How understanding behaviour leads to innovation: In conversation with the Nudge Lebanon Team

We chat to the Nudge Lebanon team about using creative thinking to overcome challenges in behavioural insight (BI) interventions. They've worked on interventions designed to increase the uptake of vaccinations among the vulnerable, enhance road safety, drive sustainability, and more.

Within each hemisphere, country, or target market, there are unique sets of circumstances, challenges and opportunities. However, some lessons are universal. So, what lessons can we learn from the BI interventions of the Global South?

THE BEHAVIOURAL LANDSCAPE IN LEBANON AND THE GLOBAL SOUTH

The nature of problems that Nudge Lebanon team is dealing with in the Global South are different to that of the Global North. For example, Nudge Lebanon is looking into issues such as gender-based violence, child protection and rebuilding trust between citizens and service providers or the government. It's important to note that many of these problems stem from structural barriers, and that can be disheartening, but there is always room for behavioural science to play a role in easing burdens in times of crisis.

How can we use Behavioural Insights (BI) right now?

Even before the outbreak of COVID-19, Nudge Lebanon expected families “to have difficulties putting their kids in school”, pressure to be placed on healthcare centres, and for basic needs to be scarce. In times of crisis or unrest, it’s up to us to look for ways that BI can ease the burden. With creative thinking, we can find ways to move forward.

 

THE CHALLENGES OF DEPLOYING BI IN DEVELOPING COUNTRIES

The first challenge: data recording and data collection

One of the most laborious aspects of BI in the Global South is data recording and collection, and this can deter individuals and companies from engaging with low-resource countries:

“Before diving into design or implementation, we like to look at baseline data to see what our sample looks like… A lot of [the time] gathering this baseline data is very difficult.”

Lebanon, does not have centralised databases, even within Ministries. Data collection is scarce and in some cases even when data is being collected, it's highly protected. To work around this, Nudge Lebanon has learned to design their own data collection systems and actively collect data first-hand, which in turn means learning to make it work with smaller sample sizes.

Working around data limitations

Nudge Lebanon has done a few BI interventions around increasing road safety. To start off, they were looking to retrieve data such as average speeds and numbers of cars jumping red lights. In a case like this, usually, one of two things happen. First: The Ministry might have the recordings, but no capacity or capability to aggregate the data, so the team would need to watch hours and hours of footage and manually record the data they want to measure. Second: The Ministry isn’t willing to share data because of potential sensitivity. In that case, the team would have to physically go into the field and record everything manually. This method uses more resources, team members and time, plus you can't collect as much data as when there is a digital or automated system to access.

Creative BI in practice

Four years ago, Nudge Lebanon began an intervention to increase the uptake of vaccinations among vulnerable and marginalised families. Outreach teams would go door-to-door to educate parents about the importance of vaccinations. They would also collect data through tablets: how many kids were in each household, whether they had any vaccinations, or if they had defaulted on some of their vaccines. At the end of the visit, the household would receive vouchers to use at their preferred primary healthcare centre to vaccinate their kids.

In principle, it’s sound – you're collecting a lot of data, you're digitising the process and the data is being collected. However, in practice, the social workers found it extremely difficult and time-consuming using the tablets. Some gave up the data collecting part, some persevered and did all their recording digitally, and some reverted to manual data collection. This widened the margin for potential duplications and errors. The lesson here: you may need to work with outreach workers to improve technological and digital literacy before introducing new software or programmes.

To measure the impact of the intervention, Nudge Lebanon needed to collect follow-up data on which families ended up using their vouchers and vaccinating their kids. Here, they also faced a few difficulties. For example, there is no standardised data collection process between the participating primary healthcare centres, each having their own method: some digitally, some manually, some not at all. Private healthcare centres also generally did not share patient information, giving the experiment another potential downward bias.

In the end, Nudge Lebanon had to physically visit every primary healthcare centre involved, collect the data they had on paper or digitally. They would then reconcile that manually with the data collected at the beginning of the intervention to cross-check that the same child observed in the field was vaccinated at the centre – super time and resource consuming.

While the results were robust given the data they had, the team suspects that the success of the intervention might have been underestimated due to these data collection difficulties.

Adapt as you go

The Nudge Lebanon team had to discover these data collection issues first-hand: "Knowing what we know now, we would have first created a standard protocol of data collection, and we would have trained the supply side to collect and record data before even thinking of rolling out the intervention.”

With the right mindset, almost every challenge can be changed into an opportunity: When results were reported, the Ministry was surprised at how haphazard the data collection process is, and as a result, they are now taking measures to try to standardise the process across all the primary healthcare centres.

The second challenge: A scarcity of communication channels

With BI interventions, you always need a communication channel between the agency and the individuals you're dealing with: an SMS, email, letter or verbal prompt. “A lot of the interventions done in the developed world were delivered by letters in the mail. That’s a channel that can’t be used here. In Lebanon, unfortunately, the mailing system doesn't work... so people usually have to resort to SMSs and personal contact.”

With SMSs in particular, Nudge Lebanon’s beneficiaries are bombarded with messages every day, so it can be difficult for their BI work to stand out among the clutter. This means they have to think of more creative ways to communicate with their target audience.

Integrate where you can

You have to use circumstances to your advantage. During the month of Ramadan, a lot of charities in Lebanon host dinners to raise funds. The team was working with one such charity and used their invites to deliver the invitation to their database.

Those invitations had really interesting narratives that were relatable in a behavioural context to our target population. Instead of sending an SMS or making a call, BI was integrated into the value chain of communication already existing for the stakeholder that were engaged. Instead of making the communication chain longer and more hassle-some, by integrating when and where you can, you can increase effectiveness and be a lot more efficient.

In another project, Nudge Lebanon was trying to get patients who were diagnosed with depression to come back for follow up treatments. This particular population was made up of a lot of refugees and vulnerable Lebanese, so communicating with them via mobile phone or through letters was not possible. The team had to use the services of the social worker that was diagnosing these patients to communicate with them and deliver the intervention herself, giving them a card to self-report their mood changes during the day. It was useful because they trusted her, but it meant that the sample size was diminished because the Nudge Lebanon team was relying on just one social worker to deliver the message.

Challenges force you to be innovative

The advantage of the Global North is that they often have more standardised procedures and centralised databases. However, they could still get value from the tools designed and implemented in the Global South, such as the calendar in the example below.

To encourage beneficiaries to vaccinate their children, Nudge Lebanon designed a calendar that was distributed by outreach workers on follow-up visits. The calendar served as an implementation and planning prompt, getting parents to pick a day and location to vaccinate their children. It was also a calendar they could use throughout the year. Some parents actually took their calendars along with them to the primary healthcare centres to have the nurses circle the date of the next scheduled vaccine.

There was another lesson learned from that particular intervention: “the outreach teams were doing multiple visits to households that haven't vaccinated their children in order to prompt them to do so… In that treatment group, a shift was observed and an increase in uptake because in the follow up visit the outreach workers were providing something new”. Instead of the same-old message, the calendar created a medium of interaction between the beneficiary and the outreach team.

The rolling power of positivity

When Nudge Lebanon first started this BI intervention with the calendar tool, outreach workers had already been in the field for quite some time. Many were fatigued, demotivated, or deflated. The team believes that supplying the outreach workers with this new intervention to implement, along with a simple and refreshed tool for engagement, also lifted their morale. This may have contributed to the positive results achieved in the treatment group.

 

In conclusion

While the nature of BI problems that we are dealing with in the Global South are different to that of the Global North, just as they differ from country to country, or target audience to target audience, we can apply a lot of the same learnings if we adapt them to each specific situation. Lessons such as integrating and adding BI value into existing chains of communication, giving outreach workers a physical medium for improved interaction, or giving beneficiaries a tool for long-term use and presence, can be applied to benefit behavioural interventions across the world.