Do I need a stats course to apply?

[Banxe]

Hi all,

I am entering my senior year in an interdisciplinary program, but I am thinking about doing poli sci for grad school. I have not taken any stats courses since high school and am wondering how important in might be for grad school.

I tried taking a basic stats course last year, but after looking at the sylabus, I realized I had learned it all in my IB stats course in high school (the professor also told me not to bother taking his course). I can take a special stats course for social science, which would be useful, but since I am thinking of taking time off between now and grad school, I would rather just take stats in grad school if need be.

What would your advice be on this?

Thanks!

[natofone]

You'll likely have to redo whatever you take now in graduate school. I suggest that you take an applied course on linear regression. It'll help in graduate school because you'll more fully understand what things like regression coefficients, t-scores, and p-values are for your reading seminars. It might improve your chances to get in as well, because I believe that some exposure to statistics is expected.

[polisciphd]

It almost entirely depends on the methods focus of the school.

[StudentForever]

Hi all,

I am entering my senior year in an interdisciplinary program, but I am thinking about doing poli sci for grad school. I have not taken any stats courses since high school and am wondering how important in might be for grad school.

I tried taking a basic stats course last year, but after looking at the sylabus, I realized I had learned it all in my IB stats course in high school (the professor also told me not to bother taking his course). I can take a special stats course for social science, which would be useful, but since I am thinking of taking time off between now and grad school, I would rather just take stats in grad school if need be.

What would your advice be on this?

Thanks!

From my personal experience, I would make sure you get a solid idea of the methods focus of the program from the GRADUATE students, not from the professors. I knew nothing about methods coming into my program, was told I would get training from the department, and was then left to fend for myself. I don't necessarily blame the department (though I wish the expectations of the faculty aligned with what I had been told). Rather, I wish I had taken it upon myself to find out about the methods focus and expectations of the program.

Ask the students where/what the focus is, then decide based off your previous experience whether or not you need more methods training before you get to a program.

[brouhaha]

On a similar note, I took Calc 1 and 2 from the math department (meaning not applied or anything) and got Cs in both, just because I'd always wanted to take calc and thought it would be helpful. It certainly was and I learned, but do you think this will hurt me?

[ipsqq]

Banxe, don't worry about it now. If you take it in grad school, it will count toward one of your minors and there will be a lot of people in the same boat as you.

brouhaha, I do think that might hurt you. Those are very low grades and reflect poorly on your ability to do quant work at the grad level. Unless you are going to an explicitly qual program or into theory, that doesn't look good.

[thegradstudent]

I'd agree and disagree with what ipsqq wrote above:

I think it'd really help your application if you took either a stats or an applied social stats course and it'd also help you once you get into grad school. Those that had prior exposure to calculus and stats had a much easier time adjusting and doing well in methods courses once in grad school. I took Calc 1 and 2 (from the math department) and stats (from the econ department) and it was incredibly helpful. Like one of the other commenters wrote above, this will definitely depend on what type of program you decide to enroll in. But I would say that virtually all of the top 15-20 (and maybe a bit farther down the rankings) programs are now quant oriented and make you take at least 2-3 methods courses (intro to stats, linear regression and then at least one of the following - game theory, generalized linear models, time series regression, bayesian analysis or increasingly popular is networks) as well as a generic research methodology/philosophy of social science course. They definitely value qualitative work, but they want you to be proficient in both qual and quant methods and everyone is required to take the initial methods sequence.

Where I agree with ipsqq is in the advice to brouhaha. Those C's may very well come back to bite you in the rear. They will look positively upon you for attempting to take those classes while an undergrad, but the scores in those courses will be looked at unfavorably. But, I will qualify this. If you have a sufficiently high GRE Quant score, those two grades will be dramatically less important. So if you have a quant score of let's say (and this is arbitrary) over 730, I don't think it'd hurt your app much.

[thegradstudent]

I agree. Would you say that it could work the other way as well? Would my A in a quant methods course in undergrad do anything to offset my 680Q?

Good question. Thanks for posting an example with hard numbers. A 680 is in the 66th percentile; a decent score. The A will show the adcom that you have both attempted a quant course and excelled in that course. It shows a previous track record of performance which is good. But I'd argue that the effect is not as dramatic as the effect of a top quant score to offset a lower grade in a course. They will still see the two C's and that will hurt your app, but some of it will be offset by the high quant score (but not all of it!). Your example shows ability but I don't see it making that much of a difference or as much of a difference as the higher quant score. Remember that applicants with A's and a higher quant score will have a bit of a leg up on both of you. This analysis will only benefit you at the margin when your app is stacked up with similar ones. My argument and analysis is based on cetaris paribus.

But as I've said in previous posts, let's not worry too much about the differences between a 680 and a 720. It's just not that important in the grand scheme of things.

[hawk]

Well the C's are a reflection on only your course, university, professor, specific circumstances, etc. The Quant score is a reflection on your performance and abilities relative to the rest of the nation (or at least those who take the GRE). Thus, a low grade would be more easily offset by a high GRE Quant than vice versa, and neither one nor the other nor both being high would guarantee you admission because the application is looked at as a whole and all factors are compared to all other applicants in the pool for any given year. To answer the OP's question, I would think statistics would be helpful if you think you can do well AND you have time for it. If you are not going to get a B+ or higher (minimum), you are just going to hurt yourself and you would be spending your time more efficiently by studying for the GRE Quant and performing in the mid 700's on that which would definitely be more attractive to an adcom than a B or lower in an undergrad stats course. I would think a 700 and an A would set you up pretty nicely, but remember that statistics is not all-important and that there are many other factors when considering a polisci applicant.