How to scale up successful programs

When I first started learning about youth development, I was astounded by the number of programs and interventions that claim to improve outcomes for children and youth. The Child Trends What Works LINKS (Lifecourse Interventions to Nurture Kids Successfully) database contains information on hundreds of rigorously evaluated programs. I would quickly learn, however, that not all of these programs are effective. In fact, it can be quite difficult to develop a program that has concrete, measurable impacts on children and youth. For example, in a recent LINKS synthesis, 14 of the 50 programs evaluated were not found to positively impact their target outcome of reducing problem behaviors in young children.

An effective program, then, can be a gem. Getting it out there—across the country or the world, in the communities where it could make a difference—is the ultimate goal of any program developer. We call this “scale-up,” and it comes in many forms. Schools or organizations may seek to replicate evidence-based programs in their districts or communities. Program developers may package a program and sell it to organizations alongside (or without) added services such as training or technical assistance. Finally, intermediaries such as Communities that Care or PROSPER can work with communities to select an evidence-based program focused on their unique needs and to support the implementation of that program.

While these scale-up models are diverse, a new Child Trends brief identifies the common factors of successful programs that can be scaled-up for expansion in size or new geographic areas. Based on a review of research and frameworks related to scale-up, the brief makes several recommendations for steps that program developers and implementers should take to support a program as it is replicated. One of the major recommendations of the brief is proactively planning for scale-up. Child Trends is currently developing a curriculum, El Camino, to promote goal-setting and reduce teen pregnancy among Latino teenagers, and is pilot testing the program at a high school in Washington, DC. Throughout our work on El Camino, I have seen our program developers incorporate considerations of scaling and replication into the earliest program development stages.

But even when that proactive planning is in place, it’s common that widespread, real-world implementation might not elicit the same supports or resources that were put in place during an evaluation. Findings from a forthcoming article by Robert Slavin and Alan Cheung, researchers at the Johns Hopkins Center for Research and Reform in Education, indicate that studies with larger sample sizes find smaller effect sizes for the programs being delivered. They identify several possible explanations for this phenomenon, one being that there is often more oversight and assistance in program implementation in a small study.

In developing El Camino, Child Trends is already building facilitator supports into the program materials. For example, we provide facilitators with options to modify activities if the program area is physically smaller than the one we used to test the activities. Even in these early stages of program development, we are keeping an eye on how things could look in the “real world.”

Just as all classrooms are different, no two facilitators will be exactly alike. For pregnancy prevention curricula set in schools, such as El Camino, it is crucial that developers understand that facilitators will have different levels of background knowledge or comfort with communicating reproductive health topics with adolescents. Developers should ensure that program materials bring facilitators “up to speed” on the critical background information of the topics they discuss.

Also, accommodations should be built in so facilitators do not feel the need to skip or drastically modify activities they are not comfortable with. For example, one activity in El Camino describes a situation that could escalate into a sexual encounter. If the program is being facilitated by school teachers, they may not feel comfortable narrating this to their students. The program materials therefore provide guidance on how the activity could be delivered through individual reading.

Developing an effective youth-serving program is a difficult undertaking, but scaling it up poses its own challenges. Because of this, program developers should understand the importance of providing all necessary supports to implementers, a process that can and should be intertwined with curriculum writing itself.

Of course, planning for scale-up is only the first step in supporting effective scaling and replication of a program. Developers, purveyors, implementers, and funders all take on new responsibilities during scale-up, including identifying core program components, creating an infrastructure for training and technical assistance, and making sure that new implementations take place where there is a strong fit between community needs and program impacts. I, and the rest of the team at Child Trends, hope to see El Camino in wide implementation one day— and we know that achieving that goal begins with the planning and pilot testing work we are doing today.