Archive | Networks RSS for this section

A network visualization of international migration

The UN keeps data on migrations patterns around the world, tracking origin and destination countries and number of migrants (Trends in International Migrant Stock: Migrants by Destination and Origin). I took some time out and created this network visualization of origin and destination countries from 2010. Other years were available, but this is all I had time for.

The size of each node represents the number of countries from which migrants arrive. By far, the most connected country is the United States, accepting more people from more countries than any other place on the planet. Most areas of the network represent geographic regions. Note that Africa is clustered at the top, and pacific island countries are clustered at the bottom.

An interesting result is that countries tend to send migrants to other countries which are only slightly better off than they are. For example, Malawi sends most of its migrants to Zambia and Mozambique, and Zambians go to South Africa, whereas those countries do not reciprocate to countries poorer than them. Wealthy countries tend to be more cosmopolitan in their acceptance of migrants.

Click on the picture to explore a larger version of the graphic.


Cell phone banking in Kenya protects public health

MPESAIt’s rare that I read an academic paper I can get really, really excited about, but this is one of them.

Researchers at Georgetown and MIT have shown that transactions over M-PESA, an African phone banking service can help struggling households when faced with a sudden illness, weather event or economic shock.

We explore the impact of reduced transaction costs on risk sharing by estimating the effects of a mobile money innovation on consumption. In our panel sample, adoption of the innovation increased from 43 to 70 percent. We find that, while shocks reduce consumption by 7 percent for nonusers, the consumption of user households is unaffected. The mechanisms underlying these consumption effects are increases in remittances received and the diversity of senders. We report robustness checks supporting these results and use the four-fold expansion of the mobile money agent network as a source of exogenous variation in access to the innovation.

M-PESA is a cel phone based banking system which allows users to send and receive money to friends and family. Transactions can be small; most users are transferring less than $10 at a time. Users are charge about $.40 to transfer money and a percentage to withdraw. It is free to deposit money into the system.

Anyone can be an M-PESA agent. Starting an M-PESA business requires only a small investment so that even extremely rural areas have access to the system. Agents receive a percentage of transaction costs, and often piggy back it onto existing enterprises such as grocery stores and mobile phone shops. M-PESA not only provides a needed service, but has also created profitable business opportunities for people even in isolated rural areas.

The system is wildly popular. Africans are extremely mobile but maintain deep friend and family networks often spread out over wide distances. When a person has trouble, he or she will often turn to family and friends for financial help.

Previously, people would send money by getting on a bus and travelling, or by sending it with friends who might be going to a particular destination. Transportation costs are high ($5 to go a distance of 200km) and often outweigh the amount to be sent. Sending money by hand also incurred risks of loss to theft and misuse.

The number of M-PESA users has skyrocketed since its introduction in 2007. Nearly all adults in Kenya have access to a cel phone now, and the number of M-PESA users is now 70% of all mobile phone users.

Shocks due to illness or negative weather events such as drought can be devastating for a poor household. A single bout of malaria could set a family back as much as a month’s income or more. When poor households lose money, they don’t get it back and successive events can quickly pile up so much so that families will often wait until illness has become too severe to effectively treat.

Jack and Suri, the researchers who conducted the study found that illness shocks can reduce a households consumption by at least 7%. An average household only consumes around $900 a year, nearly half of which is for food. A 7% reduction in consumption could mean that households will simply eat less given a sudden negative event.

M-PESA users, however, experience no reduction in consumption given a sudden health or economic event. Presumably, the ability to transfer money quickly over long distances provides insurance against disaster. Mutual reciprocation allows the system to effectively function to protect against financial disaster.

This has incredible implications for public health. Financial concerns are an incredible barrier to insuring prompt and effective treatment for diseases such as malaria, diarrheal disease and respiratory infections. An efficient system of moving money creates a broader social insurance scheme, protecting the public against the worst and, hopefully, reducing costly advanced treatments and mortality.

M-PESA is a private sector entity, which was never intended as a public health intervention. However, in an area where public sector health delivery is inefficient, underfunded and most broken, a private sector banking initiative could help bolster availability of life saving drugs (for example) by insuring a consistent flow of money. Shops in extremely isolated rural areas will be more likely to stock malaria drugs if they know that customers have the means to pay for them.

