Photo: RNZ / Nick Monro
The bystander effect, first written about in the 1960s, theorised that the more people witness to a violent incident, the less likely any of them were to intervene.
Earlier this year, a young schoolboy was verbally harassed and then attacked on an Auckland bus. Out of more than 10 people on the bus, only one intervened once the attack turned physical, a 75-year-old man.
But is the bystander effect a real thing? Kirsten Tilleman, a PhD student at the University of Auckland, is a researcher in this area.
"Some researchers who started to do some studies that were repeatable, and found that the more people that were around in an incident of a crime, the less likely someone was to do someone was to step in on the assumption that they might think someone else might do something," she told RNZ's Nights.
The 'bystander effect': Would you step in to help?
However, that theory has lost credence over the years, she said.
"Our expectation that people would help isn't unfounded, where we run into issues might be is people not knowing what to do in that instance.
"So maybe the desire is there to help out, but people don't know what to do."
Her research into the area involved talking to 524 Auckland public transport users, she said.
"What we found in our analysis was that gender was a very key component, the strongest influence on people's responses compared to other different demographic factors that we looked at, like ethnicity and age."
The expectation of others helping out in a violent situation is a big component in what makes us feel safe, she said.
She believes there are steps we can all take to equip ourselves if we witness someone being victimised.
"If we can identify early stages of forms of violence that are less severe, so harassment for instance, where there's less of risk to step in, we can maybe be comfortable in doing that and help prevent an incident from getting to the point where it becomes violent and physical."
Another intervention technique is to ignore the aggressor, she said, and support the person being targeted.
Another is to delegate.
"Go and help, get someone who might be in a better position, particularly position of authority to help."
Documenting the incident is also useful, she said.
"Especially if someone's already helping the situation. You can support it by documenting what's going on. In the context of public transport that's what's the time? What's the vehicle number? What do folks look like? What's going on, that sort of thing."
It is better to interrupt what is happening by focusing on the person being on the receiving end of any harassment, she said.
"Let the person who was harassed know that what happened to them was not okay, and to find out if they need anything from you or how they're doing"
She wants to look deeper into the issue with further research, she said.
"I'm specifically choosing to focus on women and non-binary folks who are part of the rainbow community, and being able to understand that experience and have that better represented in the data, so that the strategies that we use to better secure our public transport systems can be more reflective of some of the most targeted folks, and therefore be safer for everyone."