Developing culturally appropriate measures of social and emotional learning in Tanzania

• In Event: Contextualizing Social and Emotional Learning: Assessment

Proposal

Education programs in Africa increasingly aim to develop and measure social and emotional competencies. However, assessments are typically adapted from those developed in other continents and are not derived from local perspectives. This process of adaptation is problematic for a number of reasons. Existing assessments may fail to include domains of development that are important for adaptive functioning in the home, community and school in different contexts. Similarly, domains can be conceptualized differently in each context. Even when assessments target appropriate domains, those domains may manifest differently by context; that is, competence in a given domain may be demonstrated by different behaviors from one context to another. The importance of these issues has been demonstrated in the development of culturally grounded assessments of mental health in Africa (Betancourt, Speelman, Onyango, & Bolton, 2009; Betancourt, Yang, Bolton, & Normand, 2014).
However, developing instruments in a new context is not straightforward. Existing assessments of social and emotional learning have been developed through extensive research by developmental psychologists. We cannot hope to replicate such investigations in the short period of time typically available for instrument adaptation. In this paper, we describe and assess a method for developing contextually relevant assessments of SEL. The approach involved careful qualitative study to understand local perceptions of the domains of children social emotional competencies that are important for success at school and life in general. We also examined theory and evidence on differences in the cultural context of children’s development between Tanzania and high-income contexts. We used both theory and evidence to adapt and extend existing SEL frameworks to the Tanzania context.
The study took place in two phases. In the first phase, we conducted focus groups and individual interviews with teachers, parents and students in 4 randomly selected rural primary schools from Mtwara region in Tanzania, 3 of which had recently begun participation in a pre-primary education program. The aim was to understand the social and emotional competencies in early childhood that participants viewed as important for school and for life in general. Compared to existing frameworks of social and emotional competencies, participants placed more emphasis on aspects of social responsibility, for example respect, obedience and being an attentive listener. Individual competencies such as curiosity, self-direction and self-belief were valued more by teachers than parents and seen as most important for success at school. In general, most social and emotional competencies – even individual competencies - were discussed in terms of social relationships.
In the second phase, we developed a questionnaire to administer with teachers and parents. The questionnaire assessed behavioral examples of 13 culturally relevant constructs identified in the qualitative phase of research. The parent questionnaire had 71 items across the 13 domains; the teacher questionnaire had 42 items. We assessed the validity and reliability of the questionnaire in a sample of 512 students from preschool and the first two grades of primary school from Mtwara region Tanzania. For each student, we interviewed one teacher who knew them well and one parent. The data were analyzed using confirmatory factor analyses (CFA) to assess whether the underlying factor structure of the data matched the 13 constructs that we had identified through the qualitative work. We also conducted a cluster analyzed to identify social and emotional profiles of students. In particular, we assessed the hypothesis from the qualitative phase that social responsibility constructs – such as respect and obedience – were distinct from learning-oriented constructs such as curiosity and self-confidence. We assessed the proportions of students that were strong in only one of these areas and those that were strong in neither or both areas.
Findings have implications for how cultural values are taken into account in assessment, curriculum design and parent and community engagement around pre-school education.

Authors

• Matthew Jukes, RTI International

• Kellie Betts, RTI International