Nudging bystanders to fight sexual harassment: make perpetrators think twice

June 16, 2021

Reading Time: 4 minutes

BRAC, in partnership with The Behavioural Insights Team (BIT) and Bangladesh Road Transport Corporation (BRTC), installed bright yellow posters inside buses so that they would be visible when harassment actually occurs. Here is what we unpacked from that experiment.

Sexual harassment on public transport is an all too common occurrence in Bangladesh and around the world, with the majority of women in many countries – and in some cases over 90% of respondents – having experienced sexual harassment in the past.

Read more: End gender-based violence: Integrated approaches from around the world

Surveying commuters getting off public buses in Dhaka, Bangladesh.

Many of us would like to think we would do something to stop sexual harassment if it happened right in front of our eyes. When we surveyed more than 3,000 commuters getting off public buses in Dhaka, Bangladesh, 9 out of 10 respondents listed at least one helpful action they would take if they saw a woman being harassed on a bus.

Horizontal trial poster.

When trained observers took bus rides in the nation’s capital, we found bystanders took action in only one out of every five cases of sexual harassment. 

How could we close this intention-action gap? And how might we know if it worked? A partnership between the Behavioural Insights Team (BIT), Bangladesh Road Transport Corporation (BRTC) and BRAC tried to see what can be learnt by applying human-centred design and innovative measurement techniques to this challenge.

Read more: 30 ways BRAC prevents violence against women and children in Bangladesh

Active bystander programmes show promise in countering the ‘bystander effect’

The study of bystander inaction dates back to at least the 1960s, when American researchers Bibb Latané and John Darley documented the “bystander effect”.

In a series of experiments, passive bystanders were shown to inhibit others from acting in apparent emergencies. In some cases, this seems to occur because of ambiguity over whether the situation really requires action; if no one else is responding we conclude that we also do not need to respond.

A3 trial poster.

Active bystander campaigns around the world, such as Stand Up and Not On My Bus, have aimed to encourage us all to be allies in stopping harassment by running awareness campaigns and training programmes. A systematic review of 27 studies of bystander training programmes aimed at adolescents and college students found beneficial effects on knowledge, attitudes, self-efficacy and self-reported intervention behaviour.

A project led by the Behavioural Insights’ Sydney team showed that even light touch approaches can work to overcome bystander inaction. In collaboration with the Victorian Health Promotion Foundation, a series of active bystanding emails were sent to close to 30,000 staff and students at Victorian universities. It was found that people who received the emails were 30% more likely to report having taken action in a situation where they observed sexual harassment than a control group who received no emails.

Testing a bystander nudge on public buses Dhaka

The Behavioural Insights Team worked with BRAC to test another light touch approach, this time targeting the millions of individuals who catch public buses every week in Dhaka. Given the importance of ‘making it timely’, bright yellow posters inside buses designed and installed so that they would be visible to all at the moments when harassment actually occurs.

Apart from attracting attention at the right time, the posters helped bystanders to speak up by visualising three simple and safe actions to take: i) engage the person being harassed to show support; ii) politely address the harasser and ask them to move away; and iii) alert the conductor if the situation escalates, so the harasser can be asked to leave the bus.

Trial poster details.

Going one step further, surveys of passengers were combined  with structured observations conducted by trained observers. Together with D2, a Dhaka-based research firm, an app-based survey tool was developed to measure not only changes in awareness, attitudes and willingness to act (as previous work had done) but also actual behaviour.

Sobering new insights about the prevalence of sexual harassment were uncovered in this context. In the week before the posters went up, harassment was observed on six out of every 10 bus trips, with physical acts such as groping and “unnecessary touching” being most common. This amounted on average to one incident every two hours and forty-five minutes.

Women were the subject of harassment in almost all instances, and close to one in 10 women reported experiencing harassment on the bus trip they had just taken. These numbers may even underestimate the risks women routinely face, given the challenges of observing crowded spaces and eliciting candid responses about harassment.

By comparing data collected in the week before the posters went up with data from the week immediately after, a picture was obtained of the impact of the posters. No evidence was found that the posters changed bystander behaviour significantly in the following week. This result is not conclusive enough for us to say for sure that the posters had no effect at all on bystander action, but may reflect additional barriers, such as uncertainty about what constitutes harassment, that the posters failed to overcome.

Read more: Violence Against Women campaigns: Is raising awareness alone enough?

BRAC GBV trial – active bystanding to support victims.

However, other promising changes were observed: 

  • Amongst both men and women, measures of rape myth acceptance fell by over 10%.
  • Men were more likely to report sexual harassment as a safety concern – possibly because the posters made the issue more salient – while the proportion of women reporting it as their biggest concern fell, from 10% to 6% – possibly because the posters provided assurance that harassment would not be tolerated.

Most promising of all, a reduction in the number of harassment incidents per bus ride was observed. Though the size of the change is uncertain and we cannot conclusively attribute all of it to the posters, it seems plausible that salient calls for bystanders to act could make potential perpetrators think twice.

BRAC GBV trial – no incidents per bus trip.

If the threat of active bystanding really can deter individuals from harassing others in the first place, there is a huge opportunity to improve women’s lives.


Have a read of the full report to see how more evidence could be generated to progress the ongoing fight against gender-based violence around the world, and get in touch with the Behavioural Insights Team if you are interested in trying something similar in your part of the world!


Dilhan Perera is a Research Advisor working across the Home Affairs and International programmes, at the Behavioural Insights Team. Laura Litvine heads Behavioural Insights Team’s Paris office, where she leads a team working on a series of social impact projects across the policy spectrum, from health to consumer protection, via sustainability, education, or again public finance.


The Behavioural Insights Team, also known as The Nudge Unit, exists to improve people’s lives and communities. They work in partnership with governments, local authorities, businesses and charities, often using simple changes to tackle major policy problems.

Inline Feedbacks
View all comments