One of the best ways to improve the future health of a society is to invest in its youngest members: infants, toddlers and early-school-age children. This is especially true in the developing world, where access to the basic resources that young kids need most to learn and grow cannot be assumed. Hence the international aid community’s increased focus on early childhood development (ECD), resulting in – and accelerated by – the inclusion of ECD among the UN’s Sustainable Development Goals 2030.

The road to achieving this goal, however, can be a rocky one. Allison Zelkowitz, former country director for Save the Children (StC) in Thailand and Lebanon, found that the biggest roadblocks to change were often behavioural, not economic or physical. For example, in Thailand, one of the greatest killers of children is road crashes, and yet extremely few children wear helmets on motorcycles, even when their parents do. To solve this, Zelkowitz and her team collaborated with a behavioural insights advisor, as well as the AIP Foundation, to develop a toolkit for schools, encourage police enforcement of helmet laws and create attractive helmets that kids wanted to wear. This was only one of many initiatives in which Zelkowitz leveraged principles from behavioural science. Ultimately, she decided that this expanding scientific field offered such enormous benefits to the international aid community that she had to learn more.

She began attending conferences and courses in behavioural science, broadly defined as the systematic investigation of what drives human behaviour. Perhaps the trendiest area of inquiry covered under this umbrella is behavioural economics, or “nudging”, which finds light-touch solutions for influencing outcomes based on, for example, subtly manipulating the field of available choices. INSEAD’s Dean of Research Ziv Carmon recently co-authored a paper on how nudges can promote sustainability behaviours – e.g. by using visuals rather than words in pro-recycling messaging. But behavioural science bridges a number of diverse research specialties. Walking INSEAD’s corridors in pre-Covid days, one could greet many behavioural scientists (though they may not call themselves that). Think of Pierre Chandon who has worked on reducing junk food consumption through changing packaging sizes and other nudges, or Zoe Kinias who studies how simple and cheap interventions such as values-based affirmations can insulate women from the damaging effects of workplace gender bias. Another example would be Klaus Wertenbroch, who investigates how advances in AI and automation affect consumer behaviour. Of course, co-author Hilke’s work using fMRI scanners and neuroscience-based interventions to uncover buried nuances of consumer responses to products also belongs in this category.

After a 14-month journey of learning, building leadership support and fundraising, Zelkowitz founded the Centre for Utilising Behavioural Insights for Children (CUBIC), which she describes as StC’s “nudge unit”. Though CUBIC’s launch coincided with the onset of Covid-19, slowing down plans for its debut year, she and her small staff of eight are now working on projects in six Asian countries and two in Africa. For example, in the Philippines, CUBIC is rolling out a programme for parents based on the Tips by Text modality designed by Susanna Loeb of Brown University. Over 40 weeks, participants will receive reminders, activity prompts and notes of encouragement via SMS, in line with StC’s recommendations around positive discipline (parenting via communication rather than punishment), early learning and Covid-19 safety, e.g. handwashing and mask-wearing.

Behavioural science in Rwanda

According to Zelkowitz, too often international aid actors measure the impact of their initiatives via only baseline and endline evaluations, without doing the rigorous testing necessary to attribute causation. Behavioural science techniques allow aid workers to determine which interventions were the most impactful, and which may have under-delivered relative to their cost.

StC’s experience with techniques akin to those used in behavioural science pre-dates – and helped to inform – CUBIC. Caroline Dusabe, a senior ECD specialist for StC based in Rwanda, has seen how even well-resourced interventions can backfire without contextual understanding of behavioural influencers. In one community, Dusabe’s team distributed storybooks to schools, only to later find that the books were hoarded in a storeroom under lock and key instead of being used in class. “The head teacher said that books are precious and should be kept away from the students,” Dusabe explained. “In community behaviour, sometimes there’s a good thing, and at the same time a bad thing that you need to address. We had to make the teachers see that there is a way to value a book and maintain it while also using it.”

In Ghana, StC’s programme of play-based learning had to be adapted when Dusabe and her team discovered that the standard hand-on-head posture of the game Simon Says was associated with mourning and despair in the local culture. “If we had not gone through the listening process, we would have lost them,” she said. “They might have withdrawn from our programme. That is why we invest time in consulting the community, understanding their view.” These efforts often involve ethnographic techniques such as qualitative interviews and analysing the promoters and inhibitors of behavioural change, including in the political environment.

With the help of academics, StC also conducts scientific experiments to quantify the impact of various interventions. For example, Dusabe contributed to a randomised controlled trial (which spawned a 2019 paper in the International Journal of Behavioural Development) in Rwanda to measure the efficacy of First Steps, a 17-week instructional programme for parents of young children delivered via radio. Families were randomly assigned to a “full intervention” group where the programme was complemented by additional training, materials and follow-up visits, a “light-touch” group receiving the programme only and a control group that received no intervention. Surprisingly, although the “full-touch” treatment had the greatest behavioural impact, the no-frills programme also effected significant change (compared to the control arm) despite requiring far fewer resources. The experimental findings thus supported Dusabe’s objective “to have something the government could run at manageable cost”. Currently, she and her co-authors are continuing the study to ascertain whether the “light-touch” group is more likely to lapse into pre-programme behaviours than their full-treatment counterparts.

Making it easy

Covid-19 has only increased the need for cost-efficient, scalable interventions that can be delivered remotely. In communities without heavy smartphone penetration, SMS and radio have proven to be useful delivery systems. As Zelkowitz explains, “One of the core principles of behavioural science is to make it easy.” StC’s Covid-19 nudges fulfil the three criteria for embedding lasting positive habits, as detailed by Scott Young of BVA Nudge Unit in a June 2020 Medium post.

• Reframing: Anchor desired behaviours to purposes and ideas that transcend the temporary crisis of Covid-19. For example, CUBIC’s Philippines programme of pandemic-themed text messages encourages parents to reframe the command to cough or sneeze into your elbow as “Cover your cough or sneeze like you’re an elephant!” Making hygiene fun also makes it more likely that kids will actually develop this lifelong habit.
• Re-enforcing: Provide visual cues in the physical environment to influence decisions in the moment. To promote social distancing in rural Rwanda, StC’s public messaging uses a unit of measurement familiar to all locals, regardless of educational level: “The distance between one person and another should be the length of a cow. It is an easy way to say two metres,” says Dusabe.
• Rewarding: Reward the desired behaviour, so that people unconsciously associate it with a positive feeling. The Tips by Text programme that CUBIC has adapted for the Philippines includes one SMS per week (one-third of the total) devoted to encouraging, empowering messages to parents. Studies of Tips by Text show that this part of the programme was key to retention.

INSEAD gets involved

INSEAD’s behavioural science experts found significant overlap between our “force for good” mission and StC/CUBIC’s willingness to conduct rigorous behavioural science research. Thanks to a call for proposal from the Bernard van Leer Foundation and the Conrad N. Hilton Foundation, we – together with the BVA Nudge Unit – have developed a new education programme for early childhood development leaders with StC and CUBIC on how to leverage behavioural science to facilitate a good start in life for all children. If the health crisis allows, this new programme will start in the fall in Asia, Africa and South America.

This new collaboration has also generated many ideas for joint research programmes on improving parental decision making around health and money so children can reach their highest potential – and the birth of a new cross-disciplinary campus-wide research seminar series around this topic. Exciting times ahead!