Show me the data!

by admin on April 12, 2010

As the Chinese proverb goes: “May you live in interesting times” – and so we are! The data from the ACT assessment (NOT survey!) in Haiti has arrived, and Team Boston has been busy delving into the data and determining what valuable information can be gleaned from the answers to the assessment. Also, we were fortunate to have both Dr. Israel Solbelman from Lincoln Lab and Brigadier General (Ret) Keith Holcomb come to MIT!

Lincoln Lab

Dr. Israel Solbelman is the Division Head of the Homeland Protection and Air Traffic Control at Lincoln Lab.  Dr. Solbelman gave us a brief overview of the history of Lincoln Lab (a detailed version can be found here) and also why Lincoln Lab became involved in Haiti.  It turns out that Lincoln Lab had a radar technology that USSOUTHCOM wanted to use to gather info from Haiti.  The JTF asked that someone from Lincoln Lab went to Haiti, and Marc Zissman volunteered.  The next thing they knew, Marc had “pulled this whole effort together,” and Lincoln Lab was involved to provide “decision support” for actions in Haiti.

Humanitarian Aid/Disaster Response (HADR) Operations and Brigadier General (Ret) Keith Holcomb

Marc Zissman was extremely keen on having Gen. Holcomb come down and talk to us – and one thing we have learned throughout this project is that whatever Marc wants, Marc usually does (hence the fitting description of Marc as  “a force of nature” — Dr. Solbelman).  There is something about Gen. Holcomb that immediately commands attention and respect. It is something in the way he holds himself and speaks that brings one immediately to attention – call it leading charisma.

Gen. Holcomb’s “passion is to get heterogeneously large entities to work together.” His passion could not be better suited for Humanitarian Aid/Disaster Response (HADR) Operations – which is what he came to talk to us about.  When a disaster occurs, many organizations mobilize to come provide aid.  However, the “wicked problem” is that all these different types of organizations come in with their own culture, business, mindset, etc.  How should those groups organize so they all have a positive impact on the situation? Gen. Holcomb showed us a map of the US Interagency Relationships and pointed out the “convoluted command control, and that’s only for the US!”

What each of those organizations does as they respond to the disaster will affect how a country or an area recovers in the future.  There is a particular timeline to disaster response:

Immediate Response (IR) → Early Response (ER) → → → Fixing Failed States (FFS) → Nation Building (NB)

“Everything you do at IR and ER will affect FFS and NB.” — Gen. Holcomb

For example, after the Afghan invasion, over 2400 NGOs poured into Afghanistan from various countries (e.g. France, Germany).  The problem was that most of those organizations did not have people who could speak the local language (Pashto) and thus they had to find local translators. Yet what kind of people speak both Pashto and French/German?  Since the NGOs paid relatively well, educated Afghans went to work as their translators, drivers, etc. “The unattended consequence of their [NGOs] intention to do good is to strip out the intelligencia.” — Gen. Holcomb

Gen. Holcomb told us the story of the 6-blind men and the elephant. This story has been told in many different cultures in various ways, but the gist of it is as follows:

Six blind men walk up to an elephant.  Since they are blind, they touch the creature to determine what it is.  None of them have ever encountered an elephant before, so they each come up with their own conclusion based on the part of the creature they have touched. The one  who touches the leg says it is a tree; the one  who touches the tusk says it is a spear; the one  who touches the trunk says it’s a snake; the one who touches the tail says it’s a rope; the one who touches the ear says it’s a fan;  the one who touches the belly says it’s a wall.  The blind men cannot come to a conclusion as to what the creature is as they are all adamant they are correct in their analysis.

Different people can see the same thing differently. The problem is that most people are so entrenched in their way of thinking that they consider people who think differently as  stupid. Or worse: if the others are not stupid for holding their particular beliefs, then they are evil.

What solution is there? Gen. Holcomb hopes that we can reach a level in information technology where people can make decisions based on the agglomeration of the big picture – the reconstructed elephant based on all the descriptions. In disaster response, if information and knowledge can be made readily available in a form that each person understands, everyone can benefit — from the general who needs to make timely decisions (“key of decisions-making is timing, timing often trumps decision” — Gen. Holcomb) to a local who is trying to search for survivors and is not aware that 50 meters away over 2,000 people are trapped in the rubles.

Data Analysis

This Easter weekend turned out to be absolutely beautiful.  The sun was shining after a few rainy days, and the temperature was just right for the beautiful people of Boston to don their fresh spring outfits (one downside: lots of pale arms and legs in sight… but soon to be doctored by the sun!).

Where was Team Boston? Inside a room at MIT ready to take on the task of cleaning-up and making some sense of the data from the survey in Haiti.  The JTF wanted MIT (i.e. Lincoln Lab) to do the data analysis and provide a briefing on Sunday (Easter!) – quite a fearsome task based on the volume and state of the data. Jarrod Goentzel – our fearless data-team leader – volunteered to step in and help the Lincoln Lab with the analysis of the mounds of data coming in.

And so this is how 10 people (most of them strangers) from Lincoln Lab, MIT, and Tufts joined forces to provide JTF with indicators for 5 different sectors (WASH, Shelter, Food, Security, and Health) in 12 campsites from raw data collected over the past 3 weeks.  For example, the food sector group worked on determining the percentage of people who declared having received food donations within the last 2 weeks.  Had you walked into the room at any time on Saturday, you would have seen everyone hunched over their laptop, frantically manipulating the data and trying to generate graphs for each site.

Overall, we had an intensive and extremely productive day.  We all felt, as Jarrod said, “a strong sense of satisfaction in hard work that helps provide meals and other essential items to our friends in Haiti.” Our efforts did not go unnoticed, and each of the people who worked on the data analysis received a well-written thank you letter from Dr. Solbelman.  Furthermore, Marc took us all out for a thank-you dinner, over which we had the chance to mingle and get to know each other in a relaxed setting! Mèsi!

Leave a Comment

Previous post:

Next post: