Learning Note 11: Does going to school equal learning? Conversations about learning in ten districts in Tanzania

This post was initially written for and published on with the title; "Conversations about learning in ten districts in Tanzania". It reflects on a qualitative study on the state of education in Tanzania and learning outcomes among district level decision makers.

In Tanzania, for too many children schooling does not equal learning: official statistics show high rates of failure in national primary (as well as secondary) school leaving exams, and Uwezo (an independent monitoring mechanism run by Twaweza) has over the last five years demonstrated that while millions of children attend primary school, the most basic learning outcomes in numeracy and literacy continue to be very low.

The government is by no means deaf to this — and discussions on the quality of education are actually quite loud in the policy arena and in the national media. For instance, the most recent release of Uwezo data this past September sparked a mini-storm in the media, which focused on the (frankly very poor) learning levels among children. The effect that storm had for Twaweza is another story. But the point I want to make here is that while these discussions can feel very loud in the national arena, they are quiet — or rather, not much heard from — the sub-national level, where of course implementation takes place.

So, in an effort to better understand the “production of education” in districts and schools, we conducted a qualitative study to explore the state of debate and communication about education in general, and learning outcomes in particular, among district-level decision makers in ten selected districts in mainland Tanzania. The overall purpose of the study was to generate formative insights to for Twaweza’s future communication and engagement initiatives (at national and sub-national levels), and to inform the design of our initiatives geared at improving learning outcomes for primary-school children.

It was a purposive sample — we went to districts where we had already focused some of our sub-national communication activities (so the results are illustrative, not representative). In each district, a number of key actors were interviewed, including District Education Officers, Ward Education Coordinators, and Head Teachers of selected schools. The discussions with key actors had different themes, such as financial inputs and resource mobilization, options to provide support to teachers as well as support to pupils, the link between school inputs and learning outcomes, and engagement with Uwezo materials and results.

And while as Twaweza we focus a lot on rigorous and quantitative data, it never ceases to amaze me how much valuable insight is gathered through focused, meaningful conversations.

You can read the entire report here, but below I outline a few of the core issues that surfaced, with some personal reflections.

Issue 1: Understanding what currently “powers” the system — and what are the incentives (or barriers) to change.

The discussion of quality of education was centered around the scores on the primary-school national exams (especially grade 7, when the exam mostly decides whether a student will progress to secondary school); as well as on the numbers (or proportions) of children advancing to secondary school (whether they actually passed or not). There was no (open) questioning among the various stakeholders of whether “passing” means that children have actually mastered the required skills. In a related point, when asked about what kind of support is provided to pupils, most respondents focused on extra classes to prepare pupils for examinations, especially the exams in Grade 7.

Reflecting on incentives and change: The system generally rewards passing, not learning (e.g. schools are nationally rated according to pass rates). Nowhere is this is showcased better than in the steep increases recorded for the Grade 7 passing rates between 2012 and 2015 (for the districts in this study), contrasted with the stagnating Uwezo results for the same districts (and age group; see page 14 in the report). In my view, this means that the system produces a picture of success, while independent data paints a picture of stagnation. This will likely be a very delicate topic to breach in policy dialogue. On the other hand, change is possible — a number of respondents in our study also discussed the importance of mastering basic skills (basic numeracy and literacy) in the lower primary grades. The government’s “3R” program, which trained many primary school teachers in enhanced literacy and numeracy for early grades, has very likely contributed to this. So in a centralized system, top-down approaches yield results; unclear whether bottom-up pressure would do the same. Perhaps the more relevant question is how to get the “top” engaged.

Issue 2: Being honest about which interventions will actually result in children learning more.

Discussion about the reasons for poor performance often look for faults or causes outside of the education system; when it reflects inward, it’s focused on inputs. Two main factors were given most commonly as rationale for learning outcomes not being at desired levels: lack of resources, and the “local culture.” In the latter, education stakeholders tended to single out parents as being “non-supportive” to education — and this was pervasive (and often very strongly worded) across the ten districts. In the former, all respondents noted that education resources received from central government are insufficient, as well as often untimely. When discussing how to improve learning outcomes, most respondents focused heavily on inputs (such as desks, and books).

Reflecting on interventions: The discourse maintains “old truths” which are not supported by evidence. For instance, the focus on the importance of inputs as directly linked to improving learning outcomes persists, despite ample research showing otherwise. On the other hand, reality is messy, and it’s unwise to pretend that school-level inputs aren’t an issue in Tanzanian schools, or that parents unanimously value education. Shifting the dialogue from assigning blame to shared responsibility sounds good in theory, likely challenging in practice.

Issue 3: Managing core human resources in the system.

Teachers were nearly unanimously reported as dedicated to their work. Much of the discussion around teachers centered around the support they are lacking (such as not being paid on time), and that they work under hard conditions (e.g. lack of appropriate housing, lack of teaching materials at schools). Interestingly, a handful of districts reported some innovative management practices. For example, one district implemented a consultative process, drawing on teachers, head teachers, parents and governmental stakeholders, to develop a jointly-owned plan on how to improve education. In this study, we were unable to examine whether these practices take hold and lead to any change, but it would be an interesting case to follow.

Reflections on core human resources. The picture of teacher dedication in face of adversity as reported in this study is at odds with other recent Tanzanian research showing that reality is much more varied, and it includes high absenteeism, low motivation, and a low skill base among teachers (e.g. Service Delivery Indicators, among others; citations in report). There’s a real need to thoughtfully widen the discussion on what teachers and head teachers can do better, but again, in a dialogue which recognizes the real constraints under which they work.

So this study has shed valuable light on issues underpinning the poor learning outcomes among Tanzanian children. It’s not an easy set of issues to tackle. For Twaweza, this study is providing important inputs, together with other evidence, on how to shape the next year’s implementation plan and engagement strategies more broadly. For me personally, it’s been a real insight into how systemic incentives and rewards shape the outcomes.


Read more: education Tanzania learning outcomes qualitative study


You might also like...