Publications

This paper examines how controlling statistically for baseline covariates (especially pretests) improves the precision of studies that randomize schools to measure the impacts of educational interventions on student achievement.

Early success for this ambitious employment program for public housing residents in Seattle was disrupted by a federal HOPE VI grant to tear down and revitalize the housing development.

This working paper examines employment and earnings over a four-year period for a group of disadvantaged out-of-school youth who entered the Evaluation of the Center for Employment Training (CET) Replication Sites between 1995 and 1999. It assesses the importance of three key factors as barriers to employment: lack of a high school diploma, having children, and having an arrest record.

The Center for Employment Training (CET) in San Jose, California, produced large, positive employment and earnings effects for out-of-school youth in the late 1980s. However, in this replication study, even the highest-fidelity sites did not increase employment or earnings for youth over the 54-month follow-up period, despite short-term positive effects for women.

Welfare caseloads fell, employment increased, and neighborhood conditions improved in Los Angeles during a period of economic growth and welfare reform. However, most welfare recipients still remained poor, the concentration of poverty increased, and those who worked were usually in low-wage jobs without benefits.

Four programs that supplemented the earnings of low-income adults increased employment, earnings, and income — particularly for the most disadvantaged — but these effects generally faded after the programs ended.

While positive effects on most parents’ earnings and income faded after six years, young children in some of the most disadvantaged families were still performing better in school than their counterparts in a control group. And, for the most disadvantaged parents, MFIP seems to have created a lasting “leg up” in the labor market.

First Things First, a comprehensive school reform initiative, increased student achievement in Kansas City, Kansas, the first school district to adopt the reform model. It is not yet clear if First Things First is working in four other school districts in which it has been replicated.

The Opening Doors Demonstration is designed to show how community colleges can help more low-income students remain in school and improve other outcomes, including degree attainment, labor market success, and personal and social well-being.

Opening Doors Learning Communities, a program serving mostly low-income freshmen at Kingsborough Community College in Brooklyn, NY, improved course and test pass rates, particularly in English.

Early results are mixed for Employment Retention and Advancement project programs in four sites, but programs in two sites appear to help some welfare recipients work more steadily and advance to higher-paying jobs.

Talent Development, a high school reform initiative, produced substantial positive effects on attendance, academic course credits earned, tenth-grade promotion, and algebra pass rates for students in very low-performing schools in Philadelphia.

In welfare and employment programs that provide earnings supplements, increased family income plays a key role in improving children’s school achievement.

In a study of over 3,500 women in welfare-to-work programs in three states, child care instability did not appear to be a major cause of employment instability.

Jobs-Plus, an ambitious employment program inside some of the nation’s poorest inner-city public housing developments, markedly increased the earnings of residents in the sites where it was implemented well.

This study suggests that child support can be an important income source and can help welfare recipients move toward self-sufficiency. More generous distribution rules increase payment rates, but many parents still do not understand the distribution rules.

Library-based literacy programs face serious challenges to improving adult students’ participation. This study suggests programs should be prepared to accommodate intermittent participation by adult students and to connect students to social services and other supports.

During the first three years of implementation in six urban schools, The Talent Development Middle School model—an ongoing, whole-school reform initiative—had a positive impact on math achievement for eighth-graders but appeared to produce no systematic improvement in outcomes for seventh-graders.

Community colleges can pursue many strategies for enhancing student services, including offering “one-stop shopping,” which provides students with multiple services at the same time and place.

Seattle Jobs-Plus — part of an MDRC national research demonstration designed to promote employment among public housing residents — succeeded in engaging a majority of residents, many of whom were immigrants from diverse parts of the world, in work-related services or supports.