Author: ekfletch

I am an independent researcher on issues of gender, labor, violence, education, and children.

Striking a balance in data collection

A big part of my research time is spent on violence against women, gender-based violence, domestic violence, and harmful traditional practices. Though sometimes all whipped into a category of “women’s issues,” I’ve argued before that these are problems that everyone should care about, that they exert severe effects on our health and well-being as a society, emotionally, physically and economically.

Currently, I’m mired in two data collection projects, both with various degrees of hopelessness. I’ll write more later about my time in Caracas, but suffice it to say for now that there simply isn’t data available on issues like the ones I mention above. Or if it is available, no one’s going to give it to me. No surveys, no police data, no statistics on hotline use, nothing. We don’t know anything.

Conversely, in a meta-analysis of programs for adolescent girls that I’m writing with a colleague, my coauthor came upon a study suggesting that in order to correctly assess prevalence of Female Genital Mutilation (FGM) we should submit randomly selected female villagers in rural areas to physical exams.

I was shocked and disgusted when she sent me the study. I don’t doubt for a minute that the most accurate way to gauge prevalence of FGM is to randomly select women and examine them, but seriously? I am astounded that no one thought through the psychological consequences of women who have already been victims of gender-based violence being examined by a foreigner who thinks they are lying about whether they’ve been cut.

These days, it’s a good reminder for me that in collecting data there is such a thing as too much, and such a thing as not enough. It’s all about striking a balance.

Big data and what it means for economists

Over the past few days, a couple of pieces have come out about Big Data, or rather how economists and other social scientists are incorporating the extremely large datasets that are being collected on every one of us at every minute. Justin Wolfers, at the Big Think, says “whatever question you are interested in answering, the data to analyze it exists on someone’s hard drive, somewhere.” Expanding on Wolfers, Brett Keller speculates as to whether economists will “win” the quant race and “become more empirical.” Marc Bellemare thinks (in a piece that’s older, but still relevant) that the social sciences will start to converge in their methods, with more qualitative fields adopting mathematical formalism to take advantage of how much we know about people’s lives. Justin Wolfers and Betsey Stevenson go on in a related piece at Bloomberg about the boon that big data is for economics.

Not withstanding the significant hurdles to storing and using large datasets over time (ask a data librarian today about information that’s on floppy disks or best read by a Windows XP machine. Heck, look at your own files over the past ten years: can you get all the data you want from them? What would it take to get it all in a place and format you could formally analyze it?), I find the focus on data a little short sighted. And don’t get me wrong; I love data.

Wolfers and Stevenson think that the mere existence of data should change our models, that the purpose of theory nowadays should be “to make sense of the vast, sprawling and unstructured terabytes on our hard drives.” We do have the capability to leverage big data to gain a more accurate picture of the world in which we live, but there is also the very real possibility of getting bogged down in minutiae that comes from knowing every decision a person ever makes and extrapolating its effect on the rest of their lives. It’s the butterfly flaps its wings effect, for every bite of cereal you take, for every cross word your mother said to you, for every time you considered buying those purple suede shoes and stopped yourself–or didn’t. I’m being a bit melodramatic, of course, but it’s very easy, as an economist, as a graduate student, as a pre-tenure professor short on time, to let the data drive the questions you ask. It’s also often useful, I’m not saying that finding answerable questions using existing data is universally bad, by any means. But if we have tons of information on minutiae, we’ll probably ask tons of questions on minutiae, which I don’t think brings us any closer to understanding much of anything about human behavior.

On the convergence side, I worry about losing things like the ethnography. It may not be my strong point, but it’s useful, its methods and ouput informed my own work, and if convergence and big data mean anthropologists start relying solely on econometrics and statistics and formal mathematics, we’ll lose a lot of richness in our history and academics. I’m all for interdisciplinary work, for applying an economic lens to all facets of human interaction and decisions, but I don’t think our way of thinking should supplant another field’s. Rather, it should complement it.

Finally, incorporating big data into models that already exist will mediate some problems (unobserved heterogeneity that can now be observed, for example), but not all. Controlling linearly for now observable characteristics in a regression model has plenty of downsides, which I won’t enumerate, but can be found in any basic explanation of econometrics or simple linear regression.

Similarly, our tools for causal identification keep getting knocked down. At one time, regression discontinuity design was hot, and smacked down. Propensity score matching was genius and then, not so much. Instrumental variables still has this rather pesky problem that we can’t actually prove one of its key components. It’s not to say these tools don’t have value. When implemented correctly, they can indeed point us to novel and interesting insights about human behavior. And we certainly should continue to use the tools we have and find better ways to implement them, but the existence of big data shouldn’t mean we throw more data at these same models, which we know to be flawed, and hope that we can figure out the world. If we’re indeed moving towards more empirical economics (which is truthfully the part I practice and am most familiar with), we still need better tools. The models, the theory, the strategies for identification have to keep evolving.