This also has incredible implications for development. One of the pillars of the Millennium Development Goals and the recent Rio+20 Conference on Sustainable Development is to insure that the basic health needs of the poorest people on the planet are met. This cannot happen without addressing the greater problem of financial stability of poor households, which requires the participation of the private sector. Covering basic issues of financial movement, security and access to funds by isolated households is a major step to not only helping households which are disproportionately impacted by health and weather events, but also allows flow of cash to poor regions, bolstering local economies.

Complexity in Markets: a few random thoughts

20130318-174856.jpgI was just checking out an article by Mark Buchanan on Bloomberg about the need to abandon the idea of economic markets as being inherently stable.

For several decades, academics have assumed that the economy is in a stable equilibrium. Distilled into a few elegant lines of mathematics by the economists Kenneth Arrow and Gerard Debreu back in the 1950s, the assumption has driven most thinking about business cycles and financial markets ever since. It informs the idea, still prevalent on Wall Street, that markets are efficient — that the greedy efforts of millions of individuals will inevitably push prices toward some true fundamental value.

Problem is, all efforts to show that a realistic economy might actually reach something like the Arrow-Debreu equilibrium have met with failure. Theorists haven’t been able to prove that even trivial, childlike models of economies with only a few commodities have stable equilibria. There is no reason to think that the equilibrium so prized by economists is anything more than a curiosity.

It’s as if mathematical meteorologists found beautiful equations for a glorious atmospheric state with no clouds or winds, no annoying rain or fog, just peaceful sunshine everywhere. In principle, such an atmospheric state might exist, but it tells us nothing about the reality we care about: our own weather.

This is true. Markets are inherently unstable beasts,as was proven by the crashes of 2000 and 2007/8. Personally, I am an advocate of free markets. The trouble is that no one can agree on what a free market is.

I recently watched a compelling lecture by development economist Ha Joon Chang, where he pointed out (rightly) that “free markets” are truly in the eye of the beholder, pointing out that even the most ardent of free market supporters in 2013 wouldn’t support the free marketers and libertarians who complained of the implementation of child labor laws in the early 20th century.

I should say, then, that I’m an advocate of the “freeest markets within reason” or “the freest markets as will support the moral ideals I hold to be important.” That is, the freeest markets as will support the protection of individual rights to freedom of expression and political thought, the preservation of equal opportunity through education and health, access to capital and social mobility.

Mr. Buchanan points put that where other sciences have accepted that there is no such thing as stability in the rest of the universe, desperate economists and their politically backward fans stick to the idea that, despite evidence of the irrationality of humans in every other space, markets are “self stabilizing.” That humans are rational (they are not) and customers can democratically select optimal prices vs. availability (untrue).

First, I am drawn to the incredible volatility of prices in areas that have the least power to influence them (developing countries).

If there were ever an example of the undemocratic nature of unbridled markets, food in developing countries would be it. Buyers and sellers are legion, yet bodies across the sea set prices with little regard to the demands of the many. In Sub Saharan Africa, stability is a fantastical dream.

Second, I am thinking of the work being done on complex systems in finance, specifically that coming out of Princeton at the moment.

SOME people aren’t waiting around with their heads in the sand, but rather are working to describe the phenomena of finance volatility, noting the increased complexity of financial markets in 2013. It would seem that deeper linkages between financial systems, though necessary, induce the very real problem of volatility. Ignoring it or pretending it doesn’t exist won’t make it go away.

Blaming government regulation and calling for a return to 19th century finance doesn’t work well either.

But that’s enough….

HIV, Networks and the Unknown in 1983

Figure 1. Sexual contacts among homosexual men with AIDS. Each circle represents an AIDS patient. Lines connecting the circles represent sexual exposures. Indicated city or state is place of residence of a patient at the time of diagnosis. “0” indicates Patient 0.

I was doing some research on network theory and found the 1983 paper”Cluster of cases of the acquired immune deficiency syndrome: Patients linked by sexual contact” authored by Dr. David M. Auerbach. It’s a paper which should be commended both for the novelty of its methods and the importance of its results. The authors asked 19 AIDS patients to provide the names of all of their sexual partners. Patients with AIDS were so few at the time, that they quickly spotted several of the names on lists of AIDS patients in other cities. They were then able to produce the graph at the right and it became, perhaps, the first application of network analysis to HIV.

