P/F a Fall Quant Course?

[Habermas]

I'm a current M.A. student applying to PhD programs this cycle. Things have fallen into place and I'm feeling quite good about my chances, COVID-cycle application odds aside. I am applying to about a dozen of the top schools.

The one issue is my current semester's quant course. Its my second intermediate grad-level stats course I'm taking to try and demonstrate a non-hostility toward quant methods despite going into a relatively qual sub-subfield of comparative politics. My strategy was to pair these two courses with a 162Q and some basic stats in my writing sample to show I was ready to be a quant-literate scholar who would go on to at least do simple multivariate regression work to pair with my qual focus.

Studying for the final, I'm not confident I'm going to be able to pull off the A or A- in this class, for a variety of reasons that won't affect my ability to do well in a full quant module during PhD. The rest of my graduate coursework has been stellar--including an A in another stats class. Am I better off declaring P/F than risking a B/B+?

Tldr; I have a strong application I feel good about except for my upcoming stats final. Should I P/F a course that I might get a B in?

[PoliSciGuy00]

I would not worry too much about this.  I'd say take the risk with the class.  I personally don't think anyone will hold a B/B+ against you, especially if it's a high level quant course and if you signal in your PS that you aren't applying to be a hardcore quant scholar.  Your 162Q is also pretty good, too.  Others may have a different view, but I'd say stick with keeping it graded.  

[Habermas]

1 hour ago, Mr_Spock2018 said:

I would not worry too much about this.  I'd say take the risk with the class.  I personally don't think anyone will hold a B/B+ against you, especially if it's a high level quant course and if you signal in your PS that you aren't applying to be a hardcore quant scholar.  Your 162Q is also pretty good, too.  Others may have a different view, but I'd say stick with keeping it graded.  

Thank you so much for this feedback. I will say its not a very high-level course (level 500), but point well taken. My SOP signals mixed-methods but certainly makes no promises by way of cutting edge quantitative models or anything of the sort--and I focus a lot on the qual skills I bring to the table. It also bears note that very few schools are actually going to request fall grades, so this may be moot if this cycle pans out for me.

[qxy]

I also wouldn't worry about too much if your stats score is B or B+. I believe many grad schools would expect the applicants to have some exposure to quant stuffs but not necessarily to excel in advanced statistics, unless your subfield of interest is political methodology. I would suggest to make sure about the mixed-method paradigm is highlighted.

In terms of writing sample, I am not sure whether it is smart to show basic stats if that is not your strong suit (Unless you have enough room to show your qual strength, or mixed-method way of thinking). As far as I know, in terms of empirical works, faculty tends to figure out how you arrive at the methods for your research question/hypothesis, instead of what method is being applied - what is your "data" (in a broad sense), how you justify the method selection, how you interpret the results etc.

[Clintarius]

I agree with the others for the course! For the writing sample, I think an important part is to show that you understand the method you choose to use.

For example, if you use regression analysis, it will better to do simple OLS, and then really discuss whether the result can be interpreted as causal (if that's what you're trying to do): what are potential selection biases, potential omitted variables, potential systematic measurement error, etc. rather than trying to apply a "fancier" method like IV regressions or the like.

[guest789]

On 12/6/2020 at 10:49 AM, Habermas said:

simple multivariate regression work to pair with my qual focus.

Quick note: I think you mean multiple regression. A multivariate regression, where you have multiple dependent vars, isn't that simple. Unless you do mean multivariate regression, which is cool, I just want to be sure you're not mixing up the terms before you submit your paper.

Edited December 11, 2020 by timeseries