Pages

Thursday, May 22, 2008

The "Peace Index" - Moonbattery masquerading as statistics

Yesterday a group called the "Economist Intelligence Unit" came out with a ranking of world countries based on how "peaceful" they are (h/t Michael Fridman for the link.) The US came in 97th out of 140 countries, and Israel came in 136th. It was heavily covered in newspapers worldwide.

They have a very elaborate methodology, taking into account many different factors. The factors themselves seem to be the product of an interesting mindset.

First, they try to make it sound like it is entirely the result of rigorous statistics:
Twenty-four indicators of the existence or absence of peace were chosen by the panel, which are divided into three key thematic categories. Many of the indicators have been "banded" on a scale of 1-5; qualitative indicators in the index have been scored by the Economist Intelligence Unit's extensive team of country analysts, and gaps in the quantitative data have been filled by estimates. Indicators of quantitative data such as military expenditure or jailed population have been normalised on the basis of:

x = (x-Min(x)) / (Max (x) - Min (x))

Where Min (x) and Max (x) are respectively the lowest and highest values in the 140 countries for any given indicator. The normalised value is then transformed from a 0-1 value to a 1-5 score to make it comparable with the other indicators.
Then they go into the specific indicators, which sound sort of reasonable until you dig in a little bit. They measure things like:
Number of external and internal conflicts fought: 2001-06
  • Estimated number of deaths from organised conflict (external)
  • Number of deaths from organised conflict (internal)
  • Level of organised conflict (internal)
  • Relations with neighbouring countries
  • Level of distrust in other citizens
  • Number of displaced people as a percentage of the population
  • Political instability
  • Level of disrespect for human rights (Political Terror Scale)
  • Potential for terrorist acts
  • Number of homicides per 100,000 people
  • Level of violent crime
  • Likelihood of violent demonstrations
  • Number of jailed population per 100,000 people
  • Number of internal security officers and police per 100,000 people
  • Military expenditure as a percentage of GDP
  • Number of armed services personnel per 100,000 people
  • Volume of transfers (imports) of major conventional weapons per 100,000 people
  • Volume of transfers (exports) of major conventional weapons per 100,000 people
  • UN Deployments 2007-08 (percentage of total armed forces)
  • Non-UN Deployments 2007-08 (percentage of total armed forces)
  • Aggregate number of heavy weapons per 100,000 people
  • Ease of access to small arms and light weapons
  • Military capability/sophistication

Then they weight it according to various factors.

The problem is that many of these "indicators" are purely subjective, and when subjective criteria are used to come up with objective data, the results are anything but objective.

In the case of Israel, their breakdown shows exactly where they go wrong. For example, here are some rankings where Israel did poorly according to the EIU:
Level of distrust in other citizens
Qualitative assessment of level of distrust in other citizens. Ranked 1-5 (very low-very high) by EIU analysts
Israel got a 4, on a purely subjective guess based on little knowledge.

Similarly:
Ease of access to weapons of minor destruction
Qualitative assessment of the ease of access to small arms and light weapons. Ranked 1-5 (very low-very high) by EIU analysts.
Israel got a 3 (out of 5). Unmentioned are any controls around the access to these weapons or training in their use, as Israel's handgun deaths are quite low.

Level of organised conflict (internal) - 4
Qualitative assessment of the intensity of conflicts within the country. Ranked 1-5 (very low-very high) by EIU analysts

Respect for human rights - 4
A qualitative measure of the level of political terror through an analysis of Amnesty International's Yearbook.

Potential for terriorist acts - 4
Qualitative assessment of the potential for terrorist acts. Ranked 1-5 (very low-very high) by EIU analysts
Any time it says "qualitative assessment" it is using a fancy word for "guesses based on reading newspapers and Amnesty International reports."

Number of armed services personnel per 100,000 people - 5
Active armed services personnel comprises all servicemen and women on full-time duty in the army, navy, air force and joint forces (including conscripts and long-term assignments from the Reserves)

Aggregate number of heavy weapons per 100,000 people - 5
Source: Bonn International Centre for Conversion (BICC)

Military capability/sophistication - 5
Qualitative assessment of the grade of sophistication and the extent of military research and development (R&D) Ranked 1-5 (very low-very high) by EIU analysts
See the problem here? Thius august group makes an assumption that any country that has a large and sophisticated military must be, inherently, non-peaceful. The logical fallacy of these assumptions are staggering, yet escape this think-tank.

The basic thinking of this group is that armies are inherently evil. This is breathtakingly stupid.

But there is a patina of objectivity around this extraordinarily flawed, and simply wrongheaded, analysis. The media is quick to lap these sorts of things up as if they have any real value.

Even more ironically, the EIU says that one of the biggest reasons for having such an index is to help businesses decide where to set up shop:
Business benefits greatly from an environment of peace. Understanding the attributes of peace allows governments to better understand what they can do to improve the business environment This knowledge allows business to make more confident investment decisions on the basis of actual and predicted stability in a community or nation.
They are pretty clearly saying that companies that choose to do business in Israel are idiots, because of their pseudo-scientific rankings.

Now, who do you trust more to make business decisions: a group that includes Google, IBM, Motorola, Microsoft and Warren Buffet, or the EIU?

The EIU has been doing this sham for a few years now, and one would think that they would adjust their sacred methodology to account for what is obviously a ridiculous conclusion, that Israel is less peaceful than most African nations where tens of thousands die monthly. But they get lots of press, and no one calls them on their basic methodological flaws, so why not keep it going?