Tag Archives: Jozi

MERL Tech Jozi Feedback Report

MERL Tech Jozi took place on August 1-2, 2018. Below are some highlights from the post-conference survey that was sent to participants requesting feedback on their MERL Tech Jozi experience. Thirty-four percent of our attendees filled out the post-conference survey via Google Forms.

Overall Experience

Here’s how survey participants rated their overall experience:

Participants’ favorite sessions

The sessions that were most frequently mentioned as favorites and some reasons why included:

Session title Comments
Conducting a Baseline of the ICT Ecosystem – Genesis Analytics and DIAL

 

…interactive session and felt practical. I could easily associate with what the team was saying. I really hope these learnings make it to implementation and start informing decision-making around funding! The presenters were also great.

… interesting and engaging, findings were really relevant to the space.

…shared lessons and insights resonated with my own professional experience. The discussions were fruitful and directly relevant to my line of work.

…incredibly useful.

The study confirmed a lot of my perceptions as an IT developer in the MERL space, but now I have some more solid backup. I will use this in my webinars and consulting on “IT for M&E”

Datafication Discrimination — Media Monitoring Africa, Open Data Durban, Amandla.mobi and Oxfam South Africa

 

Linked both MERL and Tech to programme and focussed on the impact of MERL Tech in terms of sustainable, inclusive development.

Great panel, very knowledgeable, something different to the usual M&E. interactive and diverse.

… probably most critical and informative in terms of understanding where the sector was at … the varied level of information across the audience and the panel was fascinating – if slightly worrying about how unclear we are as an M&E sector.

When WhatsApp Becomes About More Than Messaging – Genesis Analytics, Every1Mobile and Praekelt.org

 

As an evaluator, I have never thought of using WhatsApp as a way of communicating with potential beneficiaries. It made me think about different ways of getting in touch with beneficiaries of programme, and getting them to participate in a survey.

The different case studies included examples, great media, good Q&A session at the end, and I learnt new things. WhatsApp is only just reaching it’s potential in mHealth so it was good to learn real life lessons.

Hearing about the opportunities and challenges of applying a tool in different contexts and for different purposes gave good all-around insights

Social Network Analysis – Data Innovators and Praeklelt.org

 

I was already very familiar with SNA but had not had the opportunity to use it for a couple of years. Hearing this presentation with examples of how others have used it really inspired me and I’ve since sketched out a new project using SNA on data we’re currently gathering for a new product! I came away feeling really inspired and excited about doing the analysis.
Least favorite sessions

Where participants rated sessions as their “least favorite it was because:

  • The link to technology was not clear
  • It felt like a sales pitch
  • It felt extractive
  • Speaker went on too long
  • Views on MERL or Tech seemed old fashioned
Topics that need more focus in the future

Unpack the various parts of “M” “E” “R” “L”

  • Technology across MERL, not just monitoring. There was a lot of technology for data collection & tracking but little for ERL in MERL
  • More evaluation?
  • The focus was very much on evaluation (from the sessions I attended) and I feel like we did not talk about the monitoring, research and learning so much. This is huge for overall programme implementation and continuously learning from our data. Next time, I would like to talk a bit more about how organisations are actually USING data day-to-day to make decisions (monitoring) and learning from it to adapt programmes.
  • The R of MERL is hardly discussed at all. Target this for the next MERL Tech.

New digital approaches / data science

  • AI and how it can introduce biases, machine learning, Python
  • A data science-y stream could open new channels of communication and collaboration

Systems and interoperability

  • Technology for data management between organizations and teams.
  • Integrations between platforms.
  • Public Health, Education. Think of how do we discuss and bring more attention to the various systems out there, and ensure interoperability and systems that support the long term visions of countries.
  • Different types of MERL systems. We focused a lot on data collection systems, but there is a range of monitoring systems that programme managers can use to make decisions.

 Scale and sustainability

  • How to engage and educate governments on digital data collection systems.
  • The debate on open source: because in development sector it is pushed as the holy grail, whereas most other software worldwide is proprietary for a reason (safety, maintenance, continued support, custom solution), and open source doesn’t mean free.
  • Business opportunities. MERL as a business tool. How MERL Tech has proved ROI in business and real market settings, even if those settings were in the NGO/NPO space. What is the business case behind MERL Tech and MERL Tech developments?
Ah ha! Moments

Learning about technology / tech approaches

  • I found the design workshops enlightening, and did not as an evaluator realise how much time technies put into user testing.
  • I am a tech dinosaur – so everything I learned about a new technology and how it can be applied in evaluation was an ‘aha!’

New learning and skills

  • The SNA [social network analysis] inspiration that struck me was my big takeaway! I can’t wait to get back to the office and start working on it.
  • Really enjoyed learning about WhatsApp for SBCC.
  • The qualitative difference in engagement, structure, analysis and resource need between communicating via SMS versus IM. (And realising again how old school I am for a tech person!)

Data privacy, security, ethics

  • Ah ha moment was around how we could improve handling data
  • Data security
  • Our sector (including me) doesn’t really understand ‘big data,’ how it can discriminate, and what that might mean to our programmes.

Talking about failure

  • The fail fest was wonderful. We all theoretically know that it’s good to be honest about failure and to share what that was like, but this took honest reflection to a whole new level and set the tone for Day 2.

I’m not alone!

  • The challenges I am facing with introducing tech for MERL in my organisations aren’t unique to me.
  • There are other MERL Tech practitioners with a journalism/media background! This is exciting and makes me feel I am in the right place. The industry seems to want to gate keep (academia, rigourous training) so this is interesting to consider going forward, but also excites me to challenge this through mentorship opportunities and opening the space to others like me who were given a chance and gained experience along the way. Also had many Aha moments for using WhatsApp and its highly engaging format.
  • Learning that many other practitioners support learning on your own.
  • There are people locally interested in connecting and learning from.
Recommendations for future MERL Tech events

More of most everything…

  • More technical sessions
  • More panel discussions
  • More workshops
  • More in-depth sessions!
  • More time for socializing and guided networking like the exercise with the coloured stickers on Day 1
  • More NGOs involved, especially small NGOs.
  • More and better marketing to attract more people
  • More demo tables, or have new people set up demo tables each day
  • More engagement: is there a way that MERL Tech could be used further to shape, drive and promote the agenda of using technology for better MERL? Maybe through a joint session where we identify important future topics to focus on? Just as something that gives those who want the opportunity to further engage with and contribute to MERL Tech and its agenda-setting?
  • The conversations generally were very ‘intellectual’. Too many conversations revolved around how the world had to move on to better appreciate the value of MERL, rather than how MERL was adapted, used and applied in the real world. [It was] too dominated by MERL early adopters and proponents, rather than MERL customers… Or am I missing the point, which may be that MERL (in South Africa) is still a subculture for academic minded researchers. Hope not.
  • More and better wine!
 Kudos
  • For some reason this conference – as opposed to so many other conferences I have been to – actually worked. People were enthused, they were kind, willing to talk – and best of all by day 2 they hadn’t dropped out like flies (which is such an issue with conferences!). So whatever you did do it again next time!
  • Very interactive and group-focused! This was well balanced with informative sessions. I think creative group work is good but it wouldn’t be good to have the whole conference like this. However, this was the perfect amount of it and it was well led and organized.
  • I really had a great time at this conference. The sessions were really interesting and it was awesome to get so many different people in the same place to discuss such interesting topics and issues. Lunch was also really delicious
  • Loved the lightning talks! Also the breakaway sessions were great. The coffee was amazing thank you Fail fest is such a cool concept and looking to introduce this kind of thinking into our own organisation more – we all struggle with the same things, was good to be around likeminded professionals.
  • I really appreciated the fairly “waste-free” conference with no plastic bottles, unnecessary programmes and other things that I’ll just throw away afterwards. This was a highlight for me!
  • I really enjoyed this conference. Firstly the food was amazing (always a win). But most of all the size was perfect. It was really clever the way you forced us to sit in small lunch sizes and that way by the end of the conference I really had the confidence to speak to people. Linda was a great organiser – enthusiastic and punctual.
Who attended MERL Tech Jozi?

Who presented at MERL Tech Jozi?

 

If you’d like to experience MERL Tech, sign up now to attend in Washington, DC on September 5-7, 2018!

Using WhatsApp to improve family health

Guest post from ​Yolandi Janse van Rensburg, Head of Content & Communities at Every1Mobile. This post first appeared here.

I recently gave a talk at the MERL Tech 2018 conference in Johannesburg about the effectiveness of Whatsapp as a communication channel to reach low-income communities in the urban slums of Nairobi, Kenya and understand their health behaviours and needs.

Mobile Economy Report 2018. Communicating more effectively with a larger audience in hard-to-reach areas has never been easier. Instead of relying on paper questionnaires or instructing field workers to knock on doors, you can now communicate directly with your users, no matter where you are in the world.

With this in mind, some may choose to create a Whatsapp group, send a batch of questions and wait for quality insights to stream in, but in reality, they receive little to no participation from their users.

Why, you ask? Whatsapp can be a useful tool to engage your users, but there are a few lessons we’ve learnt along the way to encourage high levels of participation and generate important insights.

Building trust comes first

Establishing a relationship with the communities you’re targeting can easily be overlooked. Between project deadlines, coordination and insight gathering, it can be easy to neglect forging a connection with our users, offering a window into our thinking, so they can learn more about who we are and what we’re trying to achieve. This is the first step in building trust and acquiring your users’ buy-in to your programme. This lies at the core of Every1Mobile’s programming. The relationship you build with your users can unlock honest feedback that is crucial to the success of your programme going forward.

In late 2017, Every1Mobile ran a 6-week Whatsapp pilot with young mothers and mothers-to-be in Kibera and Kawangware, Nairobi, to better understand their hygiene and nutrition practices in terms of handwashing and preparing a healthy breakfast for their families. The U Afya pilot kicked off with a series of on-the-ground breakfast clubs, where we invited community members to join. It was an opportunity for the mothers to meet us, as well as one another, which made them feel more comfortable to participate in the Whatsapp groups.

Having our users meet beforehand and become acquainted with our local project team ensured that they felt confident enough to share honest feedback, talk amongst themselves and enjoy the Whatsapp chats. As a result, 60% of our users attended every Whatsapp session and 84% attended more than half of the sessions.

Design content using SBCC

At Every1Mobile, we do not simply create engaging copy, our content design is based on research into user behaviour, analytics and feedback, tailored with a human-centric approach to inspire creative content strategies and solutions that nurture an understanding of our users.

When we talk about content design, we mean taking a user need and presenting it in the best way possible. Applying content design principles means we do the hard work for the user. And the reward is communication that is simpler, clearer and faster for our communities

For the U Afya pilot, we incorporated Unilever, our partner’s, behaviour change approach, namely the Five Levers for Change, to influence attitudes and behaviours, and improve family health and nutrition. The approach aims to create sustainable habits using social and behaviour change communication (SBCC) techniques like signposting, pledging, prompts and cues, and peer support. Each week covered a different topic including pregnancy, a balanced diet, an affordable and healthy breakfast, breastfeeding, hygiene and weaning for infants.

Localisation means more than translating words

Low adult literacy in emerging markets can have a negative impact on the outcomes of your behaviour change campaigns. In Kenya, roughly  38.5% of the adult population is illiterate with bottom-of-the-pyramid communities having little formal education. This means translating your content into a local language may not be enough.

To address this challenge for the U Afya pilot, our Content Designers worked closely with our in-country Community Managers to localise the Whatsapp scripts so they are applicable to the daily lives of our users. We translated our Whatsapp scripts into Sheng, even though English and Kiswahili are the official languages in Kenya. Sheng is a local slang blend of English, Kiswahili and ethnic words from other cultures. It is widely spoken by the urban communities with over 3,900 words, idioms and phrases. It’s a language that changes and evolves constantly, which means we needed a translator who has street knowledge of urban life in Nairobi.

Beyond translating our scripts, we integrated real-life references applicable to our target audience. We worked with our project team to find out what the daily lives of the young mothers in Kibera and Kawangware looked like. What products are affordable and accessible? Do they have running water? What do they cook for their families and what time is supper served? Answers to these questions had a direct impact on our use of emojis, recipes and advice in our scripts. For example, we integrated local foods into the content like uji and mandazi for breakfast and indigenous vegetables including ndengu, ngwashi and nduma.

Can WhatsApp can drive behaviour change?

The answer is ‘yes’, mobile has the potential to drive SBCC. We observed an interesting link between shifts in attitude and engagement, with increased self-reported assimilation of new behaviour from women who actively posted during the Whatsapp sessions.

To measure the impact of our pilot on user knowledge, attitudes and behaviours, we designed interactive pre- and post-surveys, which triggered airtime incentives once completed. Surprisingly, the results showed little impact in knowledge with pre-scores registering higher than anticipated, however, we saw a notable decrease in perceived barriers of adopting these new behaviours and a positive impact on self-efficacy and confidence.

WhatsApp can inform the programme design

Your audience can become collaborators and help you design your programme. We used our insights gathered through the U Afya Whatsapp pilot to create a brand new online community platform that offers young mothers in Nairobi a series of online courses called Tunza Class.

We built the community platform based on the three key life stages identified within the motherhood journey, namely pregnancy and birth, newborn care, and mothers with children under five. The platform includes an interactive space called Sistaz Corner where users can share their views, experiences and advice with other mothers in their community.

With a range of SBCC techniques built into the platform, users can get peer support anonymously, and engage field experts on key health issues. Our Responsible Social Network functionality allows users to make friends, build their profile and show off their community activity which further drives overall user engagement on the site. The Every1Mobile platform is built in a way that enables users to access the online community using the most basic web-enabled feature phone, at the lowest cost for our end user, with fast loading and minimal data usage.

Following the site launch in early August 2018, we are now continuing to use our Whatsapp groups so we can gather real-time feedback on site navigation, design, functionality, labelling and content, in order to apply iterative design and ensure the mobile platform is exactly what our users want it to be.

 

MERL Tech Jozi: Highlights, Takeaways and To Dos

Last week 100 people gathered at Jozihub for MERL Tech Jozi — two days of sharing, learning and exploring what’s happening at the intersection of Monitoring, Evaluation, Research and Learning (MERL) and Tech.

This was our first MERL Tech event outside of Washington DC, New York or London, and it was really exciting to learn about the work that is happening in South Africa and nearby countries. The conference vibe was energetic, buzzy and friendly, with lots of opportunities to meet people and discuss this area of work.

Participants spanned backgrounds and types of institutions – one of the things that makes MERL Tech unique! Much of what we aim to do is to bridge gaps and encourage people from different approaches to talk to each other and learn from each other, and MERL Tech Jozi provided plenty of opportunity for that.

Sessions covered a range of topics, from practical, hands-on workshops on Excel, responsible data, and data visualization, to experience sharing on data quality, offline data capture, video measurement, and social network analysis, to big picture discussions on the ICT ecosystem, the future of evaluation, the fourth industrial revolution, and the need to enhance evaluator competencies when it comes to digital tools and new approaches.

Demo tables gave participants a chance to see what tools are out there and to chat with developers about their specific needs. Lightning Talks offered a glimpse into new approaches and reflections on the importance of designing with users and understanding context in which these new approaches are utilized. And at the evening “Fail Fest” we heard about evaluation failures, challenges using mobile technology for evaluation, and sustainable tool selection.

Access the MERL Tech Jozi agenda with presentations here or all the presentations here.

3 Takeaways

One key take-away for me was that there’s a gap between the ‘new school’ of younger, more tech savvy MERL Practitioners and the more established, older evaluation community. Some familiar tensions were present between those with years of experience in MERL and less expertise in tech and those who are newer to the MERL side yet highly proficient in tech-enabled approaches. The number of people who identify as having skills that span both areas is growing and will continue to do so.

It’s going to be important to continue to learn from one another and work together to bring our MERL work to the next level, both in terms of how we form MERL teams with the necessary expertise internally and how we engage with each other and interact as a whole sector. As one participant put it, we are not going find all these magical skills in one person — the “MERL Tech Unicorn” so we need to be cognizant of how we form teams that have the right variety of skills and experiences, including data management and data science where necessary.

It is critical that we all have a better understanding of the wider impacts of technologies, beyond our projects, programs, platforms and evaluations. If we don’t have a strong grip on how technology is affecting wider society, how will we understand how social change happens in increasingly digital contexts? How will we negotiate data privacy? How will we wrestle with corporate data use and the potential for government surveillance? If evaluator understanding of technology and the information society is low, how will evaluators offer relevant and meaningful insights? How do diversity, inclusion and bias manifest themselves in a tech-enabled world and in tech-enabled MERL and what do evaluators need to know about that in order to ensure representation? How do we understand data in its newer forms and manifestations? How do we ensure ethical and sound approaches? We need all the various sectors who form part of the MERL Tech community work together to come to a better understanding of both the tangible and intangible impacts of technology in development work, evaluation, and wider society.

A second key takeaway is that we need to do a better job of documenting and evaluating the use of technology in development and in MERL (e.g., the MERL of ICT4D and MERL of tech-enabled MERL). I learned so much from the practical presentations and experience sharing during MERL Tech Jozi. In many cases, the challenges and learning were very similar across projects and efforts.  We need to find better ways of ensuring that this kind of learning is found, accessed, and that it is put into practice when creating new initiatives. We need to also understand more about the power dynamics, negative incentives and other barriers that prevent us from using what we know.

As “MERL Tech”, we are planning to pull some resources and learning together over the next year or two, to trace the shifts in the space over the past 5 years, and to highlight some of the trends we are seeing for the future. (Please get in touch with me if you’d like to participate in this “MERL of MERL Tech” research with a case study, an academic paper, other related research, or as a key informant!)

A third takeaway, as highlighted by Victor Naidu from the South African Monitoring and Evaluation Association (SAMEA), is that we need to focus on developing the competencies that evaluators require for the near future. And we need to think about how the tech sector can better serve the MERL community. SAMEA has created a set of draft competencies for evaluators, but these are missing digital competencies. SAMEA would love your comments and thoughts on what digital competencies evaluators require. They would also like to see you as part of their community and at their next event! (More info on joining SAMEA).

What digital competencies should be added to this list of evaluator competencies? Please add your suggestions and comments here on the google doc.

MERL Tech will be collaborating more closely with SAMEA to include a “MERL Tech Track” at SAMEA’s 2019 conference, and we hope to be back at JoziHub again in 2020 with MERL Tech Jozi as its own separate event.

Be sure to follow us on Twitter or sign up (in the side bar) to receive MERL Tech news if you’d like to stay in the loop! And thanks to all our sponsors – Genesis Analytics, Praekelt.org, The Digital Impact Alliance  (DIAL) and JoziHub!

MERL Tech DC is coming up on September 6-7, with pre-workshops on September 5 on Big Data and Evaluation and Blockchain and MERL! Register here.

 

 

 

 

 

Check out the agenda for MERL Tech Jozi!

We’re thrilled that the first MERL Tech conference is happening in Johannesburg, South Africa, August 1-2, 2018!

MERL Tech Jozi will be two days of in-depth sharing and exploration with 100 of your peers.  We’ll look at what’s been happening across the multi-disciplinary MERL field, including what we’ve been learning and the complex barriers that still need resolving. We’ll also generate lively debates around the possibilities and the challenges that our field needs to address as we move ahead.

The agenda for MERL Tech Jozi 2018 is now available. Take a look at register to attend!

Register to attend MERL Tech Jozi!

We’ll have workshops, panels, discussions, case studies, lightning talks, demo tables, community building, socializing, and an evening reception with a Fail Fest!

Session areas include:

  • digital data collection and management
  • data visualization
  • social network analysis
  • data quality
  • remote monitoring
  • organizational capacity for digital MERL
  • big data
  • small data
  • ethics, bias and privacy when using digital data in MERL
  • biometrics, spatial analysis, machine learning
  • WhatsApp, SMS, IVR and USSD

Take a look at the agenda to find the topics, themes and tools that are most interesting to you and to learn more about the different speakers and facilitators and their work.

Tickets are going fast, so be sure to snap yours up before it’s too late! (Register here)

MERL Tech Jozi is supported by: