Scale-up of Early Childhood Programs Must Be Informed by Research to Benefit Kids
Evidence-based early childhood programs such as prekindergarten, home visiting, and child care have the potential to change children’s lives and trajectories. Commonly cited research also estimates a return on investment of 7 to 13 percent for early childhood programs. However, challenges with scaling up these programs from an original research or evaluation setting to a broader population—and in different circumstances—threaten their ability to realize this potential. A growing body of research aims to identify factors that threaten the successful scaling of programs, as well as strategies to overcome those threats and ensure that children and families get the full range of benefits from early childhood programs.
The early childhood years are a critical time for supporting children’s cognitive and emotional development. From birth to age 3, for example, children’s brains form over 1 million neural connections each second. Children who participate in high-quality programs during their early childhood years have been found to experience positive long-term outcomes such as higher rates of education completion, lower criminal justice system involvement, and lower rates of substance use.
Although research studies have shown great potential, not all early childhood programs show the same positive outcomes when scaled. In other words, what works in a tightly controlled research setting or a specific evaluation setting may not work the same in a real-world setting with outside factors that cannot be as controlled. Recent work, from researchers at the University of Chicago’s TMW Center for Early Learning + Public Health and their collaborators, identifies four key factors that might account for differences between initial research and evaluation outcomes and scaled outcomes.
- Programs implemented at scale (i.e., in the “real world”) may involve children and families who are different from research participants—for example, from different demographic backgrounds or with different motivations for participating in the program.
- A program implemented in a real-world setting will probably have a dramatically different context than a research study or tightly controlled evaluation. For example, staff implementing a small program may have different credentials than the broader workforce that will implement a program at a larger scale.
- The level of evidence required for researchers to publish results in a peer-reviewed journal may not rise to the level of evidence that policymakers need to be confident a program is ready to scale. For example, policymakers might need to see evidence from multiple studies suggesting that a program works before they feel confident about implementing it in their own community.
- Finally, in real-world settings—due to social relationships—program elements may potentially affect systems and communities in ways that researchers did not anticipate in their study. For example, families enrolled in a program may share materials with non-participating families in their neighborhood, leading to broader improved outcomes that would not have been measured in the original study.