low CS grade for Phd stats application

[causalinf]

I went to a top 10 undergrad with a major in statistics and minor in CS. My overall GPA for statistics is 3.82 but my CS grade wasn't ideal (a couple B range grade). I know the admission process is holistic, but I am worried my low CS grade will negatively impact my application. What should I do to make my application stronger? 

Edited November 15, 2018 by causalinf

[Stat Assistant Professor]

There's not really anything you can do about the B grades at this point, so I would focus on the things that you can control (namely, making sure your recommendation letters are *very* strong). I think the B grades in CS will be mitigated by the fact that you attended a top 10 undergrad. I assume your overall GPA is still pretty good too, even with the B's in the CS classes?

[causalinf]

27 minutes ago, Stat PhD Now Postdoc said:

There's not really anything you can do about the B grades at this point, so I would focus on the things that you can control (namely, making sure your recommendation letters are *very* strong). I think the B grades in CS will be mitigated by the fact that you attended a top 10 undergrad. I assume your overall GPA is still pretty good too, even with the B's in the CS classes?

Yeah. My overall GPA is around 3.79, which is pretty close to my major GPA. I am wondering how important CS grades count for PhD admission. My grades in core math classes are all A. I actually took some grad level CS classes (machine learning), and I got all A in them as well.  The CS classes where I received B grades are computer architecture and database systems. I didn't do well in them since I was never interested at the first place but had to take it for minor requirements.

[bayessays]

Don't worry about it at all. It won't be an issue. 

[fluffyj]

I got a B- in a data analysis class, a B+ in algorithms, and a B in real analysis II all of which are much more damning than a B in computer architecture, but still got into a top 5 program and waitlisted at another. My major/overall GPA was very similar to yours. I wouldn't worry about it. It's more important to have an upward trend in grades and have other factors like research or letters of rec in your favor.

[Stat Assistant Professor]

Based on your elaboration, I would echo the two posters above and say not to worry about this. Classes on computer architecture do not seem relevant to Statistics. People who have a few B's on their transcripts still get admitted to top PhD programs in Statistics, and adcoms are more forgiving about this if you went to a top school and your GPA in math/stat classes is good. If I recall correctly, one poster on this board was admitted to Berkeley with a 3.6ish GPA from University of Chicago (but a 3.8ish GPA in math classes). You should be in good shape to get into a good PhD program, assuming the rest of your application is equally strong.

Edited November 15, 2018 by Stat PhD Now Postdoc

[causalinf]

Thank you all for the information! I am wondering since computer architecture isn't really related to math/stats, do I still need to report them when self-reporting math/stats/computing class? Thanks!

[bayessays]

54 minutes ago, causalinf said:

Thank you all for the information! I am wondering since computer architecture isn't really related to math/stats, do I still need to report them when self-reporting math/stats/computing class? Thanks!

Nah.

[Radon-Nikodym]

On 11/15/2018 at 2:38 PM, Stat PhD Now Postdoc said:

People who have a few B's on their transcripts still get admitted to top PhD programs in Statistics, and adcoms are more forgