A Closer Look Inside RWJF's Investment in the Fragile Families Initiative

An Interview with the Program Officer, Laura C. Leviton
Laura Leviton

Senior adviser for evaluation and program officer of RWJF’s Fragile Families and Child Wellbeing Study, Laura C. Leviton, shares her thoughts about the importance of the study for social research and social scientists.

RWJF augmented NIH funding for this highly ambitious study. Directed by a research team based at Princeton and Columbia universities, Fragile Families focuses on the causes and consequences of nonmarital childbearing. The study started in 1998 and is ongoing.

The researchers have sought to fill an information void about this growing group of at-risk parents and their children—what the team termed “fragile families.”

The team conducted guided interviews with approximately 5,000 parents in 20 large U.S. cities when their babies were first born, and tracked the families through follow-up interviews when the children were one, three, five, and nine years old. Approximately three-quarters of the couples were unmarried at the time of birth; the married one-quarter served as a control group.

[This interview was conducted by Michael Brown.]

The Fragile Families and Child Wellbeing Study is now in its 17th year.  To what extent is that kind of longevity unusual in the social research field?

It’s pretty unusual. In the area of child development, it’s probably less than 1 percent. But these studies tend to be the most important and influential for a whole range of reasons. Clearly when you have that kind of longitudinal information on kids, you can learn a lot more about their development, their trajectories in life, their life chances.

Beyond how long the study has gone on, is there anything in the Fragile Families study methodology that, as a social researcher, you consider particularly noteworthy?

I think there are several things. First of all, the ability to engage the families for this long is really amazing. That the lowest response rate was above 50 percent for the fathers is astonishing. So that’s one thing—the care they have taken to make sure they can get back in touch with people.

The second amazing thing is their ability to cobble together from the various funding sources such an incredibly large and rich dataset. They had many funders because everybody saw value in doing this. I’m just really pleased that although the Robert Wood Johnson Foundation portion was around 10 percent, it helped stimulate this kind of stability and continued excellence for the study.

And the third thing was the (study team’s) willingness to give complete and early access to the data set to many, many other researchers outside their circle—giving a rich source of insights into unmarried couples and their children. Their willingness to share this is, in my experience, unparalleled in social science. (The whole data set is available at the University of Michigan Institute for Social Research, Inter-university Consortium for Political and Social Research.)

You had various investigators working with different parts of this dataset. Both Sara McLanahan and Irvin Garfinkel have been enormously generous with their time and effort to make sure those sub-studies were as good as they could be.

Just among RWJF research grantees, we’ve had people looking at [the data on] incarcerated parents returning home, looking at substance abusing mothers and whether they are at risk for domestic violence, and many other issues. It’s an incredible source of information. (See sidebar Related Readings by RWJF-funded researchers.)

To date, what is the study’s key contribution to what is known about fragile families—the causes of out-of-wedlock births and the risks that unmarried parents and their children face? Do you see certain findings that stand out as particularly noteworthy?

Yes—the central theme that out-of-wedlock births put the kids at tremendous risk, and arguably the parents, too. Their life resources are going to be less; their life chances are going to be less. It’s just a bad idea. We knew that before, but we didn’t know the extent to which it is true. And there are policies that can address that without impinging on people’s freedom.

The study team’s online fact sheet says:

"In conclusion, children born to unmarried parents are disadvantaged relative to children born to married parents in terms of parental capabilities and family stability. Additionally, parents’ marital status at the time of a child birth is a good predictor of longer-term family stability and complexity, both of which influence children’s longer-term well-being."

As a layperson, that conclusion strikes me as not unexpected. Has the study essentially provided evidence for what were common assumptions about fragile families? Or has it turned up findings that point in new and unexpected directions?

You’re right. Your grandmother could have told you that marriage is a good thing if you’re going to have kids. I think there was a lot of ignorance, though, about exactly what those consequences were. And I don’t think we knew the range of consequences, the types of consequences.

Some of the things that are almost side issues but that this dataset points to—we had no idea they were so incredibly big. Like the fact that half the fathers were incarcerated is really spectacularly grave and important. We’ve got to do something on behalf of the families that are affected by that, whether married or unmarried. Because, as the authors point out, it has such devastating consequences for the whole family. They didn’t commit the crime.

We’re spending all of this money on prisons, while at the same time we’re depriving family members of a major source of resources and support. That really sticks out for me.

RWJF was an early and—according to McLanahan—critical funder of the Fragile Families study. But she and her team went on to collect millions more from others sources, most notably NIH. Are there lessons here that RWJF and other philanthropies can learn about catalyzing ambitious social research?

You bet. This experience has been a real inspiration to me to catalyze things in exactly that way. I try to tempt people with, “Let’s fund collection of the baseline information.” Then, you demonstrate the feasibility of the approach, and if it’s a good research question, other funders won’t be able to resist (supporting the follow-up work).

That has worked particularly with federal funders. For example, we have under way what is arguably one of the best, most comprehensive, and most rigorous studies of the effects of children’s environment on their eating, physical activity, and healthy weight status going on here in New Jersey. And that emerged out of a Rutgers baseline study of five communities in New Jersey that we funded for our New Jersey Partnership for Healthy Kids (NJPHK). (This is an RWJF program that empowers community partnerships across the state to implement strategies to prevent childhood obesity. Coalitions in Camden, New Brunswick, Newark, Trenton, and Vineland are leading these efforts.)

Rutgers collected the baseline information, and developed some really cool things, like an index of kids’ exposures to conditions that affect their ability to be physically active or eat a healthy diet. Then, they took that to the National Institute of Child Health and Human Development, and NICHD just absolutely grabbed it because they were assured of the feasibility, and the baseline data were already collected.

So in a very, very intense competitive environment for NIH funds, this thing has been funded through 2017. And I think it’s going to provide the best information that we have on what matters for kids’ healthy weight.

Thank you very much.

Can I say one other thing? We talked about why the Fragile Families study is so remarkable: the length of time, the amount of investment, and the huge number of studies that can be done with this dataset, with investigators from all over the country participating. This is really unusual in the social sciences, and we could do a lot more of it.

You know the physicists don’t run out and build their own particle accelerators. They collaborate and share time on a state-of-the-art particle accelerator. Well, we could do a lot more like that in the social sciences with data collection.

I mean, everybody is running around doing their own little survey, wearing out the patience of everybody in sight at tremendous expense. Data collection is the big-ticket item for any social science study, unless you’re using administrative information that’s already collected. The Fragile Families study, although it cost so much, was much more cost effective.

I’m not a social researcher, but I would think one of the problems would be that researchers have their own views of what the protocols should be to ensure good data.

To some degree that’s true, but there is a lot less of that than you would think. Sure, every social scientist, I suppose, wants to implement their own thing. But you know what? You can’t always have what you want—particularly if it demands other people’s time and resources.  


Recommended Reading

The Challenge of Causal Inference in Gene-Environment Interaction Research

October 1, 2013 | Journal Article

Combining large scale social sciences and their research methods with genetics has the potential to better understand and explain issues regarding public health concerns.

Read more
Male students take part in a team-building exercise.

Another Hispanic Paradox

December 1, 2006 | Journal Article

This study assessed whether the few findings to date suggesting weak relationships between education and health-related variables among Hispanics are indicative of a more widespread pattern.

Read more
Girl getting nebulizer treatment

Maternal Intimate Partner Violence and Increased Asthma Incidence in Children

March 1, 2009 | Journal Article

This article examines the relationship between asthma onset in children and maternal intimate partner violence. Stress and early life adversity are thought to influence children's susceptibility to asthma and other immune diseases.

Read more