For decades, language barriers have hindered researchers' ability to assess child-caregiver relationships in refugee populations - an important component in designing and implementing interventions that promote the healthy development of displaced children.
But a recent study led by Yale's Qusai Khraisha validates a novel observational tool that allows researchers to accurately measure the quality of interactions between young children and their caregivers even if they do not understand the language being spoken.
Khraisha, a postdoctoral associate at the Yale Jackson School of Global Affairs, and his coauthors developed a system called Behavioral Recognition in Displaced Group Exchanges (BRIDGE). It relies on observations of things like tone of voice and body language, instead of words, to evaluate child-caregiver relationships.
For the study, the researchers tested BRIDGE using more than a thousand videos of Arabic-speaking Syrian refugee mothers interacting with their school-age children in Jordan, recorded in two sessions over a four-month period. Observers recruited from around the globe watched videos and used the tool to analyze the mother-child interactions they saw. Even though most of the observers did not understand Arabic, they still reliably judged the quality of the mother-child bonds, the study showed.
"Language barriers have set a hard limit on research with displaced populations," said lead author Khraisha, who has worked with and studied displaced peoples for more than a decade. "Our tool works around this limitation by removing language differences from the equation - and what we can now measure more reliably matters enormously. Research has shown that the quality of the child-caregiver bond predicts outcomes across nearly every domain of a child's life: education, nutrition, health, protection. It influences how they relate to their communities and their relationships in adulthood."
The study, recently published in the journal Child Development, was coauthored by researchers from several fields, including psychology, global health, and biology, representing 11 universities across Europe, the Middle East, and North America.
Prior studies on responsive parenting in refugee settings have primarily relied on self-reported data, often collected by asking caregivers to respond to questionnaires.
"Most of what we know about parenting in refugee contexts comes from self-report surveys, but these tools struggle when literacy levels vary or when questions don't translate well across languages," said study coauthor Kristin Hadfield, director of the Trinity Centre for Global Health, at Trinity College Dublin. "Our observational coding system addresses this gap and can be used even when coders don't speak the family's language."
An additional weakness of self-reporting is that people tend to give socially desirable answers, Khraisha said.
"They'll often tell you what they think you want to hear," he said. "A more precise approach is to watch caregivers interact with their children, which captures behaviors that might be omitted or misrepresented in self-reports."
The new system, BRIDGE, directs observers to note vocal elements - tone, pitch, and volume - and nonverbal body movements, such as gestures, posture, and facial expressions that fall outside the scope of self-reports, the researchers explained.
The system allows researchers to measure two core caregiving behaviors: synchrony - whether the caregiver and child display matching behaviors, emotional states, and biological rhythms - and sensitivity, which concerns the parent's attunement to their child's emotional and cognitive states.
To evaluate BRIDGE, the researchers used 1,092 videos featuring 287 Syrian refugee caregivers and their children, aged 4 to 8 years old, which were drawn from a randomized-controlled trial to evaluate "We Love Reading," a program that seeks to improve child literacy and attitudes about reading. About half the videos depicted the mother-child pairs engaged in play activities and the other half in reading activities. An initial phase included 564 videos, while a follow-up phase four months later comprised 528 videos.
The researchers recruited and trained 18 "coders" from across the globe to evaluate the caregiver-child interactions in the videos using the BRIDGE system. All the coders were female with no experience in observational research. They came from the Bahamas, China, Germany, India, Ireland, Italy, Jordan, Luxembourg, Poland, Spain, Turkey, the United Kingdom, and the United States. (Two of the 18 coders understood Arabic.)
The study found no significant disagreement among the coders in their evaluations of the child-caregiver interactions. When the researchers compared the evaluations from newly recruited coders with those of multiple individuals experienced in observational research, who also had used the BRIDGE system to measure the child-caregiver interactions, they found no significant differences in their assessments, according to the study.
"Now that this tool is validated, we will make it available to researchers who study child development in refugee populations," Khraisha said. "The tool eliminates the need for costly and time-consuming translation work or the recruitment of multilingual coders, which can enable more research into how to improve the lives of children who have experienced the trauma of forced displacement."
