In this blogpost, Mobarak Hossain addresses the relationship between large-scale data collection in education through the Education Management Information System (EMIS) and national learning assessments. Among other findings, he shows that while EMIS can enhance data capacity for governments, it can also overlook qualitative and case-specific issues critical to improving education systems, particularly for marginalized populations.
The World Bank has played an influential role in reshaping educational governance in low- and middle-income countries (LMICs). Over the past three decades, the focus of education management and system reforms, facilitated through development assistance, has increasingly shifted towards strengthening data-driven governance. This transition is exemplified by the widespread implementation of the Education Management Information System (EMIS), a comprehensive system designed for the collection, analysis, dissemination, and utilization of education-related data (Abdul-Hamid et al., 2017). EMIS functions as an annual school census, involving three core stages: the gathering of basic statistics about demographics; the collection of advanced data, including details about students’ socioeconomic backgrounds; and the utilization of data to inform policies. The primary goal of EMIS is to furnish countries with systematic and high-quality data, empowering them to formulate and implement evidence-based education policies, with the ultimate aim of enhancing learning achievements (Abdul-Hamid, 2014).
Although the primary goal of EMIS is to enhance learning achievement, there is a lack of evidence regarding the extent to which EMIS has effectively prompted the regular implementation of national learning assessments across diverse subject areas. In a recently published study in International Journal of Educational Development, I investigate this relationship.
Assessments here refer to national learning assessments administered by national governments, utilizing standards or progress-based guidelines such as the Early Grade Mathematics Assessment (EGMA) or Early Grade Reading Assessment (EGRA). These are different from norm-referenced and low-stakes international large-scale assessments like the Program for International Student Assessment (PISA) and high-stakes public examinations, such as national leaving examinations (Benavot & Tanner, 2008).
Results: My analysis reveals a significant association between the implementation of an EMIS project by the World Bank and the conducting of approximately one additional national learning assessment across all educational levels. This finding remains consistently robust across various regression models.
Mechanisms: Firstly, while national assessments may not have been the primary focus of the World Bank through EMIS projects, they appear to be a consequential outcome. The overarching goal of EMIS extends beyond enhancing the education system’s quality to improving learning achievement, serving as a tool for informed decision-making at various levels (Abdul-Hamid et al., 2017).
Secondly, despite considerable progress in global school attendance since the 1990s, LMICs still face challenges in achieving minimum learning proficiency levels (Oketch et al., 2021; Pritchett & Viarengo, 2023). International organizations increasingly focus on tracking learning achievement progress, and national assessments have gained importance in this regard. Increased World Bank development assistance in EMIS would equip countries with technical know-how and improved data infrastructure, enabling continuous assessments. The effective management of data for an entire education system requires the expertise to gather, store, and handle the data (Jin et al., 2015). Given that comprehensive nationwide learning assessments demand such capabilities, it is plausible that the World Bank’s EMIS initiatives have spurred the development of these capacities.
The pursuit of evidence-based policymaking in the global context has heightened, with a growing emphasis on data generation and the development of robust data infrastructure, particularly within developing nations, including the education sector (Tennant & Clayton, 2010). Despite the growing push by international organizations like UNICEF’s Data Must Speak division (e.g., UNICEF Innocenti, 2022) to utilize EMIS data to inform policies, actual evidence of its widespread usage is yet to be seen globally. The following discussion underscores the challenges in translating data into actionable policies and the ongoing debate on the effectiveness of EMIS in achieving beyond large-scale data gathering.
Firstly, large-scale data collection through EMIS can enhance data capacity for governments, while it can arguably overlook qualitative and case-specific issues critical to improving education systems, particularly for marginalized populations. This underlines the need for a balanced approach, integrating quantitative data with qualitative and critical evidence.
Secondly, there is a risk that large-scale quantitative data may be misused by various actors who may highlight achievements based on broad indicators primarily focusing on dominant population groups, thus potentially neglecting the accurate representation of the status of marginalized communities.
Thirdly, while technical expertise and financial support from donors can be pivotal in various contexts, ensuring the sustainability and adaptability of these initiatives requires a shift towards developing local expertise. Establishing regular budgets and customizing tools to align with local cultures and needs becomes imperative to guarantee long-term sustainability.
About the Author:
Mobarak Hossain is an Assistant Professor in the Department of Social Policy at the London School of Economics and Political Science. His research interests include comparative education policies and institutions, inequalities, sociology of education, and cross-border policy diffusion, with a particular (but not exclusive) focus on emerging and developing economies.