*10th December 2025* *Note on Writing: I've begun using Claude as a thinking and writing partner. While the thoughts and ideas are my own I have used Claude Opus 4.5 to help me structure the document and create a coherent flow. Claude was particularly instrumental in the drafting of the introduction and concluding sections* Vismaya Kalike is an after-school learning program built on three principles - joyful learning, self-determined environments, and community leadership. Each of these principles is valuable in creating agency in learners but at the same time each makes understanding progress genuinely difficult. This essay documents five years of attempts to capture that progress and to think through future attempts. It is not a case study of what worked. Looking back at our iterations, what strikes me most is how little intention connected each version to the last. We faced challenges, moved on, and often repeated mistakes we should have learned from. The disconnect between what we observed failing and what we tried next reveals something uncomfortable about our own process. I also realise now that we also didn't co-create the process with the facilitators and were very tokenistic in their participation. I'm sharing this reflection for two reasons. First, I hope it surfaces the depth of challenge in documenting progress within initiatives like ours - there are no easy answers, and the tension between assessment and our core principles is real. Second, I hope it holds us accountable. Being more thoughtful about our experimentation is no longer optional. **So why is progress so hard to capture in a space like ours?** There is an inherent tension between any type of assessment, evaluation and the joy in learning. Traditional methods are often a source of fear and anxiety particularly for children from marginalized communities. Children make most of the decisions in self determined spaces. This means that any types of assessment or evaluation comes at the cost of this principle. The children might understand their own progress and have mechanisms to do so but it may not be their priority to communicate that with us. Another issue is that progress in self determined spaces is never linear and each child tends to have their own journey. Observation, reflection and conversations which are the primary means to understand each child is hard to structure and comes with a lot of biases. Community leadership means that our spaces are run by facilitators who are from the communities they work with. Accountability mechanisms are through parents and elders in the community. While we workshop with facilitators extensively this means that they often don't have a research or education background. To truly be community lead we must find creative mechanisms to capture the voices of the community. Given these factors the only approach that makes sense to capture progress is using qualitative mechanisms. At Vismaya Kalike, we now focus on multiple perspectives to provide a 360 degree understanding of progress. We have borrowed from Fifth Dimension's approach (Gallego, Rueda and Moll) in Southern California. They argued that documenting progress in community-based settings requires multimethod, multilevel analysis—and that the mechanical quality of traditional forms often strips away the very emotion and context that matter most. Today, we draw our understanding from learner portfolios, facilitator field notes, coordinator reports, and community feedback. The most important of these is the facilitator field notes. The facilitator is one of the most important members in the work we do; at scale they often serve as the only consistent bridge between Vismaya Kalike and the community. Their account and documentation then becomes extremely critical to understand how each learning center and each child is doing. Most of our experimentation on documentation has then been on building mechanisms to capture the facilitators understanding. **What follows is each iteration, and what we learned (or failed to learn) from it.** ## v0 - 2019 Our first facilitator documentation came from the first few days of my personal reflections at Vismaya Kalike, what I have now come to realise as an attempt at "thick description" from ethnographical practices (Bogdan, Biklen). These reflections were verbose and included my frustrations and excitement at the center. Over time I also began maintaining a separate file for each child and separated my reflections by each child. I nudged other team members towards a similar process and have sporadically reverted to this process when I go into centers. ##### How did it go? These snippets served me well to create excitement among those who I shared it with. At its peak I shared snippets each day with almost 100 well wishers who would read through my reflections. The files for children have been great to look back on to understand the progress longitudinally of children today. Here are some snippets - My anecdote dated April 10th, 2019 ([[2019-04-10]]) > I’ve noticed that sometimes the smallest, simplest of changes can lead to big results. The stationery availability has really been a great example of that. While we have been thinking of mental health, trauma, nutrition and healthcare it is important not to miss out on some simple low hanging fruits that we can easily get to. The availability of a common pool of stationery that we have invested in has really improved the productivity among the learners and it seemed that for many of them this was the biggest hurdle. I also feel that fighting and stealing for these has reduced considerably because of this. Learners also feel a sense of responsibility and belonging for the things at the center and this is definitely super encouraging to see. It is super heartening to see how such simple solutions can lead to such great results. My note dated February 19th, 2019 for a particular child > Continued to