From this, researchers were able to deduce that sexual contact was a likely transmission route, and were able to determine roughly the incubation period of contact to onset of symptoms. This paper provided strong evidence that AIDS was cause by an infectious agent.

It’s truly hard to believe that at one time, the cause of AIDS was largely unknown, even as late as 1983. While the world laughed (I lived in Mississippi), medical professionals and researchers must have felt completely powerless, having no information on what caused the disease, how it was transmitted and no way to treat it. It’s incredible that we come so far, quickly, but a cure will remain elusive for likely decades to come.

The possibility that homosexual men with the acquired immune deficiency syndrome (AIDS) had been sexual partners of each other was studied. Of the first 19 homosexual male AIDS patients reported from southern California, names of sexual partners were obtained for 13. Nine of the 13 patients had sexual contact with one or more AIDS patients within five years of the onset of symptoms. Four of the patients from southern California had contact with a non-Californian AIDS patient, who was also the sexual partner of four AIDS patients from New York City. Ultimately, 40 patients in 10 cities were linked by sexual contact. On the basis of six pairs of patients, a mean latency period of 10.5 months (range seven to 14 months) is estimated between sexual contact and symptom onset. The findlng of a cluster of AIDS patients linked by sexual contact is consistent with the hypothesis that AIDS is caused by an infectious agent.

Network Analysis Exposes Kony 2012 As Right Wing Christian Propaganda

"I want to masturbate in public"

Readers of this blog (all 2 of you) are well familiar of the intense skepticism I hold for evangelical Christian groups operating in Africa (and everywhere else).

Forbes magazine ran a fascinating piece on Social Flow. Social Flow used their network analysis software to track connections between Twitter feeds that all had “#Kony2012” in the text.

What they found was nothing short of illuminating. A visualization of the data can be seen at the bottom of this post.

We expect that connected Twitter users will be linked by geographic region and would expect more connections in large urban areas such as New York and LA. Far from being clustered in metropolises, people promoting Kony 2012 were located in smaller cities, such as Pittsburgh, Dayton, OH, Birmingham, AL and Noblesville, IN (wherever that is):

“The large cluster on the top right includes users from Birmingham Alabama who were some of the earliest to publicize the video. The cluster is substantially larger than the others, leading us to believe that Invisible Children had strong roots in Alabama. Additionally, the hashtag#Kony2012 initially trended in Birmingham on March 1st, a few days before the video was even placed online. Other clusters in the graph include Pittsburgh, Oklahoma City and Noblesville Indiana.” But not only were there geographical clusters, but cultural clusters as well, “This movement did not emerge from the big cities, but rather small-medium sized cities across the Unites States. It is heavily supported by Christian youth, many of whom post Biblical psalms as their profile bios.”

Amazing. Kony 2012 billed itself as a happy accident. The evidence indicates that this was a well coordinated, well funded campaign waged by a powerful religious group. “Stealth Evangelism.”

Talk To Action (thanks Jeff) is a secular watchdog organization devoted to exposing the (in my opinion) damaging and self-serving influence that the religious right has on American politics. They have recently done a series of posts on Invisible Children and the Kony 2012 campaign. The more I read about this and the more I find out about this organization, the weirder it appears to me.

Talk To Action has done some digging and found that Invisible Children receives funding from the Family and other right wing Christian sources. The Family are a powerful, though secretive US fundamentalist group and were behind Uganda’s reprehensible Parliamentary bill which called for the execution of anyone suspected of being a homosexual. Though the group denies this is the case, the evidence against them is fairly damning. It appears though, that the Family are behind much of the recent Kony 2012 craze.

Invisible Children’s founder (and creator of the Kony 2012 video) Jason Russell was heard saying the following, where he admits that IC is using the issue of suffering kids in Africa so that his batshit religious group can gain access to kids in public schools:

“Coming in January we are trying to hit as many high schools, churches, and colleges as possible with this movie. We are able to be the Trojan Horse in a sense, going into a secular realm and saying, guess what life is about orphans, and it’s about the widow. It’s about the oppressed. That’s God’s heart. And to sit in a public high school and tell them about that has been life-changing. Because they get so excited. And it’s not driven by guilt, it’s driven be an adventure and the adventure is God’s.”

