Exploring IDELA’s Sensitivity to Observed Classroom Quality in a Humanitarian Context

• In Event: Building the evidence base for pre-primary learning: findings from Lao PDR, Bangladesh, Cambodia, Lebanon, and Niger

Proposal

Learning in the early years is essential for children’s development. Yet research from many countries indicates that the quality of early education is often low (Yoshikawa, Wuermli, Raikes, Kim, & Kabay, 2018). For countries to monitor their progress in relation to SDG 4.2, they have to be equipped with reliable and valid tools to monitor quality of education and learning outcomes (Fernald, Prado, & Raikes, 2017).

Understanding quality of education and early learning is particularly urgent in humanitarian contexts such as the Syrian crisis. To date, more than 5.6 million Syrians have fled the country (UNHCR, 2018). Yet the evidence base on effective programs to support early learning in emergencies is scarce (Murphy, Yoshikawa, & Wuermli, 2018). Furthermore, few studies use direct assessments of child development and rely on parent or caregiver report which have been shown to be biased. Using data from a direct assessment and classroom observation of children attending the International Rescue Committee’s Preschool Healing Classroom program in Lebanon, this paper aims to answer two research questions:

1. How can we meaningfully measure early development and learning in Syrian refugee children?

2. Can observed classroom quality predict children’s early development and learning outcomes?

To respond to the aforementioned questions, this paper focuses on data collected from Syrian refugee children (Mean=3.83 years, SD=0.74; N =200). These children participated in a five-day early childhood education program over the course of 6-8 months. Children were assessed using the International Development and Early Learning Assessment (IDELA) tool developed by Save the Children. Classroom quality was assessed using the Teacher Classroom Observation (TCO) tool which was based on the classroom quality measure of MELQO.

The IDELA tool has been administered in over 50 low- and middle-income countries (LMICS) (Pisani, Borisova, & Dowd, 2018) and has shown to uphold the preconceived conceptual structure (Wolf, Halpin, Yoshikawa, Pisani, Dowd, & Borisova, 2017). The IDELA tool is hypothesized to have five main domains: Numeracy (7 tasks, 40 items), Literacy (6 tasks, 34 items), Social/Emotional Development (4 tasks, 10 items), Motor Development (4 tasks, 12 items) and Executive Function (7 tasks, 24 items).

We performed a confirmatory factor analysis for the IDELA to confirm the factor structure with respect to the five hypothesized domains. The model fit indices for the five domains at baseline, represented a goodness of fit: Literacy (RMSEA=0.06; CFI/TLI = 0.94/0.93); Numeracy (RMSEA=0.39; CFI/TLI = 0.882/0.873); Motor Development (RMSEA=0.09; CFI/TLI = 0.97/ 0.96); Social-Emotional (RMSEA=0.054; CFI/TLI = 0.96/ 0.95). We ran into some computational errors when exploring the Execute Function domain but we will investigate this further.

Using factor analytic methods, we will investigate the structure of the TCO. We will then investigate the relations between aspects of classroom quality and scores on the IDELA.

Authors

• Joyce Rafla, New York University

• Anaga Ramachandran, NYU

• Hirokazu Yoshikawa, New York University

• Alice Wuermli, New York University

• Maria del Sol Prieto, International Rescue Committee Lebanon

• Katie Maeve Murphy, International Rescue Committee