People Get Ready

[ make levees, not war ]

Brookings Institution research on Katrina

Posted by schroeder915 on December 8, 2005

The Brookings Institution just published a statistical survey of economic indicators revealing the extent to which the metropolitan New Orleans area continues to be impacted by Hurricane Katrina.

The New York Times printed the following Brookings Institution chart in the op-ed pages on Wednesday, summarizing in graphic form a few of the variables tracked by the study’s authors.

The “Katrina Index” study can be downloaded from the Brookings Institution Web site.

Of additional interest (especially to people who love maps), the Brookings Institution recently published “New Orleans After the Storm,” a study on the role that poverty and race played in the Hurricane Katrina catastrophe, how the poor were marginalized to less-desireable and lower-lying areas of New Orleans, and uses the lessons learned to make recommendations on how to rebuild New Orleans.

One gripe I have with the study is the tract-level census boundaries used to measure demographic characteristics of the city. The researchers could have used the relatively more precise block-level boundaries.

BIG KUDOS to the Greater New Orleans Community Data Center for their contribution to the study’s maps.

One could spend a lot of time at the Brookings Institution Web site reading a storm surge of studies about Hurricane Katrina.

11/08/05 update: Alright, I’m returning to this post because, as I was reading through “New Orleans After the Storm,” I found I was offended by the apparent reaching done by the authors to correlate race, income, and flooding. Just eyeball the following map and see if you could conclude that the amount of flooding in predominantly poor neighborhoods was significantly greater than in other neighborhoods:

There may be a correlation, and indeed, a huge swath of New Orleans East isn’t inside the frame (but that area is largely comprised of middle-class blacks). I am not satisfied, however, that the authors did their homework in making the case. I also take offense at the overuse of the racial component of the disaster. I have plenty — PLENTY — of white friends whose homes flooded.

Take a look at the following statement and see if it jibes with what you know about New Orleans’ demographics:

Overall, blacks and other minority residents made up 58 percent of those whose neighborhoods were flooded, though they encompassed just 45 percent of the metropolitan population.

By contrast, whites made up just 42 percent of those who lived in neighborhoods that flooded. In New Orleans the cleavage was even starker. The flooded areas there were 80 percent non-white (pp. 16-17).

In actual fact, according to the 2000 census, blacks comprised 67 percent of the New Orleans population; non-Hispanic whites comprised just 27 percent. Given those numbers, it looks more like flooding disproportionately affected white neighborhoods. Furthermore, given the extent of flooding — 80 percent of the city — I think the authors should have tested the correlation they made for statistical significance.

I’m not saying I don’t agree with the authors that blacks and poor have been geographically marginalized, and I do think the study is well worth reading. I just think the evidence the authors provided is a little sloppy.

More importantly, I think it’s very dangerous to overplay the racial component to the disaster. The flooding affected blacks and whites, wealthy and poor, indiscriminately. I think it’s more important to focus on rebuilding the entire city, with fairness and equality.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

 
%d bloggers like this: