Inquiries

INQUIRY | INSIGHT | ACTIVATION

Lab engagement starts with a question: ‘What’s going on here?’

The purpose of our inquiries is to gain enough insight to inform action. This lays the foundation for responsible experimentation, in turn driving the development of effective, street-ready and sustainable models.

 
 
 
 

Technology


 
 

Mobile telecoms

Cellphones are ubiquitous, even in poor or informal areas. Yet their application to community safety has arguably delivered below their potential, and many initiatives launched with fanfare have underwhelmed. The Lab conducted a scan of high potential applications, looking both at stand-alone opportunities and the role of technology to enable and broaden access to other safety initiatives.

This research was jointly conducted with Statt, a global crime and conflict consultancy. The resulting insights gave rise directly to our iSafety initiative, as well as informing our work more broadly.

 
 
 
 
 
 

Finance

 

Social impact bonds

In partnership with Dalberg Development Advisors we were the sole successful international applicants in 2013 for Duke MBA support through their CASEi3 program.

We posed the question ‘What’s real about social impact bonds (SIBs)?’ We wanted to separate hype from real potential, with the hypothesis that the value lies as much in the analysis underlying their pricing (who benefits from social outcomes and how can this value be quantified?) as in new flows of social innovation funds.

The Duke team conducted approximately 30 interviews with experts around the world to identify challenges, key success factors and lessons to be applied into finance innovation in safety and security.

 
 
 
 
 

Research

 
 

Violence Correlation Study

One of the earliest projects The Safety Lab did was a violence correlation study following the publication of Census 2011 data by StatsSA. The purpose was to find the strongest correlating factors with violence on a police precinct level looking at the murder rate. The study looked at over 90 different possible variables with a focus on the Western Cape.

The study found that the single strongest correlating factor, in the Western Cape, looked at was unemployment. Precincts with high rates of unemployment strongly correlated with high rates of murder.

Violence vs Unemployment 2014(1).png
 
 

 
 

Nyanga diagnostic

The Lab was asked to join a multi agency program in Nyanga, South Africa’s ‘murder capital’. We conducted a high level diagnostic of local pathways to violence, including analysis of crime stats, mortuary forensics, court inquests and census demographics. This was combined with ethnographic video chronicles to help build an understanding of the complexities of local violence.

Various Western Cape Government departments provided statistics. The Centre for the Study of Violence and Reconciliation (CSVR) supplied raw data underpinning their paper ‘Streets of Sorrow, Streets of Pain’ which dealt in part with the nature of violence in Nyanga. The Institute for Security Studies (ISS) furnished demographic data cut into SAPS police precinct boundaries. Insight Story, an innovative market research company, was commissioned for video research on the ground.

Through this process we identified (among other insights) a clear need for credible and consistent programming for youth over weekends and evenings. This was the nucleus of the program now branded One Comm.

 
 

 
 

Gang Inquiry

One of the most prevalent forms of violence in the Western Cape stems from townships gang activities. However, not enough is known about their varied forms, where they exist, operating structures or relevant interventions. In particular we are seeking to better understand the taxonomy of gangs in different parts of the Cape Flats, in order to inform program design. This links to the second part of the research scan, an assessment of a range of programs across the globe to discern local applicability and design principles.

This research is led by Nathanial Roloff, an expert in gang typology and youth development programs.