work on 2 digit multiplication with M. It seems as though while M can do the problems, he still lacks some confidence. Perhaps I can try to conceptually explain what problem we’re solving to Manish as well. ###### Challenges Faced - The files with children focus a lot more on their academic progress and inadvertently pushed me to doing more FLN work with the children. - Documentation was very tedious particularly while trying to maintain a file for each and every child. Quality and length dropped over time as it became more mechanical for me. - The process of capture took a lot of time and was not always viable for someone who was only a part time facilitator. - The openness in the centers mean that we often have children in and out of centers and there is a lot of redundancy and too much data with lot of single entry files for children. ## v1 - 2021-22 The documentation process really kicked off when we expanded to five centers across Bengaluru post the pandemic in 2022. It was no longer possible for me to know all the children by name and our donors who expected monthly reports for the work they were supporting. We had two report templates 1. [Children documentation](https://docs.google.com/document/d/1vaBZt3DiA1So6V0835K4OAG3G8zMYACjZG5ItvjXkE4/edit?usp=sharing) - Asks some questions on children's excitement, interests, and changes observed. Also included an ASER like baseline questions of the facilitators understanding of each learner's progress in Native language, English, and Math. 2. [Facilitator Reflection](https://docs.google.com/document/d/1BAHoskuXE75xqA-H4JAUyn_MXOGNVLPTHttkczCzaIU/edit?tab=t.0) - Includes qualitative questions around trying something new, giving children freedom and new learnings. Also questions on a scale of 1-10 on core values with a view of building a spider diagram. To carve out time for this process we would often use the monthly meetings to fill out these forms. In addition to the reports we also created a WhatsApp group where facilitators would share photos from their learning centres. ##### How did it go? The facilitator reports often scratched the surface of things that were happening. The focus was on learning and a few games that were being played. It left no room to expand on the questions. Here's a sample translation from one facilitator in January 2022. > My students all speak Tamil but I'm trying to teach them Kannada. In the process I'm learning Kannada as well. > I tried newspaper reading at the center. Weekly once we're bringing food and celebrating. I have also prepared some charts with the English and Kannada alphabet and multiplication tables. > I am taking in children's opinions generally and trying to be more democratic > I tried Kabbadi, Kho Kho, Yoga and singing in the center. Consistent children documentation was a struggle. We managed to get a baseline ASER like story for some children but without consistency it was difficult to measure any progress. Though there was space for qualitative responses answers tended to be one word objective type answers. Here is a sample children template from one of the facilitators reports in November 2022 > How often does the learner attend the center: Everyday > Other than academics what is the learner interested in: Kabbadi > How often does the learner ask questions outside of school doubts? 1 to 2 doubts > Is the learner excited to come to the center? Yes > Has the learner come up with any new experiments, games, song, dance etc? Games > How is the learner finding school? --blank-- > What changes have you observed in the learner? Good learner ###### Challenges Faced - Team members were hesitant to reflect honestly. This was very clear in the likert scale with almost all facilitators filling out 10s in all the questions with very rare 8s and 9s. It did not help us understand progress at all. - The children's profiles were not sustainable at all. It was not easy to create reports for each child particularly at the end of each month. - Placed an increased emphasis on FLN and did not give us any information on agency at all. - Emotion was completely lost and we ended up with very mechanical, generic answers. - Lot of answers could be dived into and delved on but the report format didn't leave room to capture more. - We had a lot of photos from WhatsApp but often they didn't communicate anything apart form the number of children attending the centre and their seating arrangement. ## v2 - 2022-2023 In 2022, we decided to tweak this process and got rid of the children documentation all together. Rather than capture every child's progress every month we decided to take a more case study approach where facilitators share anecdotes about one particular child. In hindsight this should have been a more thoughtful reflection of the challenges faced rather than only a consideration of the time available. To encourage a reflective process we asked facilitators to share a goal on what they'd like to achieve and then share an update on how it went. [Template Link](https://docs.google.com/document/d/1dm6pRaAetHxl-LvNYEV_nRCxXCYa-mrolfwr_9oFbV4/edit?tab=t.0#heading=h.opkge4khhqvl) We shared the rationale behind this approach and often gave space within the monthly meetings to fill these out. ##### How did it go? The goals as set by the facilitators often leaned very academic and around events. A sample snippet from a facilitator's goals in August 2023 > Children are having a first term exam so preparation for that. Preparing for the August 15th program. To understand agency we asked facilitators to document any requests that have come from the children. Most answers looked something like this answer in a November 2023 report > Kannada reading, biology concepts, basic general knowledge, early numbers Most of the time the goal from the month was either partially or wholly achieved largely because the goals were being filled during the time of report writing and not