For organisations needing to provide impactful training content at scale, affordably and in remote locations to people with low digital literacy — could WhatsApp be the solution?

Think WhatsApp is just a messaging platform? Think again.

A WhatsApp helpline is protecting Delhi residents from police harassment and extortion (bribery). Within the first week of launching, the helpline received 3,700 WhatsApp messages and 622 calls leading to 5 policemen being investigated.

In Brazil, WhatsApp was used to quickly identify the outbreak of Zika virus. With Doctors across the country using their personal and professional connections to share spikes seen in symptoms such as red eyes, sore joints, rashes and a particular type of birth defect.

And in Rwanda, young people are using WhatsApp to improve their spoken English skills by completing a series of daily tasks posted, as part of a six-week Speak English Programme.

And early results are promising — with 88% of participants reporting speaking more English than before the Programme began and 85% of young people changing their behaviours to speak more English on completion of the Programme.

The WhatsApp based Speak English Programme is the brainchild of Harambee Youth Employment Accelerator, and aims to assist young Rwandans to find work in jobs requiring spoken English fluency.

And in a country where only 10% of young people speak English fluently, and where thousands of new jobs requiring spoken English skills are projected, Rwanda needs an English training solution at scale.

And traditional online training approaches for scale fail to deliver in a country where both internet connectivity and access to computers are limited, where levels of digital literacy vary, and where content needs to reach people in geographically remote areas at an affordable cost for data.

In this context, an effective training programme must deliver engaging content at scale that is impactful and accessible.

And can WhatsApp deliver on these metrics?

Harambee’s Speak English Programme provides an early indication.

WhatsApp offers training at scale

WhatsApp is scale, with two billion global users.

WhatsApp usage in Africa? According to the 2018 Quartz report, “WhatsApp is absolutely dominant in the African continent”.

In Harambee’s own network of unemployed young Rwandans, only 17% of people report not having WhatsApp.

And with WhatsApp, people don’t just download the App, they use it too, with more than 100 billion messages being sent each day.

The power of this? You don’t need to direct thousands of people to your platform for training. They’re already on it.

And the near universality of WhatsApp was confirmed on the Speak English Programme. On launch, participants were “ready to practice their spoken English” after receiving the first message — regardless of where they were in the country.

WhatsApp is the definition of accessible — phone based, data light and easy to navigate

In Rwanda: Only 0.14 percent of Rwandans have computers connected to the Internet but the number of mobile connections in the Country, in January 2020, was equivalent to 73% of the total population.

So phone-based solutions are arguably the only answer to youth-based training at scale.

But access is not just about the ability to use physical hardware like a computer or phone. Training must be affordable to both deliverer and consumer.

And the case for the affordability of WhatsApp?

“To support a young person’s learning on our 3 week online programme costs $10 per person per month. For the Speak English Programme we didn’t need to provide data stipends to learners given how data lite WhatsApp is” says Hannah Adams, Country Manager for Harambee Youth Employment Accelerator.

Commenting for marketplace.org, Abraham Leno based in Kigali and Head of the not-for-profit Eastern Congo Initiative succinctly puts forward the argument for WhatsApp:

“Even the simplest phone, the cheapest phone, can allow you to connect with somebody anywhere in the world, and the data usage for WhatsApp is way cheaper than many other platforms.”

Accessibility has many layers, and also relates to a learner’s digital literacy to use a platform. And Harambee’s experience in both Rwanda and South Africa shows many young people struggle to navigate browser based applications having not had sufficient exposure to computers.

But in the case of WhatsApp, Belinda Lewis, Head of Product Development for Harambee Youth Employment Accelerator, explains: “The advantages of WhatsApp is that it’s also a near universal behaviour to chat, so you remove the digital literacy barrier.”

WhatsApp delivers engaging content making learning fun and accessible

WhatsApp is deliberately designed to distract. Each new ping draws our attention away from the present to check a message that’s just arrived.

Worse still, WhatsApp is somewhat one dimensional, designed to be primarily text based or at most to allow for a short voice note or an image.

Also, unlike many traditional online learning platforms, WhatsApp has no gamification or interactive polls and quizzes to keep a person’s interest.

But WhatsApp mimics real, live conversations conveying emotion through the use of emoji’s. Leaning into this design feature, messages in the Speak English Programme were carefully crafted to be conversational, warm and friendly, establishing an intimacy and rapport with the reader.

