Jump to content

Monody

Members
  • Posts

    197
  • Joined

  • Last visited

  • Days Won

    1

Posts posted by Monody

  1. 1 hour ago, CarefreeWritingsontheWall said:

    So, I just finished writing an exam on probability theory so perhaps I'm a little salty and burnt out from a hectic semester. I think there are two issues here. I've previously taken a 3 course methods sequence, and one month of ICPSR's summer program, so I came into my PhD with more than an average knowledge of how statistical methods are applied in the field. I've still been clobbered by the math, and the requirement that, for example, we know how to derive the variance of a standard normal bivariate distribution by hand... I think there is something to knowing the mechanics and math operating underneath the concepts. At the same time, knowing how to derive something cold certainly won't save you from poor model building. My previous training was much more oriented around applied causal analysis - meaning, for e.g. the week we learned about synthetic control methods and matching, we talked about hypothetical research questions, how this method could resolve endogeneity issues, and how to do it right (and for the right reasons). This is where standards in the field have changed the most, even in the last 5-10 years. The focus now is not only on using statistics, but using them well. Most every methods course at the graduate level will require you to replicate a previous paper at some point to demonstrate the issues you suggest (I was required to do it previously, and will again in this program). Publications today don't ride on a simple replication - the focus is on both correcting poorly developed models, but also expanding on them. I think of the Andrew Rose and Goldstein Rivers and Tomz debate on the impact of the WTO on trade flows papers as a good example of this in IR. It's hard to believe the field was okay with shoddy models, but in essence there weren't necessarily a lot of people capable of policing how statistics were used (in terms of reviewers). Now there are. 

    Of course, we're also coming to realize a lot of important issues with reliance on quantitative methods, and in some ways this comes from our field being a little behind others where statistics is the primary means of producing evidence. Consider Ai, Norton (2003) on problems with interaction terms in logit and probit models or Montgomery, Nyhan and Torres (2016) on conditioning on post treatment variables. There's another paper out there with a fantastic look at how a handful of countries in a large-N panel completely drive results due to the use of fixed effects. Any program with a thorough training in methods will have you see and talk about these things. They also need not come up in a methods class per se (I heard about the latter three in our IR seminar). Relying on substantive courses to highlight deficiencies in quantitative methods is also not a strong bet - we lucked out with a prof who is very concerned with these issues, but in other courses it was only ever raised as a cursory problem swept aside in favour of criticizing the underlying theories in papers. A good program will reinforce all of these issues, as will the diligent student. I should add this is no different for people pursuing processing tracing and interview methods, or people who are employed as faculty. I always thought it was weird that my MA advisor still went to methods workshops, but I see why now. There's always more to learn. It's part of what makes our field so dynamic, and this equally applies to survey, interview and archival research methods given the changing nature of technology and archival processes. 

    At the end of the day, what's most important is being able to walk away from a program with a strong capacity to ask interesting relevant questions, to develop logical and conceptually clear theories, and the capacity to test those theories as rigorously as possible with a combination of tactics best suited to the issue at hand. This relies on more than a knowledge of math or statistics. It also requires a thorough understanding of what it's like to be in areas experiencing the phenomena we're interested in. Field work, or even interviews with people who have been involved, are really important. If there's one piece of advice I can lay out here it's to not lose sight of the real people underlying what we see to explain. 

    Very enlightening and heartening response. Thank you very much. Maybe just to add, I think that the progress is field and probably path-dependent. For example, I would argue that the progress on this front in my area (intrastate conflict research) is far less than in IPE for example, at least from what Ive read in the recent years. 

    But coming back to my original question, what did you learn in your undergraduate method courses and what would you say is the average methodological knowledge they expect an undergraduate to have? Also in which program are you if I may ask?

  2. 3 hours ago, changeisgood said:

    So far from what I've seen, there are a lot of people doing a lot of heavy duty math in our field, but the ones that do this kind of work often struggle to attach any meaning to what they are doing.  Math is nice, math is pretty, but if you're not contributing something to improve behavior outcomes, institutional operation, etc. or whatever your particular flavor is, it's just mental gymnastics for the sake of fiddling around.  I can't tell you how many methods articles I've read that end with something like "we really can't say much about the implications of all this, except to say that we need to use this method more often".

    I dont really mean the heavy duty math as you call it. It is more the kind of stuff like: using an estimator with 200 observations that is known to be inconsistent with fewer than 500, assuming no auto-correlation in pooled models, using fixed effects plus a lagged dependent variable without clustered standard errors, not considering selection effects, using control variables that are determined in the model, etc. Then I usually look at the replication file and attempt to correct for those things just to find out that the results change and don't support the paper's argument anymore. I find that severely annoying and I think that a more in-depth training could alleviate these issues.

  3. 4 minutes ago, resDQ said:

    I think everyone has their own personal goals. After grad school, some people have time to keep up with the latest methods and others do not (heavy teaching load + family). You still need to publish to get tenure. I think math is easy to learn if you put effort into it. The ability to come up with good research questions is not obtained as easily, however.

    Well, if you see it like this, of course you are right and I see that there are other strenuous commitments. On the other hand, I have problems seeing how the community's knowledge should advance if half of it is built on sand, no matter how ingenuous they research idea. That is not to say that there isnt great work out there, just that a lot of published work disappoints more than it has to.

  4. 28 minutes ago, Comparativist said:

    Political science undergraduate majors are highly disconnected from grad school. I reckon that programs don't want to alienate potential majors by making it methodologically strenuous. Virtually any poli sci major can get through by only having to take one vague 'empirical methods' course. You may have to take a quantitative requirement to get your degree as a university-wide requirement but these are rarely very specific.

    The only ones who have any training are those who actively sought it out.

    That being said, I really don't think you need a lot to really prepare yourself for graduate studies in poli sci...differential and integral calculus, intro to stats, and maybe linear algebra + discrete math would be ample to have a good foundation going in.

    I thought more about set and measure theory and matrix algebra. I am also currently working through Econometrics by Hayashi and Ive to say that I feel that I need this kind of training to understand what I should do to get the correct results without just proclaiming "I follow x (xxxx) by applying estimator y". Maybe that's just the area I am interested in, but there appears to be so much bad work (method wise) that is being published that I think that I would like to get all the training and knowledge I can get to avoid that kind of stuff simply for knowing that I am not going to be a part of the problem. 

  5. 7 minutes ago, resDQ said:

    I have a BA in statistics. I guess that answers your question. 

    Methods training is important for me and I've looked into it extensively when I was selecting schools. A lot of your experience in grad school (from my understanding) will depend on who is teaching the methods course and how much effort you choose to put into it. If I get into schools that don't have great training, I'm going to ask during visitation days whether it would be possible to do an MA in Stats or Econ.

    Wow, I am jealous. I realized a while ago how great it would have been to have the really substantive knowledge a BA in Statistics would have been able to provide. I am currently somewhat frustrated as it feels like the more I learn the more I am missing the underlying Math to accurately relate everything to each other. Now I am working through Mathematics for Economists by Blume, but Ive to juggle all the other projects at the same time so it will probably take the next few weeks to get through it.

    Maybe someone also can share his or her experiences with undergraduate method courses for PoliSci majors?

  6. 13 minutes ago, resDQ said:

    Yes. Fit isn't the best, but I like the methods training they offer. 

    I see. Ive applied based on both, substantive research areas and methodological training. Out of interest and because you mentioned it, what kind of method training have you received so far as an undergraduate in the US and how much does it relate to the underlying Math? Formally, I only did OLS, some MLE, and factor analysis and am now basically abusing the Econ department (causal inference, time series, VAR models, panel methods, SEM), the latter of which made me realize how comparatively shoddy the PoliSci method training was with regard to the mathematical processes.

