November 27, 2005
Irrational exuberance plus indelible sniping yields delectable entertainment
In a past entry (which sadly has not yet scrolled off the bottom of the front page - sad because it indicates how infrequently I am posting these days), I briefly discussed the amusing public debate by Barabasi et al. and Souffer et al. over Barabasi's model of correspondence. At that point, I found the exchange amusing and was inclined to agree with the response article. However, let me rehash this topic and expose a little more light on the subject.
From the original abstract of the article posted on arxiv.org by Barabasi:
Current models of human dynamics, used from risk assessment to communications, assume that human actions are randomly distributed in time and thus well approximated by Poisson processes. In contrast, ... the timing of many human activities, ranging from communication to entertainment and work patterns, [are] ... characterized by bursts of rapidly occurring events separated by long periods of inactivity. Here we show that the bursty nature of human behavior is a consequence of a decision based queuing process: when individuals execute tasks based on some perceived priority, the timing of the tasks will be heavy tailed, most tasks being rapidly executed, while a few experience very long waiting times.
(Emphasis is mine.) Barabasi is not one to shy away from grand claims of universality. As such, he epitomizes the thing that many of those outside of the discipline hate about physicists, i.e., their apparent arrogance. My opinion is that most physicists accused of intellectual arrogant are misunderstood, but that's a topic for another time.
Stouffer et al. responded a few months after Barabasi's original idea, as published in Nature, with the following (abstract):
In a recent letter, Barabasi claims that the dynamics of a number of human activities are scale-free. He specifically reports that the probability distribution of time intervals tau between consecutive e-mails sent by a single user and time delays for e-mail replies follow a power-law with an exponent -1, and proposes a priority-queuing process as an explanation of the bursty nature of human activity. Here, we quantitatively demonstrate that the reported power-law distributions are solely an artifact of the analysis of the empirical data and that the proposed model is not representative of e-mail communication patterns.
(Emphasis is mine.) In this comment, Stouffer et al. strongly criticize the data analysis that Barabasi uses to argue for the plausibility and, indeed, the correctness of his priority-based queueing model. I admit that when I first read Barabasi's queueing model, I thought that surely the smart folks who have been dealing with queueing theory (a topic nearly a century old!) knew something like this already. Even if that were the case, the idea certainly qualifies as interesting, and I'm happy to see a) the idea published, although Nature was likely not the appropriate place and b) the press attention that Barabasi has brought to the discipline of complex systems and modeling. Anyway, the heart of the data-analysis based critique of Barabasi's work lies in distinguishing two different kinds of heavy-tailed distributions: the log-normal and the power law. Because of a heavy tail is an asymptotic property, these two distributions can be extremely difficult to differentiate when the data only spans a few orders of magnitude (as is the case here). Fortunately, statisticians (and occasionally, myself) enjoy this sort of thing. Stouffer et al. employ such statistical tools in the form of Bayesian model selection to choose between the two hypotheses and find the evidence of the power law lacking. It was quite dissatisfying, however, that Stouffer et al. neglected to discuss their model selection procedure in detail, and instead chose to discuss the politicking over Barabasi's publication in Nature.
And so, it should come as no surprise that a rejoinder from Barabasi was soon issued. With each iteration of this process, the veneer of professionalism cracks away a little more:
[Stouffer et al.] revisit the datasets [we] studied..., making four technical observations. Some of [their] observations ... are based on the authors' unfamiliarity with the details of the data collection process and have little relevance to [our] findings ... and others are resolved in quantitative fashion by other authors.
In the response, Barabasi discusses the details of the dataset that Stouffer et al. fixated on: that the extreme short-time behavior of the data is actually an artifact of the way messages to multiple recipients were logged. They rightly emphasize that it is the existence of a heavy tail that is primarily interesting, rather than its exact form (of course, Barabasi made some noise about the exact form in the original paper). However, it is not sufficient to simply observe a heavy tail, posit an apparently plausible model that produces some kind of such tail and then declare victory, universality and issue a press release. (I'll return to this thought in a moment.) As a result, Barabasi's response, while clarifying a few details, does not address the fundamental problems with the original work. Problems that Stouffer et al. seem to intuit, but don't directly point out.
While the rebuttal suggests the data is a better fit for the lognormal distribution, I am not a big believer in the fit-the-data approach to distinguish these distributions. The Barabasi paper actually suggested a model, which is nice, although the problem of how to verify such a model is challenge... This seems to be the real problem. Trust me, anyone can come up with a power law model. The challenge is figuring out how to show your model is actually right.