WhatsApp transmits information quickly and easily. So to capitalise off this design feature, Programme messages were designed to be short, delivering “sound-bites” of spoken English exercises that could be read, understood and acted upon in no more than two to three messages.

Posts were then varied to include images, texts and voice notes to give deeper dimension to both the platform and content. And social learning was encouraged by creating WhatsApp groups where programme participants receive messages and post messages too.

Qualitative feedback from participants of the programme suggested the mindful design of the WhatsApp content made engagement possible.

As Anick Umulisa, a young woman on the programme remarked: “I’m really into the teaching methodology because it’s easy to understand and friendly. I can’t get bored or tired during our sessions. I hope to speak English fluently within a short period of time.”

WhatsApp training content can be impactful

The Speak English Programme set out to maximise time spent speaking English, build confidence speaking English, automate access to/finding English words and to improve the ability to think in English

And the results?

After completing the Speak English Programme, 76% of participants felt more confident speaking English.

71% Programme participants also began to think a lot more in English - a key metric for second language acquisition impact, as highlighted by the English language and culture blog of fluentu:

“If you’re “speaking” English in your mind, it’s easier to speak it with your mouth.”

And 100% of programme participants also found accessing English words easier.

The Speak English Programme had impact. And Harambee Youth Employment Accelerator’s pre and post programme assessment data showed a 38% shift in spoken English competence amongst participants.

WhatsApp delivers on key metrics, but needs a few fixes…

A huge advantage of traditional online learning platforms are strong reporting functions to measure important metrics such as level of engagement and completion rates.

But right now, WhatsApp is an entirely “black-hole experience” when it comes to reporting. And while our Programme used surveys to receive feedback, this approach only serves as a proxy for gauging participant engagement.

WhatsApp also comes short when it comes to assessments of participant progress.

Although these present some serious drawbacks before WhatsApp can be considered an end-to-end training solution, Harambee remains bullish on the benefits of WhatsApp.

What other global platform is there that most young people use, that cuts through accessibility barriers, delivers at scale, and can close the skills gaps of young people?

And surely, it’s time we met young people where they are to deliver learning?

Patrick Fore on Unsplash

Young Rwandans need jobs but have low levels of English fluency. The Rwandan government is growing and attracting new jobs, but these jobs require English fluency. The good news? There are three ways to fix for spoken English fluency.

Just 10% of the nearly 7000 young Rwandans assessed by Harambee Youth Employment Accelerator speak English fluently. Yet this is arguably the most valuable skill needed to work in the formal labour market, with thousands of net new jobs projected over the next few years where English fluency will be the entry requirement.

This is because the Rwandan Government has been hard at work growing sectors to stimulate its economy, attract jobs and deal with a youth unemployment rate of 22%, according to the 2020 Labour Force Survey. And their effort is paying off, but the jobs created or attracted require fluent English speakers.

But what does it mean to speak English “fluently”?

Fluency is defined by Daniel Morgan, Head of Learning Development at the Shenker Institutes of English, as how “smoothly” and “efficiently” a second language speaker is able to speak on “a range of topics in real time”.

So, to fill a talent pipeline with “smooth” and “efficient” English speakers to meet employer demand, the 10% spoken English fluency needs fixing. We propose this could be achieved in three ways.

First, understand and then make a distinction between the four different language skills or abilities. These include reading, writing, speaking, and understanding a language. If we fix for this, on the lips of everybody will be the question: “What language skill(s) needs development and mastery?”

Second, be able to assess for competence between these four different language skills. How well does a person speak English? How well does a person understand or write English? And if we resolve this, falling from each person’s mouth will be the question: “have we assessed for spoken English competence?”

And finally, change how young people are taught English as a second or third language.

How?

By shifting from teaching which over emphasises receptive language (where young people hear the language), to teaching that promotes production or expression of language (where young people are required to creatively produce or speak the language). Fixing this would require educators to answer: “Have we designed an English language programme that helps young people to speak the language fluently?”

The demand for spoken English

Employers in the Tourism and Hospitality industry want “smooth” and “efficient” English speakers able to interact with their international clientele. This was especially true prior to Covid 19, when in 2019, tourism contributed 15,1% to GDP on the back of Government efforts to draw domestic and foreign tourists to Rwanda.

In addition, Rwanda was recently awarded: “Upcoming BPO Location in the Top Offshore Delivery CX awards of 2021”, in a survey conducted by Ryan Strategic Advisory. A clear signal that the Government’s efforts to establish a Business Process Outsourcing (BPO) sector has begun to pay off.

BPO’s typically offer thousands of entry level jobs but the ability to speak English fluently and confidently is highly valued in this world of servicing customer queries and complaints.

The emergence and growth of this demand should be seen in the context of the skills needed by Rwandan Employers in adjacent sectors. In interviews conducted by Harambee with over 100 of these Employers, fluent English speakers were identified as their single biggest barrier to finding talent.

So, for young people seeking economic opportunity in the formal labour market, speaking English fluently provides a significant competitive advantage. And for the government to grow the economy, and solve for youth unemployment, the country needs a talent pipeline of fluent English speakers.

And this can be achieved through three fixes.

Fix number one: To get young people speaking English fluently we need to better understand that any language has four language skills or abilities

To test this theory, think about your own languages. Do you speak each language equally well? Do you understand one language better than speaking it? Can you write and read at the same level in all your languages?

Even though we would likely rate our speaking, reading, writing and comprehensive abilities differently across our different languages, we commonly refer to a person’s language ability holistically.

We tend to describe a person as being good or poor at a language. And when we use this shorthand, we lose sight of the subtle but essential distinctions in a person’s ability across the four different language skills or abilities.

The four basic language skills being listening, speaking, reading and writing, as described by Columbia George Community College.

This means a person may be good at speaking a language but poor at writing it. Or a person may understand a language but not necessarily be able to speak it.

In the context of Rwanda, young people display a very different competence between the language skill of listening (understanding) to English as opposed to the language skill of speaking it. Twice as many young people assessed by Harambee can understand English than speak-it at an intermediate (fluent) level.

So, it’s in the application of this shorthand that the problem becomes clear.

Firstly, if young people made the distinction between the four different language skills or abilities they would be better empowered to:

· Realistically assess their spoken English ability;

· Sign-up for English programmes that focus on improving their specific English language skill(s) gaps; and

· Understand the need to, and spend more time speaking English to improve their fluency.

Secondly, providers of English Language Programmes would be better equipped to orientate their programmes to better understand the individual language skill needs of their learners.

Fix number two: to improve spoken English fluency we need to test the different language skills or abilities differently

Making the distinction between the four different language skills or abilities is only the first step towards remediation of spoken English fluency. The next, is to understand a young person’s competence in the different language skills. Assessments provide a picture of this.

Harambee’s own experience of using an English assessment designed to cater for different language skills, highlights the importance of making a distinction.

Using assessment results which measure ability to comprehend English and ability to speak English, Harambee has been better able to:

· Match young people to jobs by understanding the exact mix and level of the different English skills required.

· Offer young people with English fluency gaps, programmes designed to accelerate their spoken English.

Fix number three: to improve spoken English fluency, English programmes need to get young people to speak, speak, speak!

“By the time we got to A-Levels, I would say most of us were A2’s [beginner level on the Common European Framework for Languages scale] in English. This was from learning English as a subject. When we did our A Levels, most of it was listening to what the teacher was teaching us. We didn’t produce English a lot unless we did a presentation which was rare. That was our only opportunity to produce [speak] English:”

These are the words of 27-year-old Clement Nizeyimana recollecting his experience of studying English at school.

Clement lives and works in Kigali as an English Assessor at Harambee. Until he completed his O-Levels in 2009, Clement learnt English as a subject at his public school in Ruhungo District. Clement then completed his A-Levels entirely in English.

Seven-language-speaker Benny Lewis, who left school at age 21, only speaking English despite having studied Gaelic at school for 11 years offers some insight into why Clement, like many other young Rwandans, leave school speaking English at a beginner level:

“The biggest problem people have is they study, study, study and think if I study enough, I’ll just magically speak the language. But it doesn’t work like that. It’s like riding a bike, you have to get on it, [and] start practicing. It’ll be wobbly, but you’ll make progress”.

Young Rwandans complete their secondary school having studied English as a subject. And even at two hours a week, with approximately 25 weeks in a school year, and 12 years of formal schooling most young Rwandans cannot get on the “bike” and ride it.

This is with exposure to around 600 hours of guided English learning. Enough exposure, according to Cambridge University to be able to achieve English fluency.

