Quantitative Prerequisites - Online classes

[Blowinginthewind]

I'd love to get some perspective on taking online classes (think Coursera) to make up for gaps in my quantitative education.  I graduated UG a while back and did not take any math or data science classes.  Given the importance of the quantitative aspect in admissions, how do you think an AdCom would view Coursera classes in Calculus, Stats, and Data Science (with R)?  My assumption is that at the very least it would serve some signaling purpose (i.e., I recognize and acknowledge the importance of math in the field) - is that correct?  Or is it a complete waste of time?  Thanks in advance for your insights!

[Mazarin]

I think this is an excellent question to which I do not really know the answer. My instinct tells me that it can be a potentially useful asset if you have no prior exposure to calculus or stats AND your intended field of study is not pure (non-formal) theory. However, in any case, I believe that overdoing this is not a good use of your time (which for most people is a scarce resource). I know December looks far away but actually it is not! 

I think most people would not, for example, give a coursera certificate even nearly as much weight as they would to a regular class with a letter grade. So maybe my take on this is: take a single online stats class with no calculus requirement to ease you into the framework or just do a programming course on python (coursera has excellent offerings for beginners). If you can do neither, don't beat yourself up. Try to impress with other aspects of your application (get good LoR's, write amazing SoP's, etc). The exception to this is if you are going to signal an interest in going quant-heavy, or if you finished undergrad very long ago (be extra careful with your LoR's if that is the case). Then here coursera could be your best friend, e.g. the Computational Social Science Specialisation.

This forum is a bit slow this time of the year and I am sorry for not being super helpful. I have friends in the same situation as you and I just happened on your question as I came to the forum by chance. I hope my upvote to you and this reply will prompt someone more qualified than me to come and give us an authoritative answer 

EDIT: stuff with R is cool too!

Edited May 31, 2021 by Mazarin
add information

[Blowinginthewind]

Thank you for this very helpful response. 

I plan on pursuing the IR/CP route, so quant skills will be required.  I agree with your view that Coursera or other MOOC courses will not carry weight similar to a traditional class with a letter grade.  My intent in pursuing these classes would be to signal to AdCom that I am cognizant of the math component in a PhD program and that I tried to bring myself up to speed.  I already know Python - wish it was used more heavily.  It actually is quite a fun language.  I thought about R as well, but given very limited time on my end, I want to maximize endeavors that would increase admissions chances.  

I graduate UG a decade ago, so LORs will definitely be a bit challenging.  Noted on your other points of advice.  

Thanks again for your post. 

 

[Theory007]

I largely agree with the post above but I am more pessimistic that non-college online math courses will be useful. Maybe you will be able to signal that you find the quantitative aspect of your education to be important, but you are unable to signal, I think, that you have competency in this stuff. Depending on where you apply this may not be strictly necessary; if I was you I would apply only to non-quantitative programs in the second half of the top-50. I do think that all programs have quantitative requirements nowadays, but you may be able to find POIs who are not as quantitatively motivated. And yes, you will have to work on all other aspects of your application and do well on the GREs. I think in your case having a high quantitative GRE score (and verbal as well) may be particularly important.

Hope that helps - good luck!

[Blowinginthewind]

Thanks for the response!  So I am not competitive for the top 25 programs (primarily as a result of not having a quant background)?  Is that the right interpretation from the above?  

[Theory007]

14 minutes ago, Blowinginthewind said:

Thanks for the response!  So I am not competitive for the top 25 programs (primarily as a result of not having a quant background)?  Is that the right interpretation from the above?  

There will be exceptions of course, but I am almost sure that no top-25 (or so) program accepts non-theory students without a substantial quantitative background. The competition is simply too stiff so except if you are truly outstanding in other areas of your application, such as near-perfect GRE scores, I predict you'll have a hard time in really competitive programs.

[Blowinginthewind]

Thanks for the insight.  Very helpful.  And in your opinion/experience, is there anything that can be done to tilt the scales?  Would classes at a community college/university be better?  Publications (though of a more qualitative nature)? 

[Theory007]

2 minutes ago, Blowinginthewind said:

Thanks for the insight.  Very helpful.  And in your opinion/experience, is there anything that can be done to tilt the scales?  Would classes at a community college/university be better?  Publications (though of a more qualitative nature)? 

No problem.

Why not spend your time doing really well on the GREs or yes, see if you can publish a paper (of any kind) somewhere? I think that would offset not having a math background. I do think that classes at a community college would be better on balance, but I am unsure how much it would actually help, i.e. I have no idea if it would be useful. My hunch is that improving other aspects of your application would be better. And I would also write a personal statement that emphasizes the qualitative nature of your academic interests (if this is accurate) and identify POIs who would be able to supervise projects of that kind.

Another suggestion is to go for an MA degree somewhere. There are not many really good options, but completing one would benefit you I think since you graduated from college quite long ago and due to the limited quantitative background, which is something an MA program would allow you to brush up on.

[Blowinginthewind]

The added debt/opportunity cost from an MA degree doesn't make sense to me, especially when the end goal is academia.  

I knew GRE scores were important, but didn't realize that they could impact decisions within this context - I reckon anything less than 165Q would hurt me.  Highlighting qualitative aspects of my research interests in the PS is very useful.  I hadn't thought of that. 

Thank you for all your help.  Very helpful, and much appreciated.