GRAPHICALLY DEPICTING THE WAR

Data graphs are becoming something of a hobby/occupation of mine. I have to make them once in a while for work, I frequently have to edit graphs made by others for clarity and content, and I've taken the Edward Tufte class where he trashes PowerPoint (hear! hear!), and is rumored to smash overhead projectors, all in the name of getting graphic artists to think about what they're doing and why they're doing it.

When you make a data graph, you're trying to communicate something--numbers, a trend, a stance or even suggest a course of action. With that in mind, take a look at the following graphs, #1 and #2 (right-click to open them and look at them, but stay here to follow my analysis).

Both graphs are intended to say something about the conduct of the war in Iraq, and whether or not it is succeeding. Neither is, in my opinion, particularly useful scientifically, but pure science isn't really their purpose. They both want to be scientific, but neither actually is: They're both just different ways of counting our war dead, in war that is yet ongoing. And neither takes Afghanistan into account, or includes the dead of 9-11 who are every bit as much casualties of the war as any soldier killed in combat.

But they're very different graphs. Graph #1 starts high, then troughs, then peaks back up before going back down. Graph #2 starts from a zero point, then just goes up in a more or less smooth curve. What are the artists trying to say with these graphs?

Graph #1 is by Dale Amon. Here's his explanation:

The graph is rather striking in its clarity. There are three phases visible. March and April are quite obviously the period of major combat. The second is May; combat deaths plummet to almost nothing while the accident rates skyrocket. The third period is one of minor combat. Accident rates fall drastically but combat deaths climb to a minor peak before tailing off slowly. At present the combat death rate is running an almost insignificant amount over the accident rate.

My interpretation of the graph is:

March and April are clearly the period of major combat.

May is a postcombat month. Remnants of the regime are dispersed and disorganized. There are a lot of dangerous ordinance laying about. Soldiers are tired, ease up slightly and have more accidents because of it.

June through the present is a period of low intensity conflict. One can read the state of the opposing forces in the short-lived secondary peak followed by a long tail off. That tail-off is their journey into oblivion.

It will be interesting to see if the end comes with a bang or a whimper. One could imagine a last desperate and suicidal offensive by the remaining Saddamites. Alternatively, if Saddam is calling the shots and is taken out of the picture the remnants might just quit and go elsewhere. The most likely scenario - in my opinion - is an exponential tail-off in as the remnant forces are killed or captured.

In other words, Dale hopes to "prove" that we're winning the war, by showing that combat deaths are, perhaps temporarily or perhaps permanently, on the decline. It's clear from his language that he wants us to win; his graph intends to lend hope to other who agree with him and change the minds of those who don't.

Graph #2 is by Ed Stephan. Here's his explanation:

Remember how they showed a counter every night while US hostages were being held in Iran? Here's the chart I'd like to see on the nightly news these days. The URLs on the chart and the two events indicated by dots on the line are hyperlinked - just click them.

That's it. He has hyperlinked to the stories depicted on the graph, though to what purpose isn't clear. Linking to the stories merely proves that the events happened, not that they had any actual impact on the war or number of dead. He also links to the names of the dead, which is handy but does not add to the overall appeal or usefulness of the graphic itself. As for the graph, it starts from a zero point and counts the dead. That's all it does--you can't really see anything else in the graph itself. He has to circle the major dates, such as the close of major combat and President Bush's "bring 'em on" challenge, because the graph itself doesn't show them clearly, though they're important to the artist. You could argue that major combat ends when the slope angle changes, but without the "Mission Accomplished" label that he helpfully added, you'd be guessing. Dale's graph, on the other hand, visually shows you where major combat in Iraq ended because there's a significant drop in the casualty rate that corresponds to it. Even without dates applied, Dale's graph is far clearer.

Neither graph is really scientifically valid until the war is over and we have a final count of the dead. Dale's graph may, by then, show lots of zigs and zags as the war enters different phases of combat. Ed's graph, on the other hand, is predictable. The dead will remain dead; more will be added; his graph will go up and up and up until the war ends. But body counts only tell part of the story in any war. US troops killed vastly more Viet Cong soldiers than the VC's killed US GIs, and the US troops never lost a single battle in the field--yet lost the war. Any data graph based on either Dale's or Ed's model would be unlikely to show how this could occur, because the war was lost in the White House and on America's streets.

What is each artist saying in the way he interprets and presents the same data?

Here's my analysis, based on the tone and presentation. Dale wants us to win, and wants the data to show that we are winning by showing that the death rate is slowing. Ed either doesn't want us to win, or doesn't care one way or the other, as his focus is primarily on generating opposition to the war. That's why he's counting the dead in a way that will force the line to always go up. It will look worse than a trend line showing a slowing death trend via peaks and valleys.

Dale's graph is on his site, and is touted at InstaPundit, and elsewhere, largely because likeminded people agree with his premise if not his science. We want to defend the nation and win the war. Ed's graph is touted--you saw it coming--at Democrats.com. It's their masthead graphic as of tonight. What does this say about them?

"We're winning" vs "we're losing." "We want the data to represent victory" vs "we want the data to illustrate a quagmire." Which graph is more honest? Which is more useful? Which is more supportive of the effort to defeat terrorism?

And what would you guess is the agenda of a person who touts one graph as opposed to the other?