Big data is part of the solution, but it can’t be the only solution.

Sabana Grande, renovado

The first time I lived in Caracas, I had an internship at a small business and finance magazine in a part of town known as Sábana Grande. It was not the nicest part of town. The pedestrian mall, which stretches from Plaza Venezuela to Chacaíto, was filled with buhoneros, or street vendors selling socks and batteries and burned CDs. And not just filled like If you were the one copyediting late, you weren’t allowed to be there by yourself, walking around at night was not allowed, under any circumstances. Since then, the pedestrian mall has been totally repaved and the buhoneros have been exiled to a large building named after liberatadora Manuelita Saenz (one of the few famous female figures from Latin American independence movements). It’s clean. And almost totally lacking in street vendors. It’s a supremely surreal experience, to walk up and down the mall. Music is still blaring, cheap shoes are still sold in half of the storefronts, and mannequins with impossible proportions (or rather possible with surgery) grace the windows. My enduring complaints about Caracas are being eroded. Well, at least the dirty part (we won’t get into the catcalls I endured today.) In fact, I’ve been impressed with quite a few areas that were once run down and dangerous and have been renovated. I spent the morning in areas called Altagracia and Capitolio, which has a new (not yet inaugurated) mausoleum for Simón Bolívar’s remains, a renovated Plaza Bolivar, repainted municipal buildings and more. I even saw some people scrubbing the bricks in Sábana Grande today and friends tell me that the nightlife in Capitolio is where it’s at. Unthinkable a decade ago. It seems that Caracas actually has changed in the last 10 years, though perhaps not so much in other ways. I’m here for another week, trying to dig up some data. I’ll let you all know what I’m up to after I get back.

20120807-205031.jpg

20120807-205143.jpg

Peer effects, health behaviors and adolescents

Some months ago I was at a conference, listening to a presentation on breastfeeding initiation and the presenter cited a paper by Fletcher. My first and second thoughts were, “how did that person get my breastfeeding paper?” and then “I didn’t say that in my paper.” Thanks to my trusty smartphone, I went searching for the paper, thinking perhaps my Gettysburg colleague, Jean Fletcher, had actually written it (a source of endless confusion for students, believe me), but found instead that it was Jason Fletcher, at Yale’s School of Public Health. Since then, I’ve run into a number of his papers and today, one came out in the NBER Working Paper series (gated), a paper on adolescent health behaviors and network effects with Stephen L. Ross.

The paper seeks to identify the effect that adolescents’ peers’ choices have on an individual’s health. If that sounds complicated, you’re not alone. Basically, the idea is that we want to know how strongly a child’s friend’s choices affect the child’s choices. The problem of how to causally identify this effect has plagued researchers for some time. In particular, the issue is that ideally, we would want to observe one student’s choices in different peer groups. But even if we can identify an exogenous change in peer groups (or in peer groups’ choices, but most likely through a change in peer group), the change in peer group is generally coupled with a dramatic change in environment as well. For instance, Fletcher and Ross cite one paper that shows that children who move from high-poverty areas to lower poverty areas experience better outcomes. Clearly, their peer group changes because the kids in one area have access to different activities, different stimuli, etc, but also the general environment changes. Mothers of these children report reduced stress, for example, which in and of itself has been shown to improve outcomes for children (or more precisely, children in high-stress living situations have worse outcomes–memory is failing me at the moment, I’ll update when I recall a relevant paper). So, when the environment changes and the peer group changes, it’s difficult to separate out the effects.

Using Add Health, which is a really cool survey instrument, by the way, the authors identify the effect by arguing that there is rather little variation in cohorts within a grade, but friend groups that look similar (on characteristics observable to the researcher)

At any rate, I think it’s a pretty neat identification strategy. It rests on some pretty strong assumptions, primarily that when groups cluster on observable characteristics, they’re unobservable characteristics are also similar, but dissimilar on the characteristics that influence health behaviors. This assumption is a bit problematic, I think, but I’m resolving it in my head by thinking of the insertion of one student with a particular tendency to smoke (his older sister does it, perhaps?) into a peer group in 9th grade, while a similarly made-up peer group in 10th grade doesn’t receive that idiosyncratic shock. Thus, the two groups look pretty similar, but by virtue of being in different grades, they have exposure to different kids and thus end up with different health behaviors.

Neat, no?

One concern I do have, though, is the idea that these friend groups are really that separate. I’m not very familiar with the way Add Health identifies friend groups, but I seem to recall some issues arising for researchers given a) the definition changing, and b) there being a limit on the number of friends that could be identified. From my own experience (clearly the most relevant), there was also a lot of grade mixing of friends in high school, even more so in dating. Sports, off periods, electives, and activities all gave way to friends in classes above and below. I grant that I went to a rather unique high school (billed as a sort of mini college campus), but it seems like it might be even more pronounced in a small schools. The assumptions of separation might be easier to make with middle schoolers, although incidence of averse health behaviors are going to be lower there and perhaps harder to identify.