That is, first and foremost, the bursty nature of human activity is odd and, in that alluring voice only those fascinated by complex systems can hear, begs for an explanation. Second, a priority-based queueing process is merely one possible explanation (out of perhaps many) for the heaviness and burstiness. The emphasis is to point out that there is a real difficulty in nailing down causal mechanisms in human systems. often the best we can do is concoct a theory and see if the data supports it. That is, it is exceedingly difficult to go beyond mere plausibility without an overwhelming weight of empirical evidence and, preferably, the vetting of falsifiable hypotheses. The theory of natural selection is an excellent example that has been validated by just such a method (and continues to be). Unfortunately, simply looking at the response time statistics for email or letters by Darwin or Einstein, while interesting from the socio-historical perspective, does not prove the model. On the contrary: it merely suggests it.
That is, Barabasi's work demonstrates the empirical evidence (heavy-tails in the response times of correspondence) and offers a mathematical model that generates statistics of a similar form. It does not show causality, nor does it provide falsifiable hypotheses by which it could be invalidated. Barabasi's work in this case is suggestive but not explanatory, and should be judged accordingly. To me, it seems that the contention over the result derives partly from the overstatement of its generality, i.e., the authors claims their model to be explanatory. Thus, the argument over the empirical data is really just an argument about how much plausibility it imparts to the model. Had Barabasi gone beyond suggestion, I seriously doubt the controversy would exist.
Considering the issues raised here, personally, I think it's okay to publish a results that is merely suggestive so long as it is honestly made, diligently investigated and embodies a compelling and plausible story. That is to say that, ideally, authors should discuss the weakness of their model, empirical results and/or mathematical analysis, avoid overstating the generality of the result (sadly, a frequent problem in many of the papers I referee), carefully investigate possible biases and sources of error, and ideally, discuss alternative explanations. Admittedly, this last one may be asking a bit much. In a sense, these are the things I think about when I read any paper, but particularly when I referee something. This thread of thought seems to be fashionable right now, as I just noticed that Cosma's latest post discusses criteria for accepting or rejecting papers in the peer review process.
November 13, 2005
Graduate students are cool.
The sign of a good teacher is being able to convey the importance of their subject in a manner that engages their audience, making them walk away knowing (not just believing) that they've learned something valuable and novel, and wondering what other interesting things may lay down the path just revealed to them. I like to call this the "Wow"-factor, and it's what gets people engaged in a subject, whether they are just beginning their intellectual journey or have worn through several pairs of shoes already (although, probably for different reasons and in different ways).
The Value of Control Groups in Causal Inference (and Breakfast Cereal) by Gary King of The Social Science Statistics Blog, which I may have to add to my regular list now. King describes a great way to teach a fundamentally important piece of knowledge - the importance of a null-model (or a control group).
A few years ago, I taught the following lesson in my daughter's kindergarden class and my graduate methods class in the same week. It worked pretty well in both. Anyone who has a kid in kindergarten, some good graduate students, or both, might want to try this. It was especially fun for the instructor.
To start, I hold up some nails and ask "does everyone likes to eat nails?" The kindergarten kids scream, "Nooooooo." The graduate students say "No," trying to look cool. I say I'm going to convince them otherwise.
King also recommends Teaching Statistics: A Bag of Tricks for those of us interested in more compelling demos to dislodge our students from their cynicism.
November 09, 2005
The beauty of automation
I should really blog at much greater length about the beauty of the ability to automate simple (or, if you're clever, extremely complex) tasks through computer programming. Of course, with beauty comes ugliness. Every systems administrator (or at least every single one that I've ever known) will tell you that will being able to automate the maintenance of their various computers is wonderful, it is also what allows one malicious person to write a program that exploits a software vulnerability and thus enable any 13-year old "script kiddie" to be just as dangerous without half the technical knowledge.
But that point will have to wait until later. Tonight, I discovered Yahoo!'s FareChaser, which automates the searching of several airfare websites (the fine print indicates that it is only "participating partners" which suggests that money is changing hands for this service; however, the fact that orbitz.com is on the list muddies this hypothesis somewhat since Orbitz is ostensibly the same kind of automated search (what a lovely idea, search engines searching each other for results)). Why is this tool any better than the old crop of such clearing houses that have been around for years now? Because it appears to search the airlines' websites themselves, which, in my broad experience as a frequent flier, often have cheaper flights at different times than places like Orbitz or Travelocity. So, here's to technology making life even more convenient than it is now and saving me the time of hitting those websites individually.