This is, of course, before he got caught running around town publicly masturbating.

My Email Inbox

I should be spending more time studying for this stupid test I have to take in May. But screwing around with free network software is much more interesting. In a bid to convince myself that I have a social life, I used NodeXL ( to extract my emails from the past 4 years and was able to draw this cool picture in UCINET 6 ( The software extracts all emails sent to and from me and connects them also through persons CC’d to reconstruct a vast communication network of school and personal emails, spam for academic seminars, football ticket postings and people just writing to complain.  The red dot in the center left is, of course, myself. Colored dots represent distinct groups based on their level of connectedness, most of which represent particular departments that I happened to communicate in, or open mailing lists (hence my mentioning of football tickets sales). Most everyone has sent mail to me only and not cc’d anyone else. Thus, there’s a lot of isolates (people with only one or fewer connections), but the high level of overall connectedness was surprising. Mostly, though, I should get my ass to work.

Social Contact Survey II

Continuing from the previous post, I performed some more analyses of the data to better understand this network analysis business.

First, we need an exploration of basic demographics. I was able to get 63 people to participate in the survey. All but 7 of the people included in the survey were female. Most everyone was from the School of Public Health, which was not surprising since nearly all recruitment occurred through the SPH open student email group. Some recruitment was done through the Rackham Merit Fellowship email group. There were 19 people who were not from the School of Public Health. The largest group represented was Epidemiology with 21 people, 9 from HBHE, 7 from EHS and 7 from Biostat. The other 19 were from other departments.  A department/gender breakdown is as follows:

Respondents were overwhelmingly female. The SPH is also overwhelmingly female implying that is participation was a random act, then we would expect to have more women than men by chance alone. However, I am inclined to think that men are less likely to participate at all if for nothing else, men appeared to have fewer social contacts than women over all. The mean number of contacts for women was 20.34, whereas the mean number of contacts for men was a sad 11.7. Women had a max of 86 contacts. Men scored a paltry 26. The following graphic illustrates the sad social state of masculinity at the SPH:

So basically, if you want to spread some information or perhaps a fun infectious disease, you would have better success to target women rather than men. Specifially, you would want to choose the person who had 86 contacts. According to reliable information, this individual was at a professional conference, and had a reportedly inordinate number of contacts on that particular day.

Using UCINET 6, we were able to calculate the density of the network. 86% of possible connections between people are present within the data. There were 2200 possible connections between individuals within this particular group of people, suggesting that this is a tightly knit social group. Again, this is hardly surprising since participants were recruited out of an already established institutional context. All members of the network were reachable by any other member of the network, indicating that there were no isolates in the entire group. Every one of these 63 people were connected with one another through at least one person in the network.

An analysis of geodesic connections between individuals indicated that the mean shortest path for everyone on the list was 1, that is, it only takes (on average) a single step to get from person A to person B. However, even though average distances may be short, there may be a number of paths to get from person A to person B. If person A has a small number of possible connections to person B, then we would assume that person A and person B are somewhat separated from one another in terms of the level of social connectedness. If person A has many shortest possible connections to person B, then we would assume that person A and person B are highly connected within the social space. Here we see a break down of the mean number of  possible distances between individuals by department:

Overall, the mean number of possible paths between individuals was 1.58. However, we can see that some departments are more connected than others within the group, namely Epid and Biostat, whereas other groups have low number of possible connections, indicating a level of seperateness from the individuals within the survey group. These results were not surprising in that I maintain strong connections to individuals within Biostat and Epidemiology. While I’d like to make some great conclusions about this measure, I can only conclude that my having done recruiting in groups that I am already connected with biases the results somewhat.

This measure is perhaps the most important of all the measures considered in this project. The number of possible paths between people indicate the number of possible paths that an infection may pass through to get from one person to another. Although this group appears to be fairly tightly knit, infections may only have a limited number of possible paths to get from one infected host to one noninfected host.

more later, back to important stuff….