Sources:

  1. Jason M. Fletcher and Stephen L. Ross. Estimating the Effects of Friendship Networks on Health Behaviors of Adolescents. NBER Working Paper 18253. July 2012.
  2. Kling, J.R., J.B. Liebman, & L. Katz. (2007). Experimental Analysis of Neighborhood Effects. Econometrica 75(1): 83-119.

I’m back

I’m back! I’m fighting the worse jet lag I’ve ever experienced in my life. Yesterday I was up at 1:30am and today at 2:30. I figure, this is what @price_laborecon must feel like. Nonetheless, I’m stateside for a few days and going to crank out some original research and blog posts.

Here’s one picture for you, in my new saree. More on Kolkata and India and Bengali weddings later, I’m sure.

Time use and hindsight

I am in the midst of revising a paper that uses a very specific question from the Fragile Families Data set about reading to children. When I began writing the paper, I started looking for evidence with time-use surveys, such as the American Time Use Suvey (ATUS) which asks participants to record everything they do and for how many minutes on two given days (a weekday and a weekend, usually). I noticed, particularly at the PAA meetings this Spring, that there was a lot of controversy about these surveys. What, exactly, can they tell us about general effects, when we are looking at such a small sample of time for any given individual? More specifically, if we want to examine the effects of a particular policy, how does looking at one individual’s day give us a causal effect of a policy? Time use surveys are incredibly useful for seeing exactly how individual spends his time on any given day, and the possibilities for understanding the dynamics of child-rearing and marriage are far-reaching. The trade-off is that you have no way of knowing whether this is a typical day or not. On average, for the population, if we have a random sample of individuals and days are sufficiently randomly assigned, we should get an idea of what the population does, on average. But asking if a particular impetus leads to a specific behavioral change (for instance, does an increase in income mean you invest more in child’s education) is a little more problematic. The alternative is to ask questions in a survey setting about time-use behaviors without specifying the time. That’s what the Fragile Families does, and the question about how many days per week you read with your child has its own problems. I have long argued that when individuals answer the question, they must do some averaging over time. The question is not “how many days did you read with your child last week” as might be preferred or indicated by the literature on work (did you work last week?), but rather a sort of what do you usually do? I’ve been surprised at how much pushback I’ve received on this matter from discussants and reviewers. Most say the natural model to use is a count model, like negative binomial or Poisson, but I think it makes more sense to use an ordered probit, which allows for 4 to be more than 2, but not necessarily twice as much as 2. I don’t think the reading days answer is as firmly countable and identifiable as something like parking tickets, where a count model is the readily apparent model. I imagine the question is a lot like exercise. Over the weekend, I helped a friend with her match.com profile and one of the questions is how many days a week do you exercise? For some, the answer is absolutely 7, every single day. For others, zero, not lifting a finger. For most, though, I’d guess it varies from week to week. One week, you go every day, the next week is busy at work, so you go less often. Perhaps you go on a whole-day hike and tell me two days instead of one because you don’t want to seem lazy. Thus, when I ask you the question of how many days a week you exercise, you’re not really giving me a straight answer, through no fault of your own. You’re averaging over the last couple of weeks, you’re perhaps adjusting your answer to reflect what you think the surveyor is looking for, and you’re partially giving an impression of how much you value exercise. I’m having a hard time making this same argument regarding time spent with children to discussants and reviewers, and I’m not sure what I’m missing in my explanation to make it more convincing.  

Fun stuff on infant health and taxes

I’m not a huge fan of the US tax code. I think it’s far too complicated and full of ridiculous things you can do to get around paying your taxes. This makes for rent seeking and a huge time suck. That said, I’m always interested in the types of incentives that certain taxes provide to change behavior, particularly when it comes to child health and investments in children. A new NBER working paper examines the relationship between infant health and the Earned Income Tax Credit.

Without having read the paper, my first thoughts are 1) why do we think the EITC would affect infant health specifically? and 2) those are pretty large effects for an increase in income. The authors argue that an exogenous increase in income means mothers will seek out more prenatal care, but it seems that a 10% reduction in the rate of low birth weight babies would require a large portion of the tax credit (increase in income) to go towards prenatal care for most mothers (or all for some and a small amount for others). Maybe they address it later on, but this, too, will make for some good plane and train reading this week.

The abstract is here:

This paper evaluates the health impact of a central piece in the U.S. safety net for families with children: the Earned Income Tax Credit. Using tax-reform induced variation in the federal EITC, we examine the impact of the credit on infant health outcomes. We find that increased EITC income reduces the incidence of low birth weight and increases mean birth weight. For single low education (<= 12 years) mothers, a policy-induced treatment on the treated increase of $1000 in EITC income is associated with 6.7 to 10.8% reduction in the low birth weight rate, with larger impacts for births to African American mothers. These impacts are evident with difference-in-difference models and event study analyses. Our results suggest that part of the mechanism for this improvement in birth outcomes is the result of more prenatal care and less negative health behaviors (smoking). We find little role for changes in health insurance. We contribute to the literature by establishing that an exogenous increase in income can improve health, and illustrating a health impact of a non-health program. More generally, we demonstrate the potential for positive external benefits of the social safety net.

Source: “Income, the Earned Income Tax Credit, and Infant Health.” Hilary W. Hoynes, Douglas L. Miller, David Simon. NBER Working Paper No. 18206, July 2012
.

Off to the land of sights and smells and senses, or a break

As you’re reading this, I’m likely on a plane, or sitting in one of many airports or train stations that is in my future over the next month and a half or so. Today, I’m headed to India to see my dear friend and colleague get married. It’s going to be a five-day, multicity affair, with an overnight train ride in the middle. En route, I’m stopping in Mumbai and Darjeeling to do some shopping (new saree!) and hiking and to see a bit more of this huge, incredible country. I was in India a few years ago and fell in love with it. It’s overwhelming, to be sure. The smells and the colors and the throngs of people are total madness, but I love it; it’s exhilarating to be somewhere out of my comfort zone. I’m not doing research this trip, though you can guarantee my eyes will be peeled for interesting things. I’m also not taking a computer, which means unless I can get wi-fi for my phone, I’m doing this trip old school. I’m going to read some books and journals, including Casualties of Credit by Wennerland and the CESifo journal issue on malnutrition, and some stuff for fun, like Just Kids and back issues of the New Yorker, but won’t be here or on twitter much. My 30th birthday present to myself is a real break from work, this means not thinking about the papers I have under review, or the one that’s due at the end of August, or the one I have to finish for the CNEH conference in Banff in October (so excited for Banff!). A break. I’m going to sit in a big, comfy chair on a tea plantation and stare at the Himalayas (or the clouds, given that it is monsoon season, but, details). If you want to read more about down time (or the lack thereof), take a few minutes for this piece in the NYT from last week on “busyness”, a phenomenon that I’ve been complaining about since my years at Duke, and suddenly everyone is talking about, or Bryce Covert’s piece in The Nation on work-family balance. If you know something I shouldn’t miss in Darjeeling, Kolkata or Ranchi, please do share. I’ll try to check email sporadically. I’ve also been designated honorary photographer and family blocker for this wedding by my advisors, fellow grad students, and professors in the Economics department at the University of Colorado, so I hope to have some crazy wedding pictures and experiences to share when I get back. Have fun! Talk to you all soon. Enjoy your July and thanks for reading. I forgot to acknowledge my blogiversary (sp?), but I’ve loved getting to know you all over the past year. Thanks for your comments and ideas and conversation and emails and shares and links. This has been an amazing learning experience for me and I’m so excited to keep it up over the next year (I promise not to torture you all too much with job market woes in the coming months. Feel free to chastise me if it gets out of hand.)

Long chain kidney donations

The NBER working paper series this week is giving me some plane reading for the next few days:

It has been previously shown that for sufficiently large pools of patient-donor pairs, (almost) effcient kidney exchange can be achieved by using at most 3-way cycles, i.e. by using cycles among no more than 3 patient-donor pairs. However, as kidney exchange has grown in practice, cycles among n > 3 pairs have proved useful, and long chains initiated by non-directed, altruistic donors have proven to be very eff ective. We explore why this is the case, both empirically and theoretically.
We provide an analytical model of exchange when there are many highly sensitized patients, and show that large cycles of exchange or long chains can significantly increase efficiency when the opportunities for exchange are sparse. As very large cycles of exchange cannot be used in practice, long non-simultaneous chains initiated by non-directed donors significantly increase efficiency in patient pools of the size and composition that presently exist. Most importantly, long chains benefit highly sensitized patients without harming low-sensitized patients.

I have a vague recollection of reading about kidney chains a few months ago, probably in the NYT, and getting a warm, fuzzy feeling inside about altruistic donors and how long chains helped to reach those who didn’t have an altruistic donor. I love it when economics comes together and puts everything in a neat little package for me. I’m excited to read this later.

Source: The Need for (long) Chains in Kidney Exchange (gated)
by Itai Ashlagi, David Gamarnik, Michael A. Rees, Alvin E. Roth NBER Working Paper 18202 (HC)

30

It’s kind of a big one, right? Today is the beginning of my next decade, and I’m happily in the mountains, riding my bicycle, watching moose, finding secret cabins, and spending time with friends. I hope everyone is having a safe and lovely week. Happy belated 4th and talk to you next week.