Skip Main Navigation
Page Content

Save This Event

Event Saved

A Data Science Approach to Cheaper, Faster, Better Impact Evaluation

Data Analysts for Social Good

Tuesday, October 24, 2017 from 1:00 PM to 2:00 PM (CDT)

A Data Science Approach to Cheaper, Faster, Better...

Ticket Information

Ticket Type Sales End Price Fee Quantity
Reserve your seat Nov 13, 2017 $25.00 $2.37

Who's Going

Loading your connections...

Share A Data Science Approach to Cheaper, Faster, Better Impact Evaluation

Event Details

Speaker Info:

Pete York is a Principal Associate at Community Science, bringing over 23 years of expertise and field leadership in research and evaluation, including more recently the use of predictive, prescriptive, and evaluative machine learning algorithms for large public and nonprofit administrative datasets. Over the past 5+ years he has developed evaluation, predictive and prescriptive models using the administrative datasets and applying machine learning algorithms for agencies and organizations like the Florida Department of Juvenile Justice, Broward County (Florida) Sheriff’s Office’s Child Welfare Unit, and over a dozen administrative data modeling projects with large, scaling nonprofits providing adult justice, workforce development, mental health, economic development, child welfare, or youth development programs and services. Pete has published peer-reviewed articles and book chapters on the use of machine learning algorithms with administrative data to build predictive, prescriptive and rigorous evaluation models, including in leading journals like Children and Youth Services Review. Through these projects, he has developed deep insights into the life cycle of administrate data use, from access and integration of datasets from multiple sources to analysis and use. Pete is a long-time leader, presenter, speaker and author on the use of evaluation and data for social impact, and is a Leap Ambassador, which is a private community of leaders of nonprofit and civic leaders, funders, and public servants who share a core commitment to high performance.   

Event Details:

More and more social impact funders and investors are requiring rigorous experimental proof that a program causes improvements in social outcomes.  Rigorous evaluations, ideally using a randomized controlled trial (RCT) experiment or some other form of matched comparison group study, have become the gold standard for meeting the "impact" requirement needed for investment. However, for practical, cost, timing and/or ethical reasons, many effective social programs cannot conduct these types of evaluations. This session will share and provide case studies of an equally rigorous and more applicable method using program administrative data and training machine learning algorithms to approximate an RCT. This session will also present how this method improves front line case- and/or situation-specific decision making in ways not possible using traditional evaluation methods. These alternative data-driven evaluation techniques have the potential to release a growing pool of impact funding dollars that are waiting for “investable” programs that can prove predictable social and financial returns, with the added value of providing real-time feedback loops to decision makers from the front lines to the board room.

Have questions about A Data Science Approach to Cheaper, Faster, Better Impact Evaluation? Contact Data Analysts for Social Good

Save This Event

Event Saved

When

Tuesday, October 24, 2017 from 1:00 PM to 2:00 PM (CDT)


  Add to my calendar

Organizer

Data Analysts for Social Good

Data Analysts for Social Good equips, trains, and gathers nonprofit professionals with the goal of equipping them to use data more effectively. To learn more visit our website or on LinkedIn.

  Contact the Organizer
A Data Science Approach to Cheaper, Faster, Better Impact Evaluation
Class Charity & Causes

Please log in or sign up

In order to purchase these tickets in installments, you'll need an Eventbrite account. Log in or sign up for a free account to continue.