An Interview with Les Roberts

ePulse: Awesome. What sort of projects are you working on?

LR: One thing that I’m working on actively over a couple of years in our little group at Columbia is working on getting a method developed - a simple reproducible method for estimating the number of women that are raped in any window of time. Rape is about the hardest thing to interview for. I share an office with a doctoral student who is working on that and so I spend a little of my time and I expect I will spend a significant amount of time over the next couple of years with many, many master’s students going to different places and repeating this method for trying to estimate the incidence of rape in a population.

ePulse: How do you conduct your research on civilian mortality?

LR: Every war is different and so the answer is different in every setting. In Darfur, I attempted to do the survey through an international NGO. In Congo, I did it with the International Rescue Committee who had wonderful local staff; just amazing Congolese staff who made it all possible. In Iraq, it was just essentially me and a professor and we just set up everything we needed in terms of drivers and interviewers and finding out about security without any institutional support. So it is different in every setting.

ePulse: Are there others doing research similar to the civilian mortality research?

LR: Yeah. There are. I wouldn’t say lots. But there are a couple of dozen people who have done this before. And every year there is one or two more wandering into the field.

ePulse: What sort of education did you receive to get into the field?

LR: The education I had was not very relevant to what I am doing, except for introductory classes in statistics and epidemiology. My years of studying physics and engineering and the vast majority of my public health classes did not ... I take that back. The physics actually might have been the most important foundation of what I do now just because the world of physics is about making rules so that you can explain things and describe things either in categories or probabilities and that’s what measuring mortality is about as well.

ePulse: Have you ever considered doing such an analysis in a non-war zone that has high mortality such as South Los Angeles or El Salvador?

LR: The answer is yes. In a place like South Los Angeles, even though you think the mortality is high, you would knock on a lot of doors before you found someone who has died. The death rates we measured in Congo would be many times higher than the death rates in South Los Angeles. That’s thought number one. Thought number two is measuring mortality with surveys is really only a good option when the vital registration process has broken down. In the United States, it is very rare that a death does not get recorded.

So if you want to understand, what are the dynamics, or what are the neighborhoods, or what are the risks in South Los Angeles, associated with someone dying or dying violently, it would be easier and more fruitful to analyze data from a coroner’s office that has already been collected than it would to go knock on doors when nine out of ten houses hasn’t had anyone die in the last five years.

ePulse: This is a last best option.

LR: This is when everything else won’t work.