The Moving to Opportunity Study

Moving to Opportunity (MTO) is a major randomized housing mobility experiment sponsored by the U.S. Department of Housing and Urban Development (HUD). Starting in 1994, MTO provided 4,600 low-income families with children living in public housing within some of the nation’s most disadvantaged urban neighborhoods the chance to move to private-market housing in much less distressed communities. Families were randomly assigned to one of three groups: a group offered a housing voucher that could only be used to move to a low-poverty neighborhood, a group offered a traditional Section 8 housing voucher, and a control group. 

Publications with findings from the 10-15 year follow-up can be found here: 

Housing Mobility Interventions and Mental Health Disorders in Adolescence

Why concentrated poverty matters

Specal issue devoted to MTO's Long Term Impacts

Neighborhood effects on subjective well-being: evidence from a randomized experiment

Neighborhoods, Obesity and Diabetes—A Randomized Social Experiment


The Next Generation Project

Launched in 1999 with co-Investigators Pamela Morris (New York University), Aletha Huston (University of Texas at Austin), Greg Duncan (University of California, Irvine) and dozens of other researchers, the Next Generation project examined the effects of income, employment and child care on children's developmental outcomes using data from over 8 studies of anti-poverty experiments--testing over 16 different policies-- conducted through the 1990s and early 2000s. 

A few of the key publications can be found here:

How Welfare Policies Affect Child and Adolescent School Performance: Investigating Pathways of Influence with Experimental Data

How Welfare and Work Policies Affect Adolescents

A tale of two methods: Comparing regression and instrumental variables estimates of the effects of preschool child care type on the subsequent externalizing behavior of children in low-income families 

From statistical associations to causation: What developmentalists can learn from instrumental variables techniques coupled with experimental data