Carpe Datum

Terror Graph

So I'm surfing around over at The Sideshow, always full of good links and comments, and I run across a reference to this story, in which someone uses the Terrorism Knowledge Base (which uses data published by Homeland Security) to talk about the rate of attacks. I decided to use the analytical tools myself and came up with this graph of the number of terrorist incidents per year (both domestic and international):

The data prior to 1998 was measured very differently and so doesn't make for a useful continuous graph. So take a good long look at this graph, especially since September 11, 2001 and even more especially since we began the war in Iraq.

So, um, are we winning the Global War on Terror? To be fair, I acknowledge that this chart includes incidents in Iraq, which would not have taken place prior to our invasion. There were 40, 240 and 180 incidents in Iraq in the years 2003, 2004 and 2005, so that's not much of an effect on the overall chart. The graph below shows domestic incidents within the United States only and doesn't include attacks on our troops in Iraq, which would skyrocket the 2003, 2004 and 2005 numbers.

Now, I'll grant you that since 2001, the trend is downward in terms of domestic incidents only. As an honest scientist, it's only fair to point that out, but with numbers ranging from 5-40 per year, the shot noise is enormous. I could also point out that the average number per year under Bush dwarfs what we experienced under Clinton (by any measure, even going to different tools beyond the graph). Of course, analysis like this invites critics to say, "Oh, I suppose you're hoping for more incidents to make your point. You're cheering for more terrorism to prove Bush wrong or make him look bad."

Actually, no. It's just data. We scientists look at charts like this as only one tool in the overall analysis of a situation. Correlations do not mean causations. Just because the terror rates have risen overall or shrunk domestically doesn't mean Bush necessarily was the cause. But there is obviously reason to believe that what we're doing is exacerbating the problem, thanks largely to the War in Iraq.

We won't know for sure just how bad Bush has made the problem for decades. Right now, it is entirely possible that some kid in Iraq just got his parents shot up by a stressed-out soldier for no good reason. What if for the next 30 years he dedicates his life to revenge and sets off a big bomb here in the year 2036? By then, it'll be too late to blame it on Bush.

I guess my point is that it is very difficult to use statistics like this to prove anything. However, the administration loves to tout them (selectively) when they try to measure their success in the War on Terror, so that makes them fair game. A better measure, though, is common sense. And that means asking what the hell we went into Iraq for and why. To me, the question of "Are we safer?" puts the burden of proof on those who want war, not on those who oppose it.

Posted by Observer at February 8, 2006 08:55 AM