zenpundit.com » statistics

Archive for the ‘statistics’ Category

Jottings 8: of words and numbers

Sunday, May 19th, 2013

[ by Charles Cameron -- nota bene: numbers are my piano, words are my forte ]
.


.

The New Scientist in its feature Charting Syria’s civil war http://syria.newscientistapps.com/index.html claims that it “crunched the numbers on violent events in Syria” and then says “the resulting view suggests that the violence has subsided in recent months”.

Jay Ulfelder in Challenges in Measuring Violent Conflict, Syria Edition at Dart Throwing Chimp suggests “That inference is almost certainly wrong” — and proceeds to say why. Here’s the broad strokes short form:

As Deborah Gerner and Phil Schrodt describe in a paper from the late 1990s, press coverage of a sustained and intense conflicts is often high when hostilities first break out but then declines steadily thereafter. That decline can happen because editors and readers get bored, burned out, or distracted. It can also happen because the conflict gets so intense that it becomes, in a sense, too dangerous to cover.

I’m interested in close reading versus sloppy writing, and from my POV the likelihood of that sort of almost axiomatic decline not being factored into New Scientist‘s conclusions slides in when they write that they “crunched the numbers on violent events in Syria”.

If instead they’d written that they’d “crunched the numbers on reports of violent events in Syria” — wouldn’t it have been a little harder to then write, “the resulting view suggests that the violence has subsided in recent months”?

Wouldn’t it have made more sense to write, “the resulting view suggests that the reports of violence have subsided in recent months”?

I’m sorry, but from where I sit it’s not the numbers, it’s the sloppy language that seems problematic.

Share

A Meditation In Time Of War: “precision”

Wednesday, November 21st, 2012

[ by Charles Cameron -- comparing two species of precision and imprecision found in time of war, one which the camera can record, one which the heart must wait to learn -- let us pray the cease-fire holds ]
.

The key phrases here are “the mosque remained undamaged by the precision strike” and “how many Palestinians were killed and who exactly they were a tough one to answer with precision” — both of which are addressing issues of precision in the course of war.

**

What interests me here is the notion of two kinds of precision — each of them significant, but in different ways.

The IDF wants to publicize the precision with which it takes down its targets, and showing that

the mosque remained undamaged by the precision strike

is clearly preferable to admitting that

Among the Palestinians killed in Gaza this week are the 12 members of the Daloo and Manzar families, including four small children, who died when an Israel Air Force pilot bombed their home by mistake, according to the IDF.

War is not yet perfected.

**

But what of the other type of precision?

After a certain point, numbers simply numb the mind. Eighty-seven died, or ninety-six? When I, several thousand miles distant, read a statistic of this kind, the lack of precision I can tolerate is somewhere in the region of “plus or minus twenty percent”. Thus fifty deaths would differ in my mind from a hundred, but not by much, not by as much as a human life — of which the Talmud, in Sanhedrin 37a, says: –

Whosoever preserves a single soul of Israel, Scripture ascribes to him as if he had preserved a complete world

as is confirmed in Qur’an 5.32:

Therefore We prescribed for the Children of Israel that whoso slays a soul not to retaliate for a soul slain, nor for corruption done in the land, shall be as if he had slain mankind altogether; and whoso gives life to a soul, shall be as if he ha given life to mankind altogether. Our Messengers have already come to them with the clear signs; then many of them thereafter commit excesses in the earth.

Forty-seven killed, fifty-three killed — who notices the difference?

Six “complete worlds”, six times “mankind altogether” lies within the “margin of error” I find it hard to notice.

**

That, in a nutshell, is why I’m a strong Qualit advocate against the pervasive Quantification of modern life.

The eye of the camera may record how precise a given strike was, or conversely show the collateral damage — but it is the eye of the heart which must wait in an agony of suspended grief to know who, what uncle or niece, perhaps at a Sbarro pizzeria two blocks away, may have died.

Share

Switch to our mobile site