Statistics

The Lancet’s poorly choosen statistics

The Wall Street Journal opinion page has an article about a “study” appearing in the prestigious journal Lancet that purports to estimate the number of deaths in the Iraqi war. It turns out that the statistics were “funded by anti-Bush partisans and conducted by antiwar activists posing as objective researchers” and that the paper’s estimated “death toll was more than 10 times what had been estimated by the U.S. and Iraqi governments, and even by human rights groups.”

The Lancet’s editor, Richard Horton, rushed the paper into print, and was quoted as saying, “This axis of Anglo-American imperialism extends its influence through war and conflict, gathering power and wealth as it goes, so millions of people are left to die in poverty and disease.” The study’s authors have also “failed to follow the customary scientific practice of making [their] data available for inspection by other researchers.”

Peer-reviewed research that is subservient to political ideology? Is it even possible?

Categories: Statistics

2 replies »

  1. Fair comment, but this is not a subject you really want to use to prove a point. These lower figures are based on recorded deaths and everyone who compiles them expects that they are only lower bound estimates because it’s just not possible to actually record all the deaths. The Lancet study on the other hand, which was compiled by people actually visiting the place, rather than commenting from a distance, and conducting standard polling techniques, can probably be considered a higher bound estimate – but mainly because some people don’t like to think that it might be true. Of course it indeed might be true because nobody is actually counting. Likely though, the true result is between the two estimates but in any event most of us, I hope, would consider that even the lower numbers were unacceptable – because those dead people were almost entirely innocent civilians. Another point to consider is that those anti-war and anti-Bush activists were supported by most of the people on the planet, including about 90% in Spanish polls and 75% in UK polls: countries that were part of the coalition. And of course they objected not because they were against the removal of Saddam, (as indeed some pro-Bush activists still say), but that the war was based entirely on lies, would cause massive loss of innocent lives, and was probably more about oil than anything else. These of course are not just activist opinions now, they are facts, however you might sugar-coat them.

  2. I think the study stinks and the bias stinks. However, would note all the above. Also, I think epidemeology of this fancy sort is prone to problems like this.

Leave a Reply

Your email address will not be published. Required fields are marked *