The study is the first evaluation of a major ongoing national program that uses the classical experimental design of random assignment, measuring "what would have happened" by comparing people who entered Job Training Partnership Act (JTPA) programs with those who didn't. After background information on JTPA, chapters look at benefit-cost analyses; enrollment; program impacts on target groups; impacts on the earnings of subgroups; and policy implications of the findings. Distributed by University Press of America. Annotation copyright by Book News, Inc., Portland, OR

This book is about making machine learning models and their decisions interpretable. After exploring the concepts of interpretability, you will learn about simple, interpretable models such as decision trees, decision rules and linear regression. Later chapters focus on general model-agnostic methods for interpreting black box models like feature importance and accumulated local effects and explaining individual predictions with Shapley values and LIME. All interpretation methods are explained in depth and discussed critically. How do they work under the hood? What are their strengths and weaknesses? How can their outputs be interpreted? This book will enable you to select and correctly apply the interpretation method that is most suitable for your machine learning project.

Across OECD countries, almost one in every five students does not reach a basic minimum level of skills. This book presents a series of policy recommendations for education systems to help all children succeed.

Radically reimagine our ways of being, learning, and doing Education can be transformed if we eradicate our fixation on big data like standardized test scores as the supreme measure of equity and learning. Instead of the focus being on "fixing" and "filling" academic gaps, we must envision and rebuild the system from the student up—with classrooms, schools and systems built around students’ brilliance, cultural wealth, and intellectual potential. Street data reminds us that what is measurable is not the same as what is valuable and that data can be humanizing, liberatory and healing. By breaking down street data fundamentals: what it is, how to gather it, and how it can complement other forms of data to guide a school or district’s equity journey, Safir and Dugan offer an actionable framework for school transformation. Written for educators and policymakers, this book · Offers fresh ideas and innovative tools to apply immediately · Provides an asset-based model to help educators look for what’s right in our students and communities instead of seeking what’s wrong · Explores a different application of data, from its capacity to help us diagnose root causes of inequity, to its potential to transform learning, and its power to reshape adult culture Now is the time to take an antiracist stance, interrogate our assumptions about knowledge, measurement, and what really matters when it comes to educating young people.

A hearing was held on education, training, and service programs that serve disadvantaged teens. Testimony was presented on recent research findings concerning these programs and on their implementation. The major lessons learned from the Summer Training and Employment (STEP) program were presented, including those of implementation and impact. A second topic was a discussion of the effectiveness of three programs serving teenage mothers on welfare: New Chance (for mothers who have dropped out of school); Learning, Earning, and Parenting (LEAP), an Ohio program for teen parents; and Demonstrations of Innovative Approaches to Reduce Welfare Dependency among Teen Parents. A third topic was the discussion of the work provisions of the Family Support Act of 1988. The following witnesses addressed the hearing: (1) Milton J. Little, Manpower Demonstration Research Corp.; (2) Deanna Phelps, Maryland Department of Human Resources; (3) Alan M. Hershey, Mathematica Policy Research, Inc.; (4) Kevin W. Concannon, Oregon Department of Human Resources; and (5) Michael A. Bailin and Frances Vilella-Velez, Public/Private Ventures. Three submissions for the record are included. (SLD)

Many Americans seem convinced that government programs designed to help the poor have failed. Social Programs That Work shows that this is not true. Many programs have demonstrably improved the lives of people trapped at the bottom of the social and economic ladder. Social Programs That Work provides an in-depth look at some of the nation's best interventions over the past few decades, and considers their potential for national expansion. Examined here are programs designed to improve children's reading skills, curb juvenile delinquency and substance abuse, and move people off welfare into the workforce. Each contributor discusses the design and implementation of a particular program, and assesses how well particular goals were met. Among the critical issues addressed: Are good results permanent, or do they fade over time? Can they be replicated successfully under varied conditions? Are programs cost effective, and if so are the benefits seen immediately or only over the long term? How can public support be garnered for a large upfront investment whose returns may not be apparent for years? Some programs discussed in this volume were implemented only on a small, experimental scale, prompting discussion of their viability at the national level. An important concern for social policy is whether one-shot programs can lead to permanent results. Early interventions may be extremely effective at reducing future criminal behavior, as shown by the results of the High/Scope Perry preschool program. Evidence from the Life Skills Training Program suggests that a combination of initial intervention and occasional booster sessions can be an inexpensive and successful approach to reducing adolescent substance abuse. Social Programs That Work also acknowledges that simply placing welfare recipients in jobs isn't enough; they will also need long-term support to maintain those jobs. The successes and failures of social policy over the last thirty-five years have given us valuable feedback about the design of successful social policy. Social Programs That Work represents a landmark attempt to use social science criteria to identify and strengthen the programs most likely to make a real difference in addressing the nation's social ills.