Lt. Col. Robert L. Bateman is a historian, blogger and author of two books about the military. He is stationed in Washington, D.C.

Samuel Clemens, aka “Mark Twain,” famously stated that there were three forms of falsehood: “lies, damned lies and statistics.” I, like many mathematically challenged types, can agree. But there is a special sort of badness associated with comparisons of studies that lean upon numbers.

Before we go any further, let me make something clear. War is bad. Death is bad. OK? Take it as a given that death is bad. I am a soldier; I know this all too well. But death is one thing that is measured in wars. Always has been, always will be. Still, I believe there should be a special level of Hades reserved for people who mess with numbers about death to advance an agenda.

The World Health Organization has just come out with what appears to be a pretty damned thorough study of the total number of Iraqis who have died since our invasion in 2003. This study covered the same topic that the Lancet/Johns Hopkins study of Oct. 11, 2006, covered, to wit, “How many people have died, of all causes, who probably would not have died?” Both studies covered the same time period.

The Lancet/Johns Hopkins study, which came out just before the 2006 U.S. elections, said the number of additional Iraqi deaths by violence was more than 600,000. (Its total number of “additional deaths” was 654,965, of which they said 92% were due to bullets, bombs or airstrikes.) The only previous study on this topic by the same group came out immediately before the 2004 elections.

Now, the WHO report was based upon a study that contacted 9,345 homes in 1,000 “clusters.” That is pretty significant. In fact, the data set is five times as many homes, in 20 times as many areas, as was reportedly used in the Lancet study. Its resultant estimate of deaths from this much broader evaluation is 25% of the Lancet study. In other words, the WHO is estimating 151,000 deaths, in the same period as the Lancet published study (March '03-June '06).

There is one other source of estimates, which is the Iraq Body Count site. That site tracks only violent deaths and only those reported in the media. Its numbers are therefore lower. The Iraq Body Count site, however, gives a running total. It is constantly updating its material, and it doesn’t issue one-time broad reports. Moreover, its methods of collection are by definition non-academic and second-hand – as many of its critics point out.

The problem is that the headlines which came out accompanying stories about the new WHO study. To wit:

The Washington Post version of the story carried the title, “New Estimate of Violent Deaths Among Iraqis is Lower.” That makes sense. The new WHO study is much lower than the only other study to examine total violent deaths by academic statistical sampling methods in Iraq, that of the Lancet published Johns Hopkins study.

The New York Times version of the story was entitled, “W.H.O. Says Iraq Civilian Death Toll Higher Than Cited.”  Huh? Yes, they said “Higher.” Why? Because instead of comparing it to the Lancet study (about which the NYT had previously run a front-page story when it came out, which they have not done for the Iraq Body Count Web site), the NYT compared it to the Iraq Body Count site numbers (which are lower). That is apples and oranges. Downplaying the direct comparison with the only other comparable study (in methodology and process), and instead making a comparison with a dissimilar source is not good reporting, or editing for that matter. It makes no logical sense, and what is worse, it ignites paranoia.

Folks, people who deal with topics like this should not mess with numbers or timing to support a political position (as the Lancet and Johns Hopkins seem to have done), and they should not compare dissimilar items if they are writing articles and headlines for a major newspaper. Apples with apples, oranges with oranges … never apples and oranges, OK? Because if you do the latter, you feed right into the massive and growing distrust of the media as a tool with an agenda.

That is a bad thing for America, no matter who does it.

Send your thoughts to Bob Bateman, whose opinions are his own and do not reflect those of the U.S. government or the armed forces. Bateman was a military fellow at the Center for Strategic and International Studies and has written two books: "Digital War, A View from the Front Lines" (Presidio: 1999) and "No Gun Ri, A Military History of the Korean War Incident" (Stackpole, 2002).