at the start. For the anecdotes we would often get anecdotes like this from an October 2023 report > Before joining the center one student A was not studying properly at all. Once I guided him he has improved in his studies. The question on what is something new they learnt also got fairly generic responses like this in June 2023 > How to teach students, how to control students, teaching techniques ###### Challenges Faced - Once again we struggled to capture the real emotions of facilitators. While they were able to share beautifully the progress in the meetings these did not translate to the written form in the writings. - Collecting reports itself became a challenge particularly when facilitators missed monthly meetings. Facilitators found the process very tedious and reports tended to be very repetitive. - The stories of children were very thin. Often the anecdotes talked about improvement and growth but not around what the growth exactly was. ## v3 - 2024 Once again, we took a fairly narrow approach to the challenges faced. Rather than trying to understand what we were looking for, we focused on the reporting process becoming too mechanical and aimed to address that in this version. This report tried to instill ideas of action research in the facilitators. Facilitators were asked to try something new at the center, share why they tried it, what they observed and what they learned. We also retained the anecdotes as an unstructured space for facilitators to share things that were happening. [Template Link](https://docs.google.com/document/d/1hLKG711REi1OB9ef6cC7Kdk2OTUOs17jepbKhrY5uXI/edit?tab=t.0#heading=h.opkge4khhqvl) To empower them on things they could try we shared a resource book of learning activities and often brainstormed ideas on things to try in the monthly meetings. ##### How did it go? Experiments tended to be very narrow in scope and often involved trying some new art or learning a specific topic. Learnings did not follow from the experiments and the entire experiment was not very coherent. An example from a facilitator in November 2024 > We tried some art and a new electric shock game. > > Children will get a lot of happiness doing art and playing games. Children are interested in drawing so we are trying more art. > > Some children did drawing and some are busy with homework but everyone draws and plays on Saturday > > I learnt some art and a recipe from one of the students. However it is important to acknowledge that this did nudge facilitators towards trying new activities at their centres. They recognised the need to reinvent themselves constantly so children remain excited. Translation from a facilitator in February 2025 after our workshop - > The new activities tried this month included: What is a noun? (in Kannada), word building, story making, many games > > These activities were tried because many children were not aware of these concepts. > > After these activities, the children understood the concepts better, especially enjoying and having fun with story making > > I learn something new every day from the children through activities like singing, dancing, or playing games. Even when I am in the kitchen, my mind is focused on what I will teach at the center that day. ###### Challenges Faced - Observations were often limited to it went well or it didn't go well. Emotions were rarely captured. - Experiments were often limited to trying a new activity or game. Though other types of experimentation often occurs facilitators did not recognise these as experiments. - Facilitators often didn't have enough bandwidth to think about new activities. It was only the month just after the workshop that they tried something truly novel. - Stories of children were completely lost and missing. ## v4 - 2025 It was only in v4 with significant expansion and the opening of 35 centres in Hosapete did we really start to think deeply about documentation. The centers in Hosapete would add more complexity; now conversations and learning from dialoging with facilitators would no longer be viable and we truly had to build a documentation system that worked. The facilitators largely were not happy with the structured reporting style. They felt that the structures were not conducive to what they wanted to share and for the monthly structure they proposed a free flowing set of bullet points. We proposed to our facilitators a shift towards field notes. The argument we put forth was how the reports weren't capturing the stories that we tended to hear from them when we visited or met them. The hope was to go back to our v0 at thick description but now at scale. Given we had a WhatsApp group where photos were shared, we could leverage this and the photovoice technique (Wang and Burris) as a means of enabling facilitators to think critically about their learning centers. We recognised that it's impossible to track every child's progress and to lean community reporting mechanisms. What we would instead look for is case studies and an understanding of specific cases representative of different points in a distribution curve at each centre based on our discussion with Dr. Ramanujam from APU. To implement this effectively we have engaged with facilitators in group workshops and in individual settings to help them move away from the formal notion of a mechanical report. One approach that worked well was to understand who facilitators talked to about the center. Often facilitators shared stories from their centres to their friends and family much like I would do. We encouraged them to share what they were sharing with their friends and families in the group chats. Further we focused on the emotions they felt and to share stories or anecdotes that brought them some strong emotion. Facilitators were initially confused on whether these are the things they should share and we encouraged them to share stories about particular children as well not hesitating to