Social Science Exists

A student with a mathematical-physics background could easily convince himself that he has superior mathematics abilities than typical economists and superior statistical and computational skills than most economists. He might go on to conclude that, as a consequence of his superior mathematical and computational abilities, he should be able to enter economics and start contributing quickly and easily. He might also anticipate that he could easily adapt established models or techniques in physics to study economic phenomena and impress the profession.

If you are one of these people, let me try to disabuse you of these notions. Your mathematical abilities are actually not that much better than most economists (if they are better at all). You will have to spend a lot of time acclimating to the subject and the path to actually making contributions will be long and difficult. In all likelihood, there are very few (perhaps zero) off-the-shelf models or techniques in physics (or engineering, or chemistry, …) that will produce meaningful economic results. (more; HT Justin Wolfers)

Yup! I often meet scientist types who talk about some problem they are working on, which turns out to be a social problem related to ones that social scientists have explored. But they won’t believe this unless you show them work that uses methods and concepts with which they are familiar. They just can’t believe there are useful methods and concepts that they don’t already know.

GD Star Rating
loading...
Tagged as:
Trackback URL:
  • STINKY
  • Robert Koslover

    A related observation: Nearly all homework problems, examples, and test problems used in college physics and mathematics courses have clean and complete analytic solutions that can be found in less (usually much less) than one hour, if the student properly applies the techniques presented in the book(s) and/or lecture(s). As a result, many very-bright but young/inexperienced newly-employed scientists and engineers are downright shocked when their attempts to apply these same seemingly-reliable techniques to real-world problems fail to yield the complete, quick, and/or elegant solutions to which they are accustomed.

    • IMASBA

      That’s not true of master’s courses and already in bachelor courses some approximation models (such as perturbation theory in QM and lots of stuff in fluid dynamics) are treated. Courses that deal with numerical solutions (such as Monte Carlo simulations) are in principal optional (though quite often required for certain masters) but a physicist interested in things like economics will probably choose to do them.

    • froginthewell

      Bad analogy, you are comparing science homework with social science research. I think most science graduates students know that there are no clean and elegant solutions to social problems. Though they may perhaps be underestimating (as opposed to discounting) the actual amount of ‘intuitive’ aka dirty reasoning needed.

  • http://web.mac.com/deweaver Dallas Weaver

    Not sure it is a true statement. Scientists and engineers dealing with truly complex problem (life and ecologies, climate, etc.) with N interacting non-linear variables may have some very good approaches.

    For example, economists don’t seem to have gotten past the question of building truly complex systems don’t necessary require any central decision making or intelligence. A termite colony does it every day. Evolutionary biology can say more about how the design structure of social systems can evolve parasitic sub-group behavior than some of the things I see coming out of the social sciences.

    After decades and billions of dollars of social science research, we can’t solve the problems of South Central LA or nation build in Iraq. We can’t improve the performance of our schools.

    Meanwhile, in the real sciences, we are on the boundary of actually understanding universe from the big bang to present time and how life itself (real complexity) works.

    • alexander stanislaw

      To be fair, “real” scientists have the luxury of being able to do experiments. Social scientists don’t get that luxury. A social scientists working in education doesn’t get to build ten schools, randomly assign people to them and then evaluate which methods work (and even if they could do that, the experiments would occur on the timeframe of decades).

      Physicists get to brute force test theories by gathering huge swaths of experimental data. Biologists are not quite this lucky – their experiments take months, sometimes years but they at least get to test their ideas.

      • http://web.mac.com/deweaver Dallas Weaver

        In the world of applying science to real problem in biology and ecology you only know or understand a small fraction of the relevant variables and have the luxury of experiments with even a smaller subset of those. Everything from a shrimp ponds ecology (determines the performance) on in the biological world, like a school, can’t be fully specified or easily and reproducibly experimented with.

        Shrimp farmers are probably more prone to try something else when what they are doing doesn’t work. If he doesn’t adapt, he goes bankrupt. In education, we know that the bottom 10% of teachers do more damage than good, yet we keep them on the staff.

        What good ideas and understanding that have evolved in social sciences seem to be ignored when the understanding doesn’t give the desired results. For example, public choice theory provides a lot of understanding that appears to be ignored when designing governmental system creating and expanding bureaucracy.

      • Douglas Knight

        Actually, in the 70s, they did randomly assign students to 10 different teaching methodologies. But then they realized that experiments have winners and losers and stopped doing them.

  • Scott H.

    This is coming from a profession that can endlessly chase its own tail over the question of whether deficit spending leads to inflation?

    Just an example observation: The physics and engineering mindset of “totaling of the applied forces to determine acceleration of an object” seems to be almost totally absent in economics discussions. Everything devolves — for decades — into simplistic arguments between “x causes y” and “no, x does not cause y” — ad infinitum.

    Or am I wrong?

    • http://juridicalcoherence.blogspot.com/ Stephen Diamond

      Or am I wrong?

      I’m not sure you’re entirely wrong, but the state of discussion isn’t quite so primitive. Everyone recognizes, at least in principle, that both the supply and demand sides must be considered.

      Yet, economists do seem subject to ungrounded consistencies of preference for demand-side versus supply-side explanations and forecasts.

      • Scott H.

        Yes, but even in demand side and supply side explanations there are sub-forces that may or may not be present or may have different strengths at different times. The mathematical rigor and mindset of the physicist and engineer make the discovery and cataloging of these forces second nature. Economists, with their trade tied much more to the vagaries of language and narrative, get bogged down with definition of terms, zingers, and one-upsmanship — hell, good vs. evil! (Btw this is a macro-only accusation.)

