The market research world will forever debate the validity of sampling based research methodologies, and this will only increase as discernable, valuable niches proliferate. But those projection techniques certainly are not limited to marketing. They’ve moved front and center to the Iraq War to estimate fatalities, says the New Scientist:
Around 655,000 people have died in Iraq as a result of the US-led coalition invasion, according to the largest scientific analysis yet. That is 2.5% of the country’s entire population.
The study was conducted by US and Iraqi scientists to determine how many Iraqis have died since the invasion in March 2003.
Various estimates have been made of Iraqi casualties, ranging from 48,000 to 126,000. But these have been based on reporting by the press, hospitals or the military, and tend to underestimate the dead, the researchers claim.
Gilbert Burnham and colleagues at Johns Hopkins University in Baltimore, US, and Al Mustansiriya University in Baghdad, Iraq, surveyed 1849 households with a total of 12,801 inhabitants, in all but two of the 18 governorates across Iraq. The researchers asked about births, deaths and cause of death. They did not discriminate between civilian and combatants.
The death rate before the invasion was a fairly normal 5.5 per thousand people per year. Since March 2003, that figure has averaged 13.2, the researchers found. More worrying, the death rate has risen every year since the invasion: this year reaching 19.8 per thousand people per year, a near-fourfold increase over pre-invasion levels.
Methodologically sound?
Critics commenting on the study say the number of deaths in the families interviewed – 82 reported before the invasion, 547 afterwards – was too few to extrapolate to the whole country. But the researchers insist they have made statistical compensations for their sample size to pre-empt these criticisms.
They estimate that there were at least 392,976 excess deaths – those that would not have occurred, has there been no war – in Iraq since 2003, and possibly as many as 942,636. The research confirmed the results of the same group’s 2004 study.
It’s a shame that we have to rely on statistical projections versus census or absolute population verification when trying to account for who’s alive and who’s dead. But I guess that’s just another dimension of complexity in the quagmire we call Iraq.