share individual journeys. ##### How it's Going? The field notes approach has yielded mixed results. Still we find facilitators caught up in the idea of a report and sharing more of what they think sounds good. However we have started seeing some interesting stories emerge in the notes shared by facilitators particularly among the facilitators in Hosapete who don't have ready access to us and with whom I have engaged one on one. A translation of a few notes shared by a facilitators in the month of November 2025 > Today at our learning center, I told the children to do their homework. > A girl named A said, "Akka (sister), I won't do homework here- I'll go home and do it." When I asked why, she said, "Akka, I feel like I need to learn two words first." > I said okay and taught her the two words. Her face lit up with so much happiness. > Then I asked L and R, "I'll give you something to learn - will you learn it?" They said, "Akka, L and I want to play buguri (spinning top)." I said okay, go play. > The children had gotten into a fight on Friday. When I asked them about it today, they honestly admitted that they were the ones who made the mistake. They even took a promise, saying they won't make such mistakes again in the future. > Hearing this made me very happy—because they owned up to their mistake. > Also, I taught them how to use English words by writing on the board. ##### Challenges Faced - Typing is still not the most natural form of communication for many facilitators. The emotions they have when sharing in person don't quite get captured in the written form. - Power and reporting hierarchy is still evident in the field notes. The notes appear to be censored and tuned to what facilitators feel we want to hear. - FLN and academics is still the predominant theme in the field notes. - Facilitators don't always have consistent access to a smart phone and data particularly in the rural areas. - Since facilitators are the ones taking photos, they tend to be quite inorganic and do not cover what they themselves are doing. - The photos have not lent very well to the photo voice technique since there is not a lot of variability in the environments. ## Looking Ahead - 2026 While the unstructured field notes are a step in the right direction the richness in conversation with facilitators are still not getting captured in these field notes. With Generative AI there is an opportunity to leverage technology and allow facilitators to have a natural conversation and describe what's happening at their centres. Numerous studies have shown the strength of a voice based AI in marketing surveys and it'll be interesting to see if the same results translates in our work. But I want to be careful not to suggest that technology is the answer. Looking back at five years of iteration, what strikes me most is not the specific mechanisms we tried but the pattern underneath - we kept changing the container without changing what we were asking for, or who we were asking. Each version addressed a symptom. v1 responded to scale by adding structure. v2 responded to tedium by reducing scope. v3 responded to mechanical answers by asking for experiments. None of them asked the more fundamental question - what does progress actually look like in a space like ours, and who should be defining it? It was only in v4 that we began co-creating with facilitators rather than designing for them. The shift to field notes came from their feedback that structured reports didn't fit what they wanted to share. The framing of "share what you'd tell your family" came from understanding how they already talked about their work. This is still early, and the results are mixed - power dynamics persist, self-censorship is real, and FLN still dominates the notes. But the direction feels different. The answer to our challenges comes back to practicing our own principles. We want self-determined environments but did not allow facilitators to participate in designing their own documentation systems. We believe in community leadership, yet we treated facilitators as data sources rather than co-creators of knowledge about their own centres. Voice AI, if we pursue it, will be one experiment among many. What matters more is the stance we bring to it - that any documentation system must be built with the people using it, that we need to be clearer about what we're actually trying to understand, and that we need to hold ourselves accountable to learning from what we try. This reflection is a small step toward that accountability. I don't know if the next iteration will work. But I'm hopeful that, for the first time, we're asking better questions. | | | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Gallego, Margaret A., Robert Rueda, and Luis C. Moll. "Multilevel approaches to documenting change: Challenges in community-based educational research." _Teachers College Record_ 107, no. 10 (2005): 2299-2325. - [[Multilevel Approaches to Documenting Change Challenges in Community Based Educational Research]] | | Bogdan, R., & Biklen, S. K. (1997). _Qualitative research for education_ (Vol. 368). Boston, MA: Allyn & Bacon. | | Wang, Caroline, and Mary Ann Burris. "Photovoice: Concept, methodology, and use for participatory needs assessment." _Health education & behavior_ 24, no. 3 (1997): 369-387. - [Link](https://deepblue.lib.umich.edu/bitstream/handle/2027.42/67790/10.1177_109019819702400309.pdf?sequence=2) | <div class="subscribe-box"> <p>Get notified of new essays and receive an email whenever I publish something new.</p> <form action="https://buttondown.com/api/emails/embed-subscribe/vignesh-prasad" method="post" target="popupwindow" onsubmit="window.open('https://buttondown.com/vigneshprasad', 'popupwindow')" class="embeddable-buttondown-form" > <label for="bd-email">Enter your email</label> <input type="email" name="email" id="bd-email" /> <input type="submit" value="Subscribe" /> </form></div>