Identifying psychological trauma among Syrian refugee children for early intervention
Analyzing digitized drawings using machine learning
Abstract
Nearly 5.6 million Syrian refugees have been displaced by the country鈥檚 civil war, of which roughly half are children. A digital analysis of features in children鈥檚 drawings potentially represents a rapid, cost-effective, and non-invasive method for collecting information about children鈥檚 mental health. Using data collected from free drawings and self-portraits from 2480 Syrian refugee children in Jordan across two distinct datasets. We use the Least Absolute Shrinkage and Selection Operator (LASSO) machine-learning techniques to understand the relationship between psychological trauma among refugee children and digitally coded features of their drawings. We find that children鈥檚 drawing features retained using LASSO are consistent with historical correlations found between specific drawing features and psychological distress in clinical settings. We then use drawing features within LASSO to predict exposure to violence and refugee integration into host countries, with findings consistent with anticipated associations. Results serve as a proof-of-concept for the potential use of children鈥檚 drawings as a diagnostic tool in human crisis settings.
This is an output of the Gender and Adolescence: Global Evidence (GAGE) programme
Citation
Baird, S., Panlilio, R., Seager, J., Smith, S. and Wydick, B. (2022) 鈥業dentifying psychological trauma among Syrian refugee children for early intervention: Analyzing digitized drawings using machine learning鈥� Journal of Development Economics 156: 102822. https://doi.org/10.1016/j.jdeveco.2022.102822.