Evaluation 2018: Speaking Truth to Power

View Printable Version

Personalized evaluation: How evaluation methodology affects payment risk in Pay for Success projects

Session Number: 1467
Track: Social Impact Measurement
Session Type: Expert Lecture
Tags: analytics, BIG DATA, data science, Pay for success, Social impact bonds, Social Impact Measurement
First Author or Discussion Group Leader: Brian K Beachkofski [Managing Director - Third Sector]
Second Author or Discussion Group Leader: Peter York [Principal - BCT Partners]
Time: Nov 01, 2018 (03:45 PM - 04:45 PM)
Room: CC - 10

Audience Level: All Audiences

Session Abstract (150 words): 

The evaluation method for a Pay for Success (PFS or Social Impact Bond) project will affect the timing and amount of success payments. For example, a randomized controlled trial (RCT) takes years and results in a wider range of success payments, while non-experimental evaluations (e.g., rate cards) are quicker and are likely to trigger payments without proving the project “caused” the outcome. Quasi-experimental designs are a good compromise on measuring outcomes, but they take a long time and require advanced counterfactual estimates. In this age of big data, machine learning algorithms can now find counterfactuals within program datasets, evaluate precisely how many beneficiaries achieved attributable outcomes, and discover precisely what works for whom. Further, by more accurately understanding each person's outcomes, we can better price their outcomes, reduce biased selection (creaming), and make funding allocations much less uncertain.


For questions or concerns about your event registration, please contact registration@eval.org or 202-367-1173.

For questions about your account, membership status, or help logging in, please contact info@eval.org.

Cancellation Policy: Refunds less a $50 fee will be granted for requests received in writing prior to 11:59 PM EDT October 8, 2018. Email cancellation requests to registration@eval.org. All refunds are processed after the meeting. After October 8, 2018 all sales are final. For Evaluation 2018, international attendees and presenters who encounter complications due to the international travel environment will have up to 30 days after the event to request a refund and submit appropriate documentation. No administrative fee will apply for the international requests.