Week in Review

Nailing down numbers

Young journalists coming into the business have it drilled into them that every news story must address the “five Ws” — who, what, where, when and why. Some teachers add the “one H” — how.

But there is another “H” that sometimes seems to cause us more trouble than any of the others — how many?

We seem to feel compelled in the news business to attach hard numbers to any story involving human casualties, even when we have limited knowledge of what the real numbers might be.

Years ago, I recall covering the sinking of the ferry Dona Paz in the Philippines. The vessel had been designed to carry 1,400 passengers plus a crew of 50, but was widely acknowledged to have been grossly overcrowded with passengers seeking to make their way to Manila for the Christmas holiday.

Witnesses recalled watching people without tickets leap on board the ferry, even as it was pulling away from the dock at one of the nation’s outlying islands; clearly, no one will ever know how many died when the ferry collided in a fog with the oil tanker Vector.

But that didn’t stop us from trying. Most of the Manila press corps started out with estimates of about 3,000 victims, assuming the ship was carrying about double its listed capacity.

My memory is vague 17 years later, but I recall that number coming down in succeeding days and that we wound up with a consensus estimate of about 1,600, based on nothing in particular that I recall.

So I was a little surprised last week when I did a Google search for Dona Paz and found numbers for the death toll ranging from a low of 1,500 to a high of “around” 5,000. Even more surprising were a number of absurdly precise counts including 4,536, 4,381 and 4,341.

A more recent example of the media’s failures with numbers was the September 11 attacks. The initial stories counted 6,000 dead and then gradually came down during the ensuing months until we reached the current estimate of just under 3,000.

Terror in Russia

As editors, we hate it when we have to lower a previously published death toll. Casualty numbers quite naturally go up — injured persons die of their wounds and new victims are discovered. But barring resurrection of the dead, there is no explanation for a lowered death count except that we got it wrong the first time.

By that measure, we did reasonably well in our coverage of the hostage crisis at a school in the town of Beslan in southern Russia.

Nobody really knew how many people were in the school when a group of terrorist commandos rushed in with explosive belts and rocket-propelled grenades on Wednesday, the first day of the new term. The first reports from the Associated Press said simply that “it was likely that a large number of parents had accompanied their children to class.”

Story Continues →