Randomized control trials help scientists know if interventions can influence child development.

Have you ever wondered how researchers determine whether a treatment works as expected? One way to achieve this goal is through intervention research.

Simply put, in intervention research, scientists and medical practitioners give a new treatment (or intervention) to a group of people and measure if the outcome is different from a group who didn’t receive the intervention. For instance, suppose a researcher wants to study the effect of different tools and behaviors on toddlers’ napping time. The intervention would be the specific tools, and researchers would measure the outcome of whether toddlers can get longer and more consistent naps. The research may have some families use certain tools to encourage better naps for toddlers, while other families don’t. In the end, the researcher will examine whether there are significant differences between the two groups of families on the quality of the naps. Often, once the study has concluded, participants who were in the control condition (who didn’t receive the intervention) will receive the intervention later, so everyone who participated in the study has access to any benefits of the intervention! (Because everyone wants longer naps!)

However, another question arises here: Suppose the researcher finds differences between the two groups. How can they know for sure that the new nap tools caused this result? In other words, how can they guarantee that there is no other potential factor causing the differences? For example, some toddlers may have siblings that interrupt their sleep schedules while others don’t, so their naps may be less consistent than others regardless of whether their parents received the intervention. 

A child with brown curly hair wears a lab coat and a stethoscope and leans over a notebook.

Researchers are concerned with these individual differences among the participants. While controlling all of the tiny differences between participants is very hard, these small things may shake our conclusions. Fortunately, researchers have developed a specific study design called the randomized control trial (RCT) to resolve these problems. In an RCT, all participants share some factors in common, such as the diagnosis that is being treated, making them all part of one big population. The participants are then randomly split into two groups. The experimental group receives the intervention. The control group does not. This randomization gives us two similar groups, so the individual differences no longer threaten our study. The idea is that other individual differences will be pretty evenly spread across the intervention group and the control group, so any effects will “wash out” when looking at all the participants together. RCTs come in handy in scientific research because it helps researchers ensure that the intervention is really what explains any difference between the control and experimental groups. 

We can look at a real-life example of an intervention study that used an RCT. In 1999, a researcher named Sally Ward randomly divided a group of young children into two groups. These children were 8-21 months old and all had language delays. One group received an intervention to aid their language development. The control group did not. When they were measured again at three years old, only 5% of children in the intervention experimental groups showed a language delay. But in the control group, 85% of children had a language delay. Because the research used an RCT, the researchers can conclude that their language intervention was effective.

References

Ray, L. (2020). Why is randomization in clinical trials important? Cure Today. doi: https://www.curetoday.com/view/why-is-randomization-in-clinical-trials-important. 

Salkind, N. J. (2010).  Encyclopedia of research design (Vols. 1-0). Thousand Oaks, CA: SAGE Publications, Inc. doi: 10.4135/9781412961288

UNICEF Office of Research-Innocenti (2014). Randomized Controlled Trials (RCTs): Methodological Briefs – Impact Evaluation No. 7, doi:https://www.unicefirc.org/publications/752-randomized-controlled-trials-rcts-methodological-briefs-impactevaluation-no-7.html

Ward S. (1999). An investigation into the effectiveness of an early intervention method for delayed language development in young children. International journal of language & communication disorders, 34(3), 243–264. https://doi.org/10.1080/136828299247405

Cai has black hair and glasses. She wears a collared white shirt with a bow at the collar and a brown paid skirt. She stands against a bright pink wall, holding a rotary phone to her ear.

Caimiao Liu

Author

Caimiao is a sophomore at Duke University studying psychology and sociology. She is interested in how children learn to distinguish the nuances among very similar words through early interactions as well as how they respond differently to multilingual environments. On campus, Caimiao is part of the Duke Chinese Theater, Anti-Resume project, Margins Team, a Japanese tutor, and a Duke PAWS volunteer. She enjoys reading, movies, and all kinds of arts.

Elika Bergelson

PRincipal InvestigatoR