Contact Survey

As it is something I don’t know a whole lot about, I recently got the bright idea to to start working with social network analysis in infection transmission. A search of the literature turned up a few interesting gems, mainly of infection transmission through sexual networks, but little in the way of actual data. There were plenty of boiled down examples of other people’s data, but they don’t post the data for people to play with. I could easily simulate some data. A network analysis software package, UCINET, has a feature to create a random network. However, I felt this to be cheating and desired to get my hands on something real.

In a rare moment of spontaneity, I posted a call for study subjects through the School of Public Health’s open student mailing list. Surprisingly, I got about 65 responses of people willing to expose their contact networks for a day. Picking a single day, I asked the group of volunteers to fill out a form stating whether they had contact any of the other volunteers on the list and how many people they may have had contact with who were not on the list. I was intrigued by how many people were willing to answer the survey and return them to me with out compensation of any kind. I was also surprised at how difficult it is to create a survey that provides you with exactly the type of data you want in the format you wish.

The basic network of people who were on the list looks like this:

At first I was worried that the data might be worthless, due to the lack of overlap in volunteers or possibly due to too much overlap, as might be the case if all of the people on the list have a class together on the study day. However, the network appears intuitive, and knowing the individuals on the list, the clustering present is logical. The circles in the top corner are isolates who had no contact with people in the group. The red dot in the center is me. I had a wide variety of contacts since I was the one doing the survey. Although scientifically, it might not make sense to include my self in the contact network, I do have contact with many of the people on the list regularly, so I could as a member. The clump to the left of me is primarily Epid PhD students, of which I’m a member. It has to be said that they provided the most concise data.

Including the contacts people had that were not on the list, we can see that the results get a little more interesting:

In addition to the cool looking patterns, we can see that many people have contacts well outside the immediate study group. In fact, the people in this study had a mean of  20 contacts per person for the single day. The contact distribution was highly skewed, with some people having as high as 150 contacts for the single day. Contact rates varied by department and by degree. The colored circles represent what’s called “K-cores”, that is groups that are more connected to one another than with other groups. Here, in this case, the K-cores roughly turn out to represent differernt departments represented by the study group. In fact, it’s is fairly surprising how well it pegged individuals into their respective groups. It even positioned me right between Epid and Biostat. I am one of the larger blue dots up top, and Epid (blue) and Biostat (green) are to the left and right of me. HBHE is mostly scattered black dots. The size of the dot represents the relative proportion that the member constitutes of the K-core.

Most of the study group were folks from the School of Public Health. HBHE is by far the most connected of all the departments. Other (a mixture of departments spread around campus) was the least connected of the people who bothered to report their affiliation. Not surprisingly, PhD students reported a lower level of contact than Master’s students, a difference confirmed by a Wilcoxon test with a p-value of .033.


Grad school is marked by long periods of isolation and silence. To make up for a lack of social life (and skills), I, like many others, resort to creating semblances of social circles through social networking sites like Facebook. In doing research for a paper I’m working on, I am reading up on network theory and happened upon some software that allows me to export my friends list from FB with all of the internal connections between them. UICNET ( is a small but powerful program designed to perform social network analysis. Basically, it takes all the connections between people you know and is able to draw them for display using a number of criteria. The raw list read into it ends of looking like this:

I have approximately 300 friends, most of which I actually know. Using a neutral criteria for display, I can see distinct groups from various points of my life. There’s an EMU group in the bottom right corner, a University of Michigan Group on the left, and a huge clump of people that are mostly music related, i.e. pre-grad school. Within the pre-grad school group, I can see distinct clouds of varying time points, Boston, Noise related folks, Mississippi folks and some others. There are some isolates, basically people I know from various disconnected events such as the time I spent in Germany and students of mine from JCC, among others. I find it interesting that Joe Kacemi is my bridge from UM to my cloudy music world. Thom Klepach is my link from EMU to my music cloud. Without Thom and Joe, there would not be a single link between pre-grad school music world and my grad school life.

Using the software to isolate 6 specific “Factions” within the entire list I am able to produce this:

Now, it’s much more clear. There is one group of complete isolates, a random bag that combines Boston and EMU, some odd group consisting of people that I rarely talk to, a Mississippi group, a University of Michigan group, a collection of msuic related people that I knew from Boston and San Francisco (basically, 99-2002) and a group of Michigan music people. Mostly, it makes sense in the grand scheme of my life. The amalgam of Boston, EMU  and Canada is quite strange, however. I think that the level of connections between the major music-related groups is fascinating. The program divides them up into two distinct factions, despite the large number of connections between the two.

Finally, I did an eignevalue analysis and got this one:

The only reason I present this is because, oddly, all three members of Wolf Eyes appear to form the base of the organized list. I’m not sure why. Basically, according to this, all of my social relationships start with Aaron Dilloway.

Racist Model

In trying to decide what courses to take next semester, I did some exploring on complex systems modeling. I created a lame model of two distinct classes of people (or things) who have a certain dislike for one another and prefer to be surrounded with a certain number of like individuals within an arbitrary radius. An distributed number of individuals of both categories are initially randomly placed on a grid. Each individual then scans the area of certain radius around them, figures the percentage of individuals like them relative to the total number of individuals within the scan radius, then makes a decision to move based on an arbitrary threshold. The individual them moves to some randomly chosen open space on the grid within a certain move radius.

I am assuming that the model represents to distinguishable groups who have some dislike for one another. Consider African Americans and white people, Hutus and Tutsis, poor people and rich people, Republicans and Democrats, etc. Upon seeing that an unacceptable percentage of the people around them are of the other category, they then can only move within a certain distance, assuming that resources are limited or moving too far will remove them from some desirable geographic proximity to work, resources, etc.

The model is quite simple, but the results are rather interesting. First we start with a 100 x 100 grid, yielding 10,000 possible occupable spaces. We assume 5000 total individuals, and a 50/50 distribution of each group. Placing them randomly, we obtain an initial grid that looks something like this:

Blue represents an unoccupied space, and yellow and red represent spaces occupied by one of the groups.

I started by assuming that individuals would not tolerate any less than 50 percent representation of their group within a radius of 10 squares. If they happen to occupy a square where the percentage of their own group compared to the total number of individuals within 10 squares is less than 50 percent, they will move to a randomly selected open square somewhere within a 20 square radius of their present position. I repeat this process 25 times. At the end, we see that even after 25 steps, people have already formed segregated clusters of individuals that are not necessarily contiguous. In fact, the entire grid is completely segregated after a mere 10 steps:

Adjusting the parameters a bit, we increase the percentage that people will tolerate to 80%. We can see that given a higher level of “racism” and a dense population, groups have a more difficult time clustering and are thus relegated to a life of constant movement and avoidance, with no resolve. I found this behavior to be true given a smaller population, and even a wider radius of movement. Given a higly level of intolerance for the other group, individuals have a difficult time forming clusters but there is little stability.

Assuming a high tolerance for the other group, leads to the opposite effect, leaving more individuals happy with their present position and less willing to move. This leads to high stability and less segregation, as one would expect.

The “sweet spot” for total segregation appeared to be approximately 50% tolerance. Individuals are happy as long as they make up 50% of the community, but this level of tolerance leads to the highest level of segregation overall. It is assumed that if an individual were to randomly move to an area occupied by the other group, they would immediately move as they felt overwhelmed by the presence of a majority that consisted of individuals of a group other than their own.

I also ran a model assuming that one group only made up for a quarter of the overall population and a moderate level of racism. The results were interesting. The minority group was forced to maintain a nomadic existence while the majority group hardly moved at all. When adjusting for extreme levels of racism, the majority group clusters almost immediately and the whole grid basically becomes a segregated urban area after approximately 25 steps. I call this the Jackson, Mississippi model.

My conclusions were simple and expected, but I was surprised that even this simple model was able to bear them out. High levels of racism lead to high levels of instability but low clustering due to the random nature of movement in the model. Low levels of racism lead to low clustering, but high stability of movement. Moderate levels of racism that we likely see within the US, lead to high levels of segregation and clustering with high levels of stability as people, once they have clustered are unlikely to leave what they consider to be a favorable situation. When creating a less than even distribution of groups, the minority group must maintain a nomadic state of existence, while the majority group remains fixed. Having a high level of racism under these conditions, creates a segregated society as seen above demonstrating the interplay between racist attitudes and imbalance in group representation.

%d bloggers like this: