Okkhor

Okkhor seeks to improve English language literacy for young people in Bangladesh (and beyond) to enable them to access greater academic and career opportunities and realize their full potentials.

English language education in Bangladesh is largely outdated and ineffective, and as a result Bangladesh ranks 66th globally according to the EF English Proficiency Index. Despite their enthusiasm and technical skills, poor English language literacy excludes Bangladesh’s 50 million strong youth population from accessing better educational and professional opportunities, and making bigger contributions to the global economy.

Okkhor seeks to develop an English language learning platform that will serve not just individual learners directly, but ideally also be adopted by the private education system so that it becomes commonly ised in the classroom. In order to do so, our focus is on developing simple but effective solutions that lead to clear and measurable learning outcome improvements. Therefore, our focus would be on:

1) Impact-driven solutions

2) System level adoption/change

I am the founder/CEO of Okkhor, and will also serve as Team Lead. As the founder, I set the vision and direction of the organization and ensure the team is empowered to execute on that direction.

As building a solution that is evidence-based is our core principle, creating a framework for understanding our impact is a priority. We also want to take a data-driven approach to refining our product over time. I had the intention of spending a dedicated amount of time for this purpose by finding, engaging and supporting appropriate experts - engaging with LEAP is perfectly aligned with this priority objective.  

I will be supported by one or two team members who would also have otherwise been (partly) engaged for this purpose, even though they have other roles as well. They include an English language studies graduate and an English language teacher.  

Online platform for learners to improve their English reading and listening skills through daily 15 minute practice sessions.

English language teaching methods in the mainstream Bangladeshi education institutions (including schools and coaching centers) are largely outdated and ineffective. The methodology followed is typically grammar-heavy, placing excessive emphasis on rote memorization - also referred to as the Grammar-Translation Method. Additionally, English language instructors themselves often lack fluency, have to face overcrowded classrooms, and struggle with curating learning materials and providing learners with feedback.

Individual learners also have no learning resources that are designed for native Bengali speakers, and are adaptive to their specific skill level.

Okkhor is a simple website where we publish English reading/listening practice content daily. Fresh content is published daily, based on news and current affairs, so that the material is both engaging and effective (“authentic text”). Reading exercises include hints for difficult words, audio and quizzes. We also show learners progress reports to give them meaningful feedback and keep them motivated (gamification). Learner progress can also be viewed by teachers on a separate teacher dashboard. We do not have courses or teach lessons, but emphasize daily practice for language skill acquisition (intrinsic language learning).

We certainly consider addressing learning variability a core design factor: each content we publish is available in multiple versions to suit different difficulty levels: Level 1 (suitable for young learners or beginners) starts with approximately 50 words, using very plain language, going up to approximately 500 words and more complex language for level 5. Individual learners can pick the difficulty level that suits them, regardless their peers are at attempting the same or different difficulty levels. Individual learners can attempt to move up a level once they feel more confident, and all this data will be captured in the system and be visible to teachers as well.

Although we wish to, we have not conducted any extensive studies on the effectiveness of our solution. We do have user data for about 12 months that can provide some insights, specifically:

- User retention: how many sessions each user has completed since they joined a platform, which could give an indication of how useful and engaging they find the platform

- Difficulty levels selected during each session (may be able to see if the difficulty level chosen goes up over time)

- Self-rating for each exercise (can see if the self-rating improves up over time)

I would not consider this data to be formal research, though it could at most be considered formative research.

The data-point we have focused on most till now is user-retention. We use this as an indicator of both possible learning outcomes as well as usability of our platform.  

So far, we have engaged approximately 7000 users, who have generated approximately 51,000 sessions (over the last 12 months) 

During our trials so far - on a very basic version of the product, and with no involvement of teachers or follow-up from the Okkhor team- we have seen a range of user responses:

* 80% of users have had less than 5 lifetime sessions

* A segment of users has been highly engaged and self-motivated, with 200+ lifetime sessions.  