And this remains true for whether you leave school or university. Harambee’s assessment results show no significant distinction in the level of spoken English fluency between young people leaving secondary school and those graduating from university.

So why can’t young people get on the “bike” and ride it?

Clement’s experience of speaking English is common to many of the young people within Harambee’s network of work seekers. And it is reflective of the Rwandan education system’s tendency to favour receptive language learning rather than the production or expression of English, which would benefit spoken English fluency.

But as Dr Lee Conway, Clinical Psychologist explains: “The medium of education lends itself to receptive language. It’s easier to teach receptive language to large groups of people. It’s harder to teach expressive language. Teaching expressive language is a more individualised process.”

So, to get young people speaking English fluently all educational providers need to double-down on providing young people with spoken English practice.

This may be more difficult to design for and requires a radical shift from the way languages have traditionally been taught, but it is essential for improved spoken English fluency.

And educational institutions, such as universities, provide a unique opportunity to create ideal spaces for young Rwandans to learn to speak English fluently. These are safe spaces where mistakes can be made without fear, and where the culture of speaking English may be encouraged and fostered.

If we plan to fix spoken English fluency in Rwanda, we need everyone to make the distinction between the four different language skills or abilities. This will stop us from speaking about a person’s language ability holistically. And in doing so, we will be able to assess young people’s language competence in these four different skills, empowering them to close their own gaps. And when educators design programmes that aim to measure their success by competence in spoken English fluency, we will begin to see a positive shift in the 10% spoken fluency rate.

As Erin from Education First, says: “Basically, the best way to speak better is to, well — speak!”

So, let’s get English fluency on everyone’s lips.

Sometimes you set your own audacious goal and sometimes your hand is forced by a once-in-a-100-year pandemic.

The goal? Within one month, transition 300 young work-seekers across Rwanda and South Africa from in-person work readiness programmes to mobile learning. Create a curriculum that can be delivered on a standard smartphone, but sensitive to the cost of data. Prepare youth for work in the new COVID-19 reality, while designing for an audience that lacks fluency in English.

Harambee Youth Employment Accelerator has been on a nine-year mission to help marginalised youth into earning opportunities. We didn’t want them held back further by the pandemic, so we were ready for the challenge presented by the lockdowns in South Africa and Rwanda.

Over a four-week period, we pivoted from a face-to-face intervention by designing, developing and implementing a completely online (mobile) solution. And, in doing so, we learnt six key lessons.

Lesson 1: Put young people at the centre. Listen to feedback.

“Good solutions are unlikely to come from within the comforts of a fancy glass office, far removed from the problem at hand. The best social entrepreneurs start by understanding their customers. This means being proximate, a dose of humility, and lots of listening” says Ann Mei Chang, in Lean Impact: How to Innovate for Radically Greater Social Good.

Years of face-to-face interventions meant we had a huge advantage in this respect. We knew our learners well. We understood smartphone penetration and we were aware of their daily struggles with connectivity. We also had a good understanding of our learners’ levels of English.

While this accelerated the initial design of a mobile learning programme, we knew we had to stay proximate. And we had to do this virtually.

We decided to use simple surveys administered immediately on completion of each module. We supplemented this with phone calls to gather more qualitative data about our learners’ experience of mobile learning.

We asked learners variations of the questions “how did you experience that?” and “did we get that right?”

Some learners told us they would like more practice writing professional emails. Many of our learners asked us to extend the English Acceleration curriculum to include more of an emphasis on vocabulary acquisition.

These were just two examples of how being proximate [and listening] paid off. Learner feedback throughout the course meant we could constantly assess the programme’s impact, levels of engagement, ideal curriculum length and preferred learning styles.

Lesson 2: Start small. Iterate.

Harambee subscribes to lean impact design principles when innovating. Central to this methodology is the notion of quick cycles of testing and iterating. The idea being that rapid success and failure allows you to learn in real time, ultimately leading to greater impact.

And in terms of this methodology, the innovation journey starts with a minimum viable product (MVP). The MVP is essentially the quickest prototype released to market for testing and experimentation.

We released our first MVP a week after taking the decision to move to mobile learning. Our MVP got us going, then built momentum and created the space for us to learn from failure instead of fear failure.

Our MVP was a curriculum that spanned one week. It taught our learners how to write professional emails and was loaded onto a learner management system that could be accessed from a mobile phone.

Lesson 3: Design for agility. Listen for feedback.

At Harambee we spend a lot of time reminding each other of Ann Mei Chang’s quote: “fall in love with the problem and not the solution.” And this keeps us agile.

Again, we listened, and we heard some of our learners struggled to sign up to the Learner Management System (LMS). We found out that our learners had vastly different levels of digital literacy in using phone applications. We had instant feedback that our on-boarding of learners required rethinking.

We also kept agile in the space of curriculum design. When our learners told us they liked variety in learning from audio clips, to polls, to quizzes and videos, we made sure we kept designing for this.

But when we tried to maximise the benefits of social learning for millennials by including a lot of discussion forums, our learners responded with “too much and too often”. So we reduced these discussion forums and are on the hunt for new and varied approaches to social learning.

By listening to our users, we learnt that we needed to move away from our first-choice LMS, towards a more engaging platform, one that also allowed for gamification and customized reporting. We have redoubled our efforts to identify and pilot new learner management systems all while aiming to be nimble and responsive.

Lesson 4: Creatively redeploying resources can reap benefits

On-line learning programmes are famously criticised for low completion rates. A study cited by Sarah Cordiner, in her blog entitled “9 ways to increase completion rates online”, found that “students were working individually only 2% completed their course; whereas when there was a mentor present a massive 44% completed their course.”

Our skilled face-to-face facilitators were readily available — and we could re-imagine their role in driving engagement. So we took the decision to pivot the responsibilities of our in-person facilitators to assume a new role which we called an “accountability coach”.

The accountability coach was required to track data on learner performance, assist with learner remediation, check-in with learners and “mark or give feedback” on assignments. And all this support was rendered virtually.

To test the success of this experiment, we collected data from students on their levels of engagement. We also asked students why they felt engaged. We achieved a staggering 95% average engagement score for our learning modules across the mobile learning programme. Learners attributed this to 1) support and access to an accountability coach, 2) feeling part of a cohort and 3) learning with colleagues or social learning. We’re excited about this and hope to build on these findings.

Lesson 5: Flexible learning arrangements can particularly benefit young women

Ernestine Turabayo, a leaner in Kigali, had this to say about her mobile learning experience: “I am so grateful to Harambee, especially for initiating this on-line programme. Now I can follow the courses and still get many new skills, and also take care of my two men. I mean my husband and my son. I am also enjoying the on-line learning because I am not spending as much money in transport”.

In Rwanda, like many other countries, the burden of housework and child care falls on women. We were pleasantly surprised that our mobile learning programme offered a an opportunity for some of our female learners to learn both at convenient times and at a suitable pace.

Ernestine’s story has us thinking differently about our second iteration of the programme. We intend to offer different time slots for live on-line lessons thereby providing for a more gender inclusive learning approach.

Lesson 6: To bridge the digital divide in Africa, we need data light programs

Rwanda has made impressive strides in ensuring cheap connectivity for all — a 1GB of data costs roughly $1 USD, making it one of the cheapest on the African continent. While cheap, this is still not enough to fully engage in online learning.

Our learners loved the engaging content of our videos and podcasts, but complained about the data-heavy content — 30 minutes of video streaming can easily take 1 GB, and a single learning module could demand up to 2GB to complete. Data bundles could take up to 10% of monthly spending — all while earnings are scarce, and the internet is viewed as a luxury and not an essential good.

Again, we listened. Young Rwandans told us they chose to work after 10pm at night because that is when data is considerably cheaper, and in some cases free on certain networks.

A few weeks into the mobile learning programme, our accountability coaches began to spot another trend with some learners in Kigali. Learners seemed to access the online content for short bursts of time repeatedly throughout the day. The reason? They were not used to spending extended periods of time using costly data on their phones.

In response to this, we started a radio broadcast on employability skills as part of our Tuzatsinda campaign, launched to provide reliable information on COVID-19 to young people. Going forward, Harambee plans to use radio as a channel to increase learning and respond to accessibility challenges.

#Datamustfall for young people, and Harambee is keen to support the Rwandan government’s plans for digital inclusion in the years ahead.

So what happens next?

These six lessons have given us much food for thought. Our “v2.0” will be more than just a second iteration of a design for a mobile learning programme.

As Harambee, we have now seen first-hand just some of the benefits of mobile learning . These include scale, replicability, gender inclusivity and the democratisation of learning.

What started as one big and necessary experiment may well become Harambee’s future response to closing young people’s gaps across the African continent.