SERAIS: Social Emotional Response and Information Scenarios Evidence on Construct Validity, Measurement Invariance, and Reliability in use with Syrian Refugee Children in Lebanon
Demonstrating that social-emotional learning (SEL) programs lead to improvements in children’s social-emotional skills requires the use of measures that provide accurate data capturing meaningful changes in children’s development over time. In contexts affected by crisis and conflict, few measures have the evidence required to support their use in program evaluations, limiting stakeholders’ ability to determine whether programs are working, how, and for whom. The Social Emotional Response and Information Scenarios (SERAIS) holds promise for addressing this gap. SERAIS (“I would” in French) employs a scenario-based format in which children are asked to report what they would do and feel in a variety of different social situations. Responses are designed to capture information about a suite of social, emotional, and cognitive skills among elementary school-aged children in fragile, conflict-affected settings. The measure was tested in Lebanon in school year 2017-18 with a sample of 3,661 Syrian refugee children (ages 5-16) who were enrolled in Lebanese formal schools and had access to IRC programming in the Bekaa and Akkar regions of Lebanon. Evidence on the psychometric properties of this version of SERAIS support its use as an outcome measure in program evaluations and in research with Syrian refugee children in Lebanon. Specifically, we provide evidence that SERAIS assesses key social and emotional skills reliably. We also provide evidence that the measure functions and is understood in the same way by children with access to SEL programming and those without, as well as by children at the beginning and the end of the school year. This criteria is known as measurement invariance, and establishing the measurement invariance of an assessment used in a rigorous program impact evaluation enables us to confidently assess whether children’s skills are improving or declining over time – and whether such changes are the result of our SEL programming.
