that made them likely to be placed into special education, to drop out of school, to face many other bad life outcomes.
Rather than simply waiting for the worst to happen, Weikart’s team designed and fielded a high-quality preschool intervention. They did two other things that changed American social policy. First, they implemented a rigorous randomized trial. Fifty-eight children were randomly assigned to the preschool program; while sixty-five were assigned to a control group. Second, researchers tracked both the treatment and the control groups for decades to see how people fared.
The results were—and are–striking. Perry Preschool kids were more likely to graduate high school and to become gainfully employed. They were less likely to be arrested. They were more likely to have decent earnings at age 40. The social return, per dollar invested, was huge.
Scholars still argue about the proper interpretation of study findings. No one disputes their importance. Perry played a huge part in winning public support for Head Start, and for other early interventions for at-risk children. It exemplified the power of the right randomized trial at the right moment to change public policy. As one policy analyst put it, Perry Preschool subjects may be the most influential 123 children in the history of American social policy.
Like many home runs, Perry was a terrific efficacy trial. You need rigorous small-scale trials of best-practice interventions to establish the upper limits of a promising program model–and just to show people that progress is possible when there are so many reasons for cynicism. We need many more such randomized trials, especially effectiveness trials of interventions that can feasibly be implemented in real-world settings for large numbers of people. That’s a main goal of the University of Chicago Crime Lab, which I help to run.
Without in any way disparaging interventions such as Perry—or other terrific examples such as David Olds’ studies of home visiting–I wonder whether we’re asking the wrong question by putting a heavy premium on finding the next similar big thing. Finding and backing a real breakthrough is sometimes a great way to make progress, sometimes not.
Methodical, low-tech, incremental process improvement exemplified by Atul Gawande’s Checklist manifesto is more typical of organizational improvements in health care that actually stick. Donald Berwick pioneered, with others, the Institute for Health Improvement. Its 100,000 lives campaign saved thousands of people, often through basic quality and safety efforts. Many of the clinical and organizational health innovations discussed on this blog might fit the same description.
Even in cases such as Perry Preschool, where a genuinely powerful, evidence-based approach received experimental support, there remain a host of questions regarding how well the intervention can be brought to scale and sustained, how processes can be improved, cost-effectiveness, proper targeting of services to particular populations, program effectiveness when implemented by typical rather than pioneering staff, and so on. That’s the blocking and tackling that’s really needed to make public policy and field interventions that help millions of people over a long period of time.
Moreover, the search for the next killer app or the next big star makes us vulnerable to fads and, simply, to the characteristic failures of many organizations focused on a single person or a single gripping idea. This is a tough dilemma, because the reality is that you need gripping people and gripping new ideas to excite funders, policymakers, and the public, to attract the resources and support one needs to make things happen.
It’s sad to see the likely unraveling of Greg Mortenson’s efforts. It would be even worse if this led people to be cynical about global development when so much good work deserves support. Whatever the truth behind this specific controversy, we should ask how our continued search for big stars and new breakthroughs sometimes leads us astray.