Actually, my statistical analysis is pretty good. Has to be, both to do what I do for a living, and to keep up in conversations with my physicist bf. And I'd be glad to help with that side of things...but the first thing (and I'm sure you know this) is that lines are meaningless if you don't know what into them. What constitutes a terrorist incident here? Are they source specific, or are they collapsing across various terrorist organizations? Something as simple as whether Sept 11 (to use one that we're all too familiar with) count as 1, 3 or 4 attacks makes a difference.

Also, line graphs like this aren't terribly meaningful in this sort of complex system. Verbal analysis of what is and isn't relevant would be a big help in interpreting what are, effectively, raw numbers. More than that, though...IF you put any faith in these for purposes of prediction (and I'm not sure if they're useful for that), there are a couple of things that throw up some problems for the assertion that the Gulf War Redux has had any impact on the war on terror.

The US Dos number seem more relevant (don't fold domestic terror in...although if Palestinian attacks on Israel are counted as domestic I may be 100% off about that), and according to them...while the number of fatalities dropped (although given the range even in non-spike years I'm not sure it's statistically significant), the number of attacks stays the same between 2002 and 2003 (the closest we can come to before and after measures, I disagree that the 1998-2003 bin is telling...both cause I don't see the dip you talk about around the spike...and the Bush pre-emptive defense doctrine wasn't applied until 2002). Arguably, that means we're not making anything better. Data from 2004 would be particularly interesting, but isn't available until the end of the year apparently. So if you come back with "this doesn't show a long range effective decrease" I can't argue on the numbers. But I would want at least a logical argument to show why it should be expected apriori that the effect would be long range. Most folks seemed to think the effect would be instantaneous. Then again, most folks think Iraq was responsible for 9/11....so there you go.

You might say that we aren't making anything worse, to defend the invasion. But I think that's problematic...as the positive effect on the War on Terror is used to justify the damage we've done. So we have on ballance made something worse, without making Terror better.

Anyway....could you throw me a rope on the folks who put these numbers together in the first place so I can get a better idea about where the numbers come from? Thanks.

G