Our interpretation is that if learners can be engaged long enough to start seeing some initial benefits, they are more likely to continue long-term, and see more meaningful improvements in their language skills. This is why moving forward, we want to prioritize teacher/classroom engagement, even though we can improve features and attempt to improve retention for individual learners that are not using Okkhor as part of their class.

There are already many options for English language learners in Bangladesh, including traditional coaching centers as well as online courses. However, these appear to be of limited effectiveness. Okkhor is only entering this segment because we seek to build something that will truly make an impact for our learners, and we want to be methodical and maintain data-driven rigor to ensure this.

We also want to engage the mainstream education system by taking Okkhor to schools and attempting to get integrated into their curriculum and teaching plans. In order to be credible and be convincing for education administrators, having a formal study of our effectiveness in partnership with MIT LEAP could be decisive.

Engaging with LEAP now is ideal as we are about to iterate on our product and develop the next version. With our limited funds, we do not have the option of iterating multiple times before commercializing, so it is very important that we can ensure the impact our solution has before we attempt to raise funds or generate revenue.

1) Does regular reading practice based on difficulty-appropriate texts lead to increased confidence and reading skill development (even without additional/formal lessons). In how long, and to what extent?

2) Does regular reading and listening practice also have positive impacts on other language skills (writing and speaking)?

There are a number of specific areas that where we want to focus our development:

1) Content development guidelines: what are the best practices - or what could be the good guidelines - for developing reading practice content for Bangladeshi learners? There are a number of readability and skill measurements frameworks including CEFR (Common European Framework of Reference for Languages), Lexile Levels, and more, and a better understanding of these could be a good starting point. 

2) Can we measure the impact on reading skill development on a group of students that regularly use Okkhor vs students in a similar setting that are not using Okkhor? How can we design and administer these evaluations?

3) Can we develop a framework for collecting feedback from all stakeholders (school students, teachers, non-school learners) on both the effectiveness of our current solution but also in a manner to help guide improvements in product features?

4) What sort of usage data should the system capture, and how can we use this data to continuously generate learning-outcome related insights?

This may be a lot for a single sprint, so we will be happy to prioritize based on what is feasible.

As a direct outcome of the above-mentioned study objectives:  

1) Develop and implement guideline for daily content development. Ideal guideline would be local-appropriate (keeping in mind biases of native Bengali speakers).  

2) Develop and implement periodic reading-skill evaluations with partner institutions, to not just establish initial credibility, but to also gauge impacts of updates we will continue making to our solution over time.  

3) Develop automatically-reported metrics form platform usage data as real-time indicators of learning progress of users.  

Again, we would focus on only the outcomes that would be feasible for a single sprint with the LEAP project.

The core long-term outcome we are seeking is of course to refine our solution to ensure maximum learning impact for our stakeholders. Our ultimate vision is to have a macro-level impact, which could be reflected in global rankings, economic impacts, and so on. In order to partner with schools on a broad level, credibility from an evidence-based solution design, as well as partnership with MIT Solve, will go a long way.

Our short-term objective is to orient Okkhor as an evidence-based learning solution developer, by creating tools for collecting meaningful learning impact data, and building a company culture around that. The data we collect from the tools/frameworks we develop during a LEAP sprint would go on to form the basis for both daily content development as well as product development and improvement.

The core long-term outcome we are seeking is of course to refine our solution to ensure maximum learning impact for our stakeholders. Our ultimate vision is to have a macro-level impact, which could be reflected in global rankings, economic impacts, and so on. In order to partner with schools on a broad level, credibility from an evidence-based solution design, as well as partnership with MIT Solve, will go a long way.

Our short-term objective is to orient Okkhor as an evidence-based learning solution developer, by creating tools for collecting meaningful learning impact data, and building a company culture around that. The data we collect from the tools/frameworks we develop during a LEAP sprint would go on to form the basis for both daily content development as well as product development and improvement.