RECENTLY, Dr Faisal Bari and a colleague published an insightful article in this paper (`Evidence for policy`, Feb 14, 2025), raising a pertinent question: why is the vast amount of data collected by government and non-government organisations routinely ignored in shaping education policy? They pointed to large-scale data efforts like the Annual Status of Education Report (Aser) and the government-run Education Management Information Systems (Emis), which consistently generate provinceand district-level statistics.

These tools provide a holistic picture of the education system, but seldom find mention in the policies they should inform.

I have firsthand experience with this disconnect. Between 2012 and 2017, I was part of the design and implementation of the Standardised Achievement Test (SAT) in Sindh, a World Bank-funded initiative assessing grade five and eight students in maths, science, and languages across all public schools. Each year, data from nearly three million students were collected, analysed and reported.

Despite its scale and rigour, SAT findings were never cited in any major policy document. Since the project concluded, Sindh has introduced multiple reform measures, including the Right of Children to Free and Compulsory Education Act (2013), the Teachers Education Development Authority Act (2012), two sector plans (SESP 2014-2018 and SESP&R 2019-2024), three teacher recruitment policies, and a curriculum reform plan in 2019. Not one made reference to the SAT dataset.

Why? Because most of these reforms were driven by large-scale international funding managed through international NGOs. These INGOs typically appoint foreign consultants to set the goals and define metrics for success, while local partners and institutions are brought in only to implement what has already been decided. With hefty budgets and international branding, these actors command the attention of both policymakers and bureaucrats.

This process creates an illusion of local ownership. In reality, the priorities reflect donor agendas.

Success is measured not by improved learning outcomes, but by budget utilisation reports. Multiple agencies may operate in parallel, each pursuing its own version of `reform`, often duplicating efforts and ignoring existing data. Once the funding ends, so does the reform initiative, leaving little institutional memory or sustainable change.Take, for example, the $75 million USAID-funded Pre-STEP project (20082013). It organised four high-profile national dialogues, leading to a consensus: provinces would hire only primary teachers with a four-year B.Ed degree.

The policy was enthusiastically adopted until the funding ended. Faced with real-world constraints, Khyber Pakhtunkhwa and Balochistan were the first to retract their commitments, followed by Punjab. Sindh waited, hoping to find enough qualified candidates, but with vacancies soaring past 60,000 by 2018, it too abandoned the requirement.

Over the years, Sindh`s education reforms have followed a similar pattern.

The Sindh Education Reform Programme (2006-2012) and the Benazir Bhutto Shaheed Youth Development Programme (2008) were supported by the World Bank and Unesco. The Sindh Education Sector Plan (2014-2018) received assistance from the Asian Development Bank, while the Sindh Education Sector Support Prog-ramme (2013-2018), which helped shape the Curriculum Implementation Framework, was backed by the European Union.

The Sindh School Monitoring System (2015-2017) was funded by the Global Partnership for Education.

More recently, reforms such as the Teaching License Policy (2023) and the current sector roadmap (2019-2024) have also involved internatio-nal support.

Each funding organisation came with its own goals, timelines, consultants, and frameworks. Rarely did they build on existing local efforts or datasets like SAT.

Instead, every new project marked a fresh start until the funding clock ran out.

If policymakers truly wish to create sustainable, context-sensitive reforms, they must break this cycle. That begins by using the data we already have. Evidence like SAT, Emis, and Aser should inform future policies, not be overshadowed by the next funding opportunity. Otherwise, we will remain stuck in a reform carousel: spinning fast, going nowhere. • The wnter is a professor of education at Sukkur IBA University, and currently working as dean, Faculty of Language Studies at Sohar University, Oman.