  • Michael Edward Vassar

    If a person doubts that social science exists, or alternatively, that post-New-Deal social science exists, how would you convince them?

  • Daublin

    Scott: I agree about macro, but economics seems to do much better with price theory. The researchers have converged, and they frequently make accurate predictions.

  • Anonymous

    Here is Mark Buchanan’s reply: https://medium.com/the-physics-of-finance/b91a9e6ee184 .

    I agree with that. Physicists don’t think that social sciences are easy and can be conquered by simply applying mathematical methods that every theoretical physicist knows. From the physicists viewpoint, it seems that social scientists themselves don’t appreciate how difficult their field is.

  • IMASBA

    Many economists are very proficient at math, no doubt about that. I do get the feeling though that they’re often like engineers: applying super complicated math without looking at the big picture first (except that the engineers are usually right anyway). It also takes forever before ideas are updated (it seems economists themselves act as if sociology and psychology don’t exist) and even then sizeable minorities cling to the old ideas (I mean seriously, how can economists in 2014 still maintain that large monetary bonuses work, or that executives are really worth millions?) Economics has all the tools to be a real science but it needs a mentality change to actually become one.

  • lambdaphage

    From a signaling perspective, what should we make of the fact that social scientists protest loudly about physicists poaching in their fields, but we never hear reciprocal complaints from physicists about social scientists?

    • IMASBA

      When has an economist (let alone a psychologist or sociologist) ever gone into physics?

      • lambdaphage

        Exactly. If it is true that:

        [one] often meet[s] scientist types who talk about some problem they are working on, which turns out to be a social problem related to ones that social scientists have explored

        then one should naturally expect a flux of social scientists into physics to poach the problems that physicists find difficult but sociologists et al. find familiar due to their different palette of methods and concepts. The fact that this never happens would seem to be a difficulty for the view quoted above, unless I’m misreading it.

      • Clinton McMurray

        It doesn’t follow. Why exactly should “..one naturally expect a flux of social scientists into physics to poach…”? This might be correct but not for any reason pointed out above. Perhaps social problems are simply more interesting to people who are capable of grasping at solutions to complex systems. We live in a physical universe but relate to other minds in a social universe. The consumption of the physical world is a social phenomenon. In a mind inhabited artificial world (in which agents were aware it was artificial, at least to begin with) would the agents care more about the physics or the social conventions and boundaries?

        All should welcome more brains to economics. Many many more would be nice. The way I see it currently is that lots of general science smarts does not necessarily transfer as easily to social problems as some very smart physical scientists would like to think. I have seen this first hand. Of course, those scientists can ultimately make contributions, but not as easily as one might think, which is the point of the article.

      • IMASBA

        “Perhaps social problems are simply more interesting to people who are capable of grasping at solutions to complex systems.”

        Probably, but that would make physical scientists a rare subset with a broader range of interests than usual, it would be wise not to dismiss them when they have an idea.

        If showing that a problem cannot be solved, or only after future breakthroughs in psychology/neuroscience, or showing showing that some incredibly complicated theory performs worse than a model that assumes randomness, or showing that some questions are silly to begin with count as contributions (they should), then physical scientists (including mathematicians) have contributed a lot more to economics than economists would like to admit.

      • Clinton McMurray

        “..then physical scientists (including mathematicians) have contributed a lot more to economics than economists would like to admit.”

        Do you think economists really downplay those contributions in general? It seems to me those contributions are made and absorbed and science moves on, but that they’re acknowledged when acknowledging is the aim and for the reasons you listed.

      • lambdaphage

        If there are secret methods and concepts in the social sciences that could solve open problems in physics, but which no social scientist has ever disclosed for lack of interest, and no physicist has ever caught wind of despite their notoriety for adopting whatever techniques will yield a halfway-reasonable answer, I suppose we have to leave it there unless you’re willing to offer an example. I searched in good faith and didn’t come up with anything, and I suspect I’m just being trolled.

        As to your suggestion that social scientists don’t bother with physics because of their confidence that they live inside a simulation, have you met any social scientists besides Robin Hanson?

      • Clinton McMurray

        I didn’t say there were secret methods and concepts in the social sciences that could solve open problems in physics. There are reasons that the set of tools and methods used by economists are different from the set of tools and methods used by physicists.

        I also did not say anything about social scientists having confidence that they live in a simulation. Or anything like it.

        You were not being trolled. This would be a troll: http://xkcd.com/793/

      • lambdaphage

        I suppose we don’t really disagree then. My original comment was motivated RH’s claim that natural scientists often work on problems that turn out to be social problems related to ones studied by social scientists, which still wants for an example.

  • Meteorologist at large

    Points well taken, but comparatively to meteorologists, Nate Silver is correct – meteorology has made enormous progress the past 50 years and one should save their jokes not for “the weatherman” but for the overconfident economists who speak authoritatively about their predictions! Indeed “the weatherman is not a moron”!

  • Philip Goetz

    I believe that I have better mathematical, statistical, & computational skills than biologists, and that I could quickly improve many solutions in biology. In fact, I believe I’ve already done this repeatedly.

    (But the NIH does not give funding to bioinformaticians, not even for bioinformatics projects. If you find a bioinformatics project the NIH has funded, it is more likely being run by an MD than by someone with mathematical background.)

    Do you think I’m wrong?

    • http://juridicalcoherence.blogspot.com/ Stephen Diamond

      In fact, I believe I’ve already done this repeatedly

      Can you link to an example?