I just finished a big research project. Here’s what I wish I’d known at the start.

Jangala’s Impact & Advocacy Lead, Kat Dixon, reflects on learnings from our recent report publication, Digital Lifelines.

At Jangala we recently finished a big research project focused on digital inclusion with people living in temporary accommodation. This was evaluating our Get Box, and understanding the social impact of internet access, resulting in this report. 

We’ve been through our retrospective process, and here are some of my top highlights to share:

Lesson 1: Incentivise at the beginning, especially for snowball recruitment*

*Snowball recruitment is where researchers recruit participants through other participants. In this instance, we interviewed community partners, who then helped us recruit service user participants.

We incentivised our community partners (local charities and services) with a £50 voucher or donation. An upfront incentive really helped us start on the right foot. We had an additional advantage of being introduced by the local council to help build trust, but we had zero no shows and a really good participant pool of end users, all recruited through partners.

In contrast, we tried to do a series of surveys, where survey 1 had no incentive and a £15 incentive for survey 2 and 3. This was definitely a mistake - we didn’t get a great survey response. We didn’t incentivise the first survey because we were worried about blowing our budget. But we learned there’s no point having a budget if we get limited responses. We incentivised all our qual interviews with end users, which was a good move for us. Next time, I’ll incentivise from the start. 

In the wider charity sector, there’s lots of different views on this practice and incentivising can come with pros and cons. At Jangala, we recognised that incentivising can present some bias, but we felt it was important to show respect for the time and energy that participants offer in those interview settings. All of the end user participants were living in temporary accommodation, many had faced a form of recent trauma and most were facing financial hardship. To ask them to offer their time for free felt tone-deaf to their situation.

Lesson 2: Focus on making people comfortable and opening up

Easily the most successful thing we did was making digital inclusion research cards. We made an in-house, budget deck of cards, with a different digital activity written on each. Participants could sort through the cards and tell us what was important to them, eg. ‘Banking, Streaming Video, Calls and Messages’ and what wasn’t eg. ‘Maps and Navigation, Gaming, Podcasts’. 

Using digital inclusion activity cards meant we could skip to the end of digital inclusion conversations, rapidly moving through what people did online and getting to why it was important. This definitely helped us have deeper conversations, and get much richer data. It also helped participants relax and helped shift them into the driving seat of the interviews. It was one of a few different participatory approaches we took to empower people to engage on their terms, and people responded well to the gamified approach. It’s definitely a style of approach I would take again. 

We’ve made the cards into an open source resource which is free to download and use. There is also a blog here focused specifically on the development of the cards if you’d like to read more. 

Lesson 3: Be prepared to flex and change on the fly

We did a ton of prep for our fieldwork, and we had thought through important things - consent, safeguarding processes, how to structure conversations, ethics, ways to help people relax, recruitment, incentives, safety and many other elements. We’d also thought about the specific challenges of the groups we were working with. But during our retro process I realised there were things we could have done better. 

One key example was gender matching; we’d incorrectly assumed that the participants who were survivors of domestic abuse would be female. This meant when we arrived we realised we weren’t gender matched with all of the participants. A bit more prep and talking more to our local partner could have helped address this, perhaps by asking a gender-matched case worker to be present. 

I also reflected that I need to increase my confidence to interrupt interviews or change things on the fly. During one interview, we were in a community centre with flickering lights and one of our participants had shared that he is autistic. I noticed he seemed to be affected by the lights so I asked if he wanted me to switch them off. He declined and I left it. Looking back, I realise with the ‘high status’ role of being a researcher, I could easily have taken the initiative to make him more comfortable and just switched them off, especially when it may well have affected him disproportionately due to neurodivergence. 

Lesson 4: Ask your partners for help, they know so much

One thing I learned a lot about was anonymisation. I sought out best practice standards and asked advice from other researchers, but there’s surprisingly little practical advice about how to anonymise case studies for high risk cases. It was particularly important to keep identities anonymous because many of our participants were survivors of domestic abuse or had experience of the criminal justice system. 

Fortunately, a colleague suggested I ring a partner, who had been particularly helpful to us. We had a great call where we discussed the balance of doing justice to the person and their story, whilst making sure that we keep participants safe. She also explained how perpetrators try to find survivors and the things I needed to focus on obscuring, for example age, gender and number of children, intimate details like a favourite music artist, or a recent bereavement of a specific family member.

We discussed how combinations of identifying details (eg. type of neurodivergence, country of origin, time spent in prison) could be identifying and how revealing gender can make someone very identifiable, due to the much lower presenting numbers of male survivors of domestic abuse.

Later on, our partner also reviewed our case studies to help us double check we hadn't accidentally revealed an identity. Her experience and understanding was invaluable in making sure we did justice to the insights participants shared whilst keeping them safe. I’ll definitely be asking to collaborate more in future.

Lesson 5: Plan what you’re going to do with the data, and be creative with it

Jangala is a small team - I led the work and we had a freelancer researcher supporting us, so we had to be creative about resourcing and make the most of this data. Once we’d been through synthesis and analysis, we ran internal learning sessions and recruited Learning Champions. This meant we got wider reaching engagement for what we’d discovered, and then helped colleagues feel engaged with how they could use the data. 

In our learning cycle, we ended up making some great changes to the product - making the data balance visible to end users, creating video instructions and instructions in different languages, making the power cable longer, creating a whole new set of information pages - all because we had people internally who were bought into doing something practical with the findings.

We also got creative with our findings. We released a report and did an online report launch, but we also created one page case studies and short form information sheets on particular topics. These downloadable assets are all hosted on the report page on the website. The idea is that they pull out relevant information so that someone with an interest in a subject, such as mental health, can understand how digital inclusion is relevant to their work. The aim is that this will help the findings reach more people. 

Lesson 6: Keep learning!

Across the last 12 months, I’ve made mistakes, learned, laughed and been deeply moved from the interactions I’ve had with participants. I take notes regularly, jotting down things that strike me, things that went well and things I missed.

It sounds obvious but when things move fast, it’s important to jot stuff down. And it’s equally important to carve out retrospective moments to integrate that knowledge. Doing retros with a team is more fun, more productive and helps humanise those moments where you wish you’d done it differently.