The_Bear

The battle for control of Iraq, in America, is not being fought with weapons or rhetoric; it is being fought with different analysis of numbers. The republicans claim that the troops surge is 'undeniably working', and to this effect, they cite numerous facts; anbar province is now restive when it was once violent, areas where US troops have performed now experience little violence, etc. Contrarily, the Demcrats , with their opposition, cite that attacks against US forces have increased, that civilian deaths have doubled since the beginning of the surge, that nationwide violence isn't down, etc.

Therefore, it is necessary to make a proper analysis of how to understand the relationship between numbers, and the actual situation on the ground.
----

Every analysis must begin with some basic tenet that people cannot deny. In this case, let us talk about the way that data is gathered; data is gathered through, what is essentially, government run programs. Furthermore, no amount of statistical data could ever be completly accurate - one thing that evidences this is the advent of polling, and that it always has 'random error' associated with it.

Now, how do these programs collect the data? The programs collect the data because in the course of the daily lives of the people of a certain country, these people need to enter what I'll call chokepoints - that is, people need to buy and sell goods in a particular place, and social law dictates that these places will not change unless some outside circumstance changes them.
One example is that of a commodity store; these stores are fairly permanent structures, and because of this, this provides a stable place for the buying and selling of goods. This, however, also provides a stable place for the authorities to take a census of what it is that goes on in this place; what goods are bought and sold, if the activites there conform to some sense of right or lawfulness, etc. So it is at these chokepoints that we are ble to truly understand what goes on in those places that are associated with the people around them. That is, in an undisputable manner.

Another point that I wish to bring up before I begin the main point of my argument is simpsons paradox. Simpsons paradox is:

Essentially what this means is that if you take a country that is made up of many states, and if the country elects a leader, you cannot take the overall result of each state and transform that result to the national level. If someone wins by 10% in each state, that does not mean that they will win the national election.
I referance this paradox because this is a necessary point to understand if you are trying to analyze data. If data shows that overall attacks in a country are up, that does not mean that a certain province must have an increase of attacks; it could just mean that there are a few really bad provinces, and other provinces are relatively peaceful. The converse point is true also; if attacks are down nationwide, it doesn't mean that in a certain province attack aren't up.

Ultimately, what needs to be measured are the rates of progress; how fast are the attacks nationwide falling, how fast are the Iraqi forces standing up, how fast is the civic life of the average Iraqi - be he sunni, shiite, or kurd, improving.

Why?

It is pointless and partisan to argue about absolute numbers in anything. If you argue using absolute numbers, you are taking a perspective that lacks context; one example of this is that although the US is the nation with the most guns in the world, it does not have the highest crime rate - this is of course because many of the gun owners in our country are mindful of having guns, and have them not to kill other people. Similarly, you cannot argue that there have been millions of Iraqi's displaced because of the sectarian violence, because they have already been displaced; arguing that policies are bad doesn't take into account if the policies are actually doing something about that situation.
Therefore, ignore statistics about absolute numbers; concentrate instead on the rates of the important statistics. Specifically, how much something happens in a given amount of time; and whether that is decreasing or increasing from month to month. Rates tell the true picture of whether policies are working or not.

Lastly, before I sum up my overall thoughts on the looming Iraq report, there must be an emphasis upon the overall progress that Iraq has made. If we are to talk about Iraq as a nation, if we are to say that Iraq should be one nation, rather than three separate ones, then we cannot focus on any one particular area. Success in one region of the country doesn't mean that we are any closer to achieveing our overall goal.

In sum:
- In order to truly judge whether the surge has worked thus far, you need to take the overall rates of Iraqi society deteriorating, and see if it has increased, decreased, or stayed the same as it was pre-surge. I am unsure of any statistic that could accurately convery this; civilian deaths, usually the greatest marker, could be declining in a particular area because of the increase of homogenaity of the people. Does a decrease in civilian deaths in a particular area correspond more with an increased security presence, or with an increasing homogenization of a neighborhood?

- In addition to the rates of civilian deaths, the rates of peoples perceptions of the biggest ills in their life must also be analyzed. If more sunni's now fear the shiites, or vice versa, then the surge will be having progessively worse effect on prospects of true national reconciliation.

- If we are to follow the Bush/Petraeus strategy of -clear, hold, and transfer- we must look towards how fast the Iraqi people are able to truly take control of the security sitation in that particular region. A useful statistic here might be the amount of combat-capable Iraqi army/police/guard forces; or, the amount of forces from the local militas willing to provide security and not be allied with anti-national government forces (terrorists). It is through analyzing how fast they are stepping up in relation to the amount of violence they encounter, to perform real policing duties, can we extrapolate as to whether Iraq will be a peaceful nation relatively soon - the time when we can leave 'with honor'.

- Regression analysis of broader Iraqi society is what is necessary here, or, regression analysis of how fast the province to province strategy can achieve complete victory. If it is fast enough, the province to province strategy could be victorious.

In my opinion, the necessary statistics and necessary questions of these statsitics, as to their nature if they are increasing or decreasing, and what statistics correlate best with what statistics - they are:
- Civilian deaths
- Sectarian attacks
- Influence of militas/partisans
- Amount of combat capable troops/police
- If militias/tribes are used to keep order, the amount that their ideaology conforms to that of national Iraqi ideaology.
- Ethnic heterogenaity.

At the moment, those are all the statistics I would look at, if I were to really judge whether the surge is working.

The ignorant ask "what is my effect on the world?"
The enlightened ask "What does the world know about itself?"

The Bear

I am The Bear
The Bear

Advertising