On a related note, I like that Yahoo!, Microsoft and Google are all competing quite vigorously to create compelling online applications for users (typically using Ajax, a really great hack of a technology). All the better for us, and, ultimately, the better for them, too. No one likes a stagnating, entrenched corporation bent on extracting ever greater revenue from the same (formerly compelling but now just prosaic) offerings. Oh, I'm not thinking of anyone in particular. Really.
November 06, 2005
Finding your audience
Some time ago, a discussion erupted on Crooked Timber about the ettiquete of interdisciplinary research. This conversation was originally sparked by Eszter Hargittai, a sociologist with a distinct interest in social network analysis, who complained about some physicists working on social networks and failing to appropriately cite previous work in the area. I won't rehash the details, since you can read them for yourself. However, the point of the discussion that is salient for this post is the question of where and how one should publish and promote interdisciplinary work.
Over the better half of this past year, I have had my own journey with doing interdisciplinary research in political science. Long-time readers will know that I'm referring to my work with here, here and here). In our paper (old version via arxiv), we use tools from extremal statistics and physics to think carefully about the nature and evolution of terrorism, and, I think, uncover some interesting properties and trends at the global level. Throughout the process of getting our results published in an appropriate technical venue, I have espoused the belief that it should either go to an interdisciplinary journal or one that political scientists will read. That is, I felt that it should go to a journal with an audience that would both appreciate the results and understand their implications.
This idea of appropriateness and audience, I think, is a central problem for interdisciplinary researchers. In an ideal world, every piece of novel research would be communicated to exactly that group of people who would get the most out of learning about the new result and who would be able to utilize the advance to further deepen our knowledge of the natural world. Academic journals and conferences are a poor approximation of this ideal, but currently they're the best institutional mechanism we have. To correct for the non-idealness of these institutions, academics have always distributed preprints of their work to their colleagues (who often pass them to their own friends, etc.). Blogs, e-print archives and the world wide web in general constitute interesting new developments in this practice and show how the fundamental need to communicate ideas will co-opt whatever technology is available. Returning to the point, however, what is interesting about interdisciplinary research is that by definition it has multiple target audiences to which it could, or should, be communicated. Choosing that audience can become a question of choosing what aspects of the work you think are most important to science in general, i.e., what audience has the most potential to further develop your ideas? For physicists working on networks, some of their work can and should be sent to sociology journals, as its main contribution is in the form of understanding social structure and implication, and sociologists are best able to use these discoveries to explain other complex social phenomena and to incorporate them into their existing theoretical frameworks.
In our work on the statistics of terrorism, Maxwell and I have chosen a compromise strategy to address this question: while we selected general science or interdisciplinary journals to send our first manuscript on the topic, we have simultaneously been making contacts and promoting our ideas in political science so as to try to understand how to further develop these ideas within their framework (and perhaps how to encourage the establishment to engage in these ideas directly). This process has been educational in a number of ways, and recently has begun to bear fruit. For instance, at the end of October, Maxwell and I attended the International Security Annual Conference (in Denver this year) where we presented our work in the second of two panels on terrorism. Although it may have been because we announced ourselves as computer scientists, stood up to speak, used slides and showed lots of colorful figures, the audience (mostly political scientists, with apparently some government folk present as well) was extremely receptive to our presentation (despite the expected questions about statistics, the use of randomness and various other technical points that were unfamiliar to them). This led to several interesting contacts and conversations after the session, and an invitation to the both of us to attend a workshop in Washington DC on predictive analysis for terrorism that will be attended by people from the entire alphabet soup of spook agencies. Also, thanks to the mention of our work in The Economist over the summer, we have similarly been contacted be a handful of political scientists who are doing rigorous quantitative work in a similar vein as ours. We're cautiously optimistic that this may all lead to some fruitful collaborations, and ultimately to communicating our ideas to the people to whom they will matter the most.
Despite the current popularity of the idea of interdisciplinary research (not to be confused with excitement about the topic itself, which would take the form of funding), if you are interested in pursuing a career in it, like many aspects of an academic career, there is little education about its pitfalls. The question of etiquette in academic research deserves much more attention in graduate school than it currently receives, as does its subtopic of interdisciplinary etiquette. Essentially, it is this last idea that lays at the heart of Eszter Hargittai's original complaint about physicists working on social networks: because science is a fundamentally social exercise, there are social consequences for not observing the accepted etiquette, and those consequences can be a little unpredictable when the etiquette is still being hammered out as in the case of interdisciplinary research. For our work on terrorism, our compromise strategy has worked so far, but I fully expect that, as we continue to work in the area, we will need to more fully adopt the mode and convention of our target audience in order to communicate effectively with them.