  7. 7 minutes ago, resDQ said:

    It is a good program, but I can't imagine them having the same number of applicants as say Harvard or Stanford. We also don't know how they handle admissions (several meetings throughout a week, one long meeting, etc.). 

    I didn't want to imply that it wasn't a good program as I surely would not have applied otherwise, but I find that impressive compared to the length of the process for the other programs. I mean 2 1/2 months for Stanford compared to roughly 3 weeks. Are they receiving such fewer applications?

  8. 2 minutes ago, resDQ said:

    completely normal. Should be hearing from Penn State soon, I think. 

    Thanks. Penn's results are expected at the end of January, aren't they? Ive to say that this is interesting considering that their deadline was only recently and much later than those of universities who publicize their results much later. 

  9. 6 minutes ago, Quickmick said:

    Try and factor in everything you don't know is out there!

    Not foolish/suicidal enough... It's particularly damning as my reference manager now has over 1500 annotated papers and this feeling still persists.

    On a positive note, I noticed yesterday that Princeton seems to post some results at the end of January already.

  10. 4 minutes ago, philpot said:

    Not getting more stupid, just realizing how much there is you don't know. It's a great exercise in humility :)

    Well more stupid in terms of knowledge as a percentage of what you know is out there. That percentage is precipitously shrinking. 

    I also have to say that there are days on which I am unsure about what I should work on. Should I attempt to find every possible literature that is related to my field to be aware of it. Should I put more work into methodology? Should I hone what I know first before venturing into other areas (particularly Economics, Psychology, Criminology, etc)? Should I just focus on current working papers or is this a dishonest thing to do if I am not 100% sure about every possible connection out there?

  11. Anyone else feels like s/he is getting more stupid by reading more? I am starting to feel like I should take courses in basically every social science research area and beyond to get a comprehensive understanding of the substantive issues and methodological opportunities to create the best possible research. When I am reading Wooldridge (2010) I feel like I should really take courses in Mathematics and in extension in Microeconomics. Alternatively, there is such a rich literature in Psychology or Geography that I want to tap into but it all takes so much time to really understand the complexity and avoid creating shallow research that is not sufficiently grounded and aware of all necessary implications and assumption that undergird the methodology, just to take one example. I even caught up on links to evolutionary biology and physics...

  12. 4 hours ago, dagnabbit said:

    Hope everybody's holidays were excellent! It's very exciting and nerve-wracking to think that, if this cycle is similar to prior cycles, we'll have decisions from at least a few schools before the month is over.

    To pass the time, what is everyone interested in studying? My subfield is IR, with specific interests in international organizations and foreign aid allocation.

    I am only waiting for the 11th of February. None of my schools seems to send out results earlier than that based on previous results. I am also preparing alternative MA applications just in case.

    Regarding your question, I am situated between IR and CP and am interested in micro-processes in civil wars. I recently thought about beginning to work on a dataset, but the librarian at my university seemed to be quite annoyed when I asked her whether it would be fine to download several thousand articles from Factiva. Licensing agreements are quite a hassle. :/

  13. Judging from your profile, you are already too late. The application season for international/non-EU student was the 15th of December. I applied in case that nobody in the US takes me (also for the scholarship) and I found the information they provide on their website quite extensive actually. -->(http://www.cis.ethz.ch/education/MACISnew.html). If you have any questions on the application procedure for the next season I may be able to provide you with some answers.

  14. On 12/25/2016 at 4:02 AM, GradNYC said:

    Has anyone submitted their application to Princeton before the deadline then UPDATED their personal statement before the deadline but notices that the original personal statement is recorded still..... 

     

    Thanks!!!

    The original statement is not recorded still. It only states the date of the original submission but it said on the top of the checklist before the deadline that they keep the date of the first submission but change the file. It is different from Columbia's process where you explicitly upload an updated version.

×
×
  • Create New...

Important Information

This website uses cookies to ensure you get the best experience on our website. See our Privacy Policy and Terms of Use