No research backs widely shared statistic about gun ownership risk in South Africa

An awareness campaign in South Africa about the dangers of owning a firearm relies on an often-used claim

“From home invasions to hijackings and muggings, many South Africans own a gun out of fear for falling victim to violent crime,” reads the introduction of a press release by the organisation Gun Safe Cities. But living in a house with a gun increases your odds of harm, it claims.

“Research in South Africa shows that you are four times more likely to have your gun used against you than to be able to use it successfully in self-defence.”    

The claim has also appeared in publications by the non-profit organisation Gun Free South Africa. It’s been published on South African sites News24, Brand South Africa, Safer Spaces and IOL in the last couple of years.

Where does this statistic come from? And is it true today? Africa Check went down the rabbit hole.

‘Plenty of research freely available’

There’s little information online about Gun Safe Cities, but it was described as a “charity” by John Davenport. He is the chief creative officer of Havas South Africa, the ad agency that Gun Safe Cities partnered with for the awareness campaign.   

Davenport told Africa Check “there is plenty of research freely available around the ‘four times’ statistic from a number of countries including South Africa”. But he is yet to respond to queries about the specific research they relied on. (Note: We will update this report should he do so.) 

Claim linked to 1999 study

Lizette Lancaster manages the Crime Hub, a source for statistics on crime and safety in South Africa, at the Institute for Security Studies. She told Africa Check the “frequently quoted” statistic comes from studies published by author and crime expert Antony Altbeker in 1999 and 2000.   

Gun Free South Africa, who includes the statistic in their online toolkit, also confirmed that Altbeker was their source. 

We contacted Altbeker. “This is the piece of research that has generated the most reruns of any work I ever did,” he said. 

Altbeker referred us to a chapter he wrote in a 2003 book called Justice Gained? Crime and Crime Control in South Africa's Transition. It sets out the “validity” of his studies and “what can and cannot be claimed about it”, he said.

‘A different claim altogether’

In the first study, Altbeker reviewed 602 police dockets opened for gun-related crimes in the Johannesburg suburbs of Alexandra and Bramley that were reported during the first three months of 1997. 

In 8% of the cases, the victims were armed. And in those cases, nearly 80% of victims lost their guns to their attackers without being able to defend themselves. 

The study also found that those who used their own gun to defend themselves were “four times more likely to have been fired upon by their attackers”.   

This finding is therefore different to the claim made by Gun Safe Cities – that you are “four times more likely to have your gun used against you than be able to use it successfully in self-defence”.

“My study does not support that claim,” Altbeker told Africa Check. “And whatever the limitations of my study, it's got nothing to do with the gun being used against you, that’s a different claim altogether.”

Altbeker revisited the question of victims’ ability to defend themselves in a second study in 2000. But again, “data quality issues”, specifically the possibility that people who lost their guns might have lied to avoid a charge of negligence, meant it was impossible to draw clear conclusions.

Another reason, Altbeker noted in Justice Gained?, is that many who scared off their attackers would not have reported the case to the police.

Study too localised for broad conclusions

Dr Samara McPhedran is the deputy director of the Violence Research and Prevention Program at Griffith University in Australia. She specialises in the study of firearm violence and gun policy. 

McPhedran told Africa Check she is always cautious when using a study that is more than 20 years old. “Typically, patterns of firearm violence change over time, so it is important to work with up to date information whenever possible.”  

And rates of firearm violence and misuse also tend to vary in different places, so it is important to avoid generalisations based on a geographically localised study, she said.

“What may be true of one city, for example, may not be the case in another. The same applies to suburbs. Even within one province, different areas can have very different experiences.” 

Firearm law specialist Martin Hood of MJ Hood and Associates Attorneys in Johannesburg told Africa Check that Altbeker’s study was based on police dockets before the cases had gone to trial, so the contents “were not necessarily tested to determine factual correctness”.  

No recent studies available

“I always think that studies like this suggest the need for more research,” Prof Jonathan Metzl, director of the Center for Medicine, Health, and Society at Vanderbilt University in the US, told Africa Check. 

However, there are “no large databases like we have for all other kinds of injury and death”, said Metzl.

Richard Matzopoulos, a specialist scientist at the South African Medical Research Council's Burden of Disease Research Unit, told Africa Check he was “not aware of more recent studies” about armed victims’ ability to defend themselves. Naeemah Abrahams from the council’s Gender and Health Research Unit said she also had not seen research on ownership and risk within homes from South African studies. 

Conclusion: No recent, reliable data on the risks facing gun owners

Gun Safe Cities, a South African charity, claimed that “research in South Africa shows that you are four times more likely to have your gun used against you than to be able to use it successfully in self-defence.”    

The statistic is often attributed to South African studies published in 1999 and 2000. But the study’s author says his research does not support the charity’s claim. 

Various experts told Africa Check the survey is too old and too inconclusive to make any claims about the risks of owning a gun in South Africa today. 

There is no recent, reliable data on the topic. We therefore rate the claim unproven.