written by Michelle Kaffenberger, Danielle Sobol, Deborah Spindelman, Marla Spivack
A new paper shows that girls who are learning are more likely to stay in school. Improving learning could be key to achieving both schooling and learning goals.
The G7 recently agreed to two new education objectives: ensure 40 million more girls attend school and 20 million more girls are able to read by 2026. The UK government has reiterated its pledge to ensure 12 years of quality education for every girl, and committed to making girls’ education a key feature of their Covid recovery agenda. A new RISE working paper suggests good news – that progress on girls’ learning goals may actually be one of the keys to delivering on girls’ schooling goals.
The paper draws on longitudinal quantitative and qualitative data from the Young Lives Surveys in Ethiopia, India, Peru, and Vietnam, to understand why children dropout of school. The quantitative data reveals a strong link between low learning and later dropout. The qualitative findings reveal that low learning often underlies other, more commonly cited reasons girls drop out such as marriage or work. Girls report seeking ways to provide for their futures, and when it becomes clear that they are learning too little for school to provide future security, they seek other means such as a husband or a job.
After completing the Implementing Public Policy Program and joining the IPP Community of Practice, I was thrilled to receive an invitation to work with a group of master students taking a class at the Harvard Kennedy School (HKS) titled, “Problem Driven Iterative Adaptation (PDIA) in Action: Development through facilitated emergence” (MLD 103).
The course objectives were to (i) introduce the students to the PDIA methodology and (ii) give them an opportunity to immediately apply what they learned in class to a specific policy challenge that I had a privilege to nominate. Working on advancing legal education reform in Ukraine, I asked the group of five students to approach the following development problem:
“The supply side of Ukraine’s legal system is inadequate for fulfilling the role and responsibilities of the legal profession in a modern democratic society, contributing to the legal system’s self-perpetuating failure to ensure the rule of law and deliver justice in Ukrainian society.”
To help the students get up to speed and hit the ground running, I provided them with a list of reading materials and other resources that gave them background information on their policy challenge and a list of stakeholders ranging from senior government officials, leaders of the bar to law deans, local experts, and student union leaders that the students could contact to learn more about the local context and better understand the problem they were about to start working on. This support was important to engage the students in problem solving from the start. One of the students reflected on this experience in the anonymous feedback:
“[Authorizer] was a great supporter of our work, and has provided excellent guidance in understanding the problem of legal education in Ukraine. He […] kept us highly engaged.“
The course spanned seven weeks starting in January 2021. The students met twice a week on Tuesdays for lectures delivered by Matt Andrews and Salimah Samji and Thursdays for check-ins with me as their authorizer. Each week, the students did research on the development problem, interviewed stakeholders, turned in individual and team assignments. Even after delivering their final presentation on March 11, 2021, the students willingly continued their action learning to complete remaining interviews. When providing anonymous feedback, one of the team members even noted:
“At first, I thought, this is kind of an abstract topic that I never really had any explicit interest in. But honestly, I really enjoyed using the PDIA process to explore this topic and learn more about Ukraine and the context in which challenges present themselves. [I]t was great to get into it as much as possible. I would be happy to support this USAID effort in the future.”
The PDIA process taught us how to turn a ‘wicked problem’—a highly complex tangle of many problems with high uncertainty—into manageable components that we can begin to address. We learned a strategy for how to deconstruct an abstract problem with the fishbone framework. Most importantly, we learned that complex problems in unfamiliar contexts can be addressed through a structured approach. We had the chance to put theory into practice by working on radicalization in France.
There was a lot to unpack for the problem of radicalization in France. We had the opportunity to work with our authorizer, Raphaël, whom currently serves as a cyber security expert to the BNP Paribas Bank and Board members of think tank “Les Jeunes IHEDN.” His initial problem statement was to detect, react to, and prevent radicalization within private companies. However, it is very difficult for private companies to play a constructive role in the radicalization debate because of how sensitive the issue is and because there is a lack of dialogue even at a community level. But before we could start a conversation, we had to zoom out on the big picture to grasp the full complexity of radicalization.
On a winter’s afternoon in early February this year, a Mexican MPP1, a Brazilian MPP2, a Zimbabwean MC-MPA and an American MC-MPA randomly stepped up to the plate of abandoned projects in Nigeria. We, the four students and travelers, had never crossed our paths before (more accurately, we had never seen each other over Zoom). Additionally, none of us had ever worked in Nigeria. Before you think it could not get more chaotic, we had only 8 weeks to learn and experiment as much as we could on the assigned problem before coming up with novel and actionable ideas to expand its change space. Ready. Steady. Go! We weren’t ready, the journey wasn’t steady, but we definitely went on.
Maybe one of our first and most powerful realizations in our PDIA journey was that there was no silver bullet fix to the problem of abandoned projects in Nigeria. It took us two entire weeks to look at the problem with more curious and deconstructive eyes until we managed to draft a set of plausible causes and sub-causes that could be at its roots. We had to remain patient and above all curious and collaborative to shift from our initial planners approach to the searchers perspective required by the PDIA process.
As we deconstructed the problem through interviews and research, the Ishikawa fish diagram and the “five whys” heuristics helped us organize our insights in a meaningful fashion. At this stage, we also started to become more wary of our language usage versus our authorizer’s language usage (more on that later). And as our inquiry and knowledge deepened, so grew our ability to ask smarter questions and to find viable entry points.
As a pedagogical procedure for learning Problem-Driven Iterative Adaptation a group of four students from Mexico, Nepal, the United States, and South Sudan studied bilateral trade between Kenya and Canada with the help of an external authorizer: Dr. George Imbenzi, Honorary Consul General of Kenya to Canada. This global team, codenamed “Canadian Safari,” met with several Kenyan government officials, as well as, a Kenyan student studying in the US, a Canadian educator with non-profit experience in Africa, and an academic/practitioner of Kenyan-origin who leads a Harvard-based program, Building State Capability.
Uncovering Unseen Challenges in Kenya-Canada Trade
Our first thought was that the lack of a trade agreement was the major cause for limited trade between Kenya and Canada. However, when we broke down the problem of fledgling trade between the two countries into subproblems, we ended up with some causes we didn’t expect. (see fishbone diagram in figure 3).
One cause we noticed was the lack of capacity of Kenyan diplomats – in terms of technical knowledge and negotiation skills. Also, due to the frequent turnover of Kenyan officials, there was limited institutional memory.
MLD 103MA: Problem Driven Iterative Adaptation (PDIA) in Action: Development Through Facilitated Emergence is among the best classes at Harvard Kennedy School. This hidden infinity stone, 2-credit class challenges you to solve real-world, complex problems using the PDIA approach.
The tried-and-true PDIA process puts a learning structure in the way we look at complexity in local contexts from multiple perspectives. From a high-level, implementation includes a step-by-step approach of breaking down problems into its root cause, finding entry points, searching for possible solutions, taking actions, reflecting on what you learned, adapting, and repeating until the true solution is developed.
This semester, we were divided into teams to tackle real-world solutions. Our team, MY FM Inspiration, were given the challenge of examining legal education reform in Ukraine. Our authorizer was Artem Shaipov, a legal specialist and task leader for the USAID New Justice Program in Ukraine. In the first week, our team realized this problem had many dimensions to it.
There was an abundance of information to consume, and competing literature on what the problem actually was with legal education. To make the problem more difficult, many of us came from western legal education structures, but the Ukrainian legal education structure was very different, and in many ways still based off a Soviet Union era paradigm. Our team dived thickly into the topic with great humility and was focused on gathering as much information and learning as fast as possible. Our first fishbone diagram had nearly ~50 ribs and reflected the discoveries we obtained after the first two weeks.
It was hard to see a clear picture at the beginning. We found ourselves trying to dig past fake problems and problems that were just a lack of a specific solution. It was clear that PDIA was the correct method to use in this case because there was nothing linear about the challenges and potential solutions facing legal education in Ukraine. We had to fight the urge to try and find answers too quickly. The problem seemed to have a hundred gaps that each required individual keys and mastery.
Its ever-changing nature will make you question every move,
Build it up, break it down, and you shall find the truth.
When the ‘problem’ came to us, it was really a solution in the guise of a problem, for the original task was to make childcare in Burien a portable benefit that families could take with them. Even as we transformed it into a problem statement of families in Burien not having access to affordable and quality childcare, our problem construction work did not end there – we had painstakingly asked ourselves over and over again why this mattered and why it was a problem, not just a condition. Replicating this thought process with our authorizer Councilmember Kevin Schilling, we found that naming the distinction between the two created a pause and an opportunity for a deeper contemplation to give shape to the initially undefined problem.
Following the PDIA approach, we proceeded to problem deconstruction, which shed light on a number of insights, including underlying causes that did not seem to be obvious and inherent to the problem itself. Firstly, while stakeholders knew that affordable childcare was an issue, their understanding of its complexity was rather limited, contributing to insufficient motivation and urgency to take action. Secondly, the problem was not simply a lack of a solution, implying that no amount of expansion to Burien’s currently restricted budget will solve the childcare problem permanently. Our problem deconstruction pointed to much deeper societal issues that needed to be simultaneously or first addressed, including the need for a wider recognition that childcare is not an individual problem but in fact, one that weighs upon the community as a whole.
After we finally decided on three potential entry points to tackle first (awareness, lack of business support, and lack of city support), we began to fully appreciate the dynamicity of both the problem and the change space surrounding it. Through continually gathering information from a broad network of people and sources and updating our prior, we came face to face with the possibility that a change in one piece of information may trace back and require corrections to all of our past decisions. This realization, alongside the uncertainty that came with it, was difficult to embrace, and it also manifested in our AAA analysis. Kevin reminded us that authority, acceptance, and ability can change quickly, so does the feasibility of every solution that has been generated as a result of this analysis. It struck us that, perhaps we were too static in unpacking the problem and building the change space around the authorizer. Therefore, a dynamic mindset and an understanding of the problem as an evolving object, be it in the context of a six-week project or a five-year one, is an absolute necessity.
We started MLD103M as six complete strangers scattered across three continents trying to learn better ways to tackle complex problems like those we expect to face in our careers. The class was different, though, from what we were used to. We were divided into teams, given real-life problems, and asked to learn in practice. Our project was on Community and Police relations in a city in the US. Over the seven weeks working on this, we experienced quite the journey!
The magnitude of the problem felt the biggest in the first week. When we had just learned about the topic and hadn’t started the process of learning about and understanding the problem, it was difficult for us to imagine what contributions we could make over seven weeks. We had a difficult time figuring out where to start. But it was also difficult not to understand the problem in simple terms: a mistrust between the police and the community that was the result of last summer events, including the police-involved shooting and killing of a resident in the city. At the beginning, the problem seemed as if it started last summer.
After receiving our brief and the initial set of meetings we buried our heads in desk research in the second week. We were trying to construct the problem is: what is the problem is, why does it matter, and how would it look if it were solved. We also had conversations with the authorizers on what they think the “solved problem” would look like. As one of them put it, “we want to build a bridge of communication back and forth with our community… it’s truly a concerted effort between community/police to improve our community”. The authorizers’ investment in solving the problem was a great motivation for the team.
During the third week, we were still relying on what we read from public documents and the media on what the problem is. We started deconstructing the problem and thinking about possible causes of the problem. We started developing a fishbone diagram for what we thought the causes and sub-causes might be. We were clear that these are hypotheses to test and that this was an early draft at breaking down the problem, but it was an important starting point. During this week we started reaching out to people and getting out of our team’s bubble.
Over the past 8 weeks, we had the opportunity to work with the National Blood Transfusion Service (NBTS) on the lack of safe blood in Nigeria. The lack of safe blood during emergencies such as car accidents or postpartum hemorrhages has led to high numbers of preventable deaths.
Upon learning about our project, we were afraid that our lack of knowledge and experience in public health would limit our progress, but the Problem Driven Iterative Adaptation (PDIA) process showed us how addressing major problems such the lack of safe blood in Nigeria requires learning on the fly, using the diverse perspectives and contributions of our teammates, and constantly reflecting and improving on our work.
Here are some of our key learnings:
Focus on the problem, not the solution.
It is our nature as humans to be solution-oriented and not problem-focused. Is the lack of safe blood in Nigeria due to the low number of voluntary donors the problem? Or is it a combination of supply-sideand demand-side factors? Instead of assuming what the possible solutions could be, the PDIA process slowed us down and forced us to get uncomfortable and ask hard questions. This helped us identify the problem at hand and helped us construct our fishbone diagram.
MLD103, otherwise known as “PDIA in Action”, is a one-of-a-kind experience at HKS. On day one, we were randomly assigned teammates we would spend the next 7 weeks working with and given a problem to focus on. Quickly, we needed to get to know one another, build trust, and become experts in racial justice and how city governments operate.
Tasked with exploring viable funding mechanisms to enable the Asheville reparations process to progress, our team waded into a conceptualization of “the problem” that, we soon realized, was just a tributary flowing into a larger set of circumstances and hurdles. This early lesson served as a road sign reminding us to be ready to rethink at any time, framing our discoveries about policy and problem-solving along the way.
The Power of Iteration in Coping with Uncertainty
Approaching a task like implementing reparations for four centuries of harm inflicted on the Black community in the United States can be daunting to say the least. It’s instinctual to want to take it slow, refining all of the details of a comprehensive plan before it goes into action in order to ensure that it is done well and done correctly. At the same time, justice delayed is often justice denied. Advocates are justifiably trying to capitalize on the momentum of the moment given the unprecedented support for reparations. But there’s a reason reparations have never been implemented at such a scale before: we don’t know how. Never before has a society tried to repair numerous years and countless incidents of harm, but many of the disparities facing the Black community are centuries in the making, not the result of one isolated event.
Iteration gives us a way to cope with this very uncertainty. Accepting that we do not know the right answer can liberate us from the burden of needing to be right. We know that we’re not going to get it right immediately because the problem isn’t that simple. Rather, we have decomposed the problem and formulated small, incremental steps that we think could make a difference. If we’re wrong, that’s okay. We haven’t sunk years of time and energy into any one idea. After a week or two, we can stop, reflect, and refocus. As we try new things, we’ll learn more and more about what a solution could look like. Eventually, the uncertainty will disappear and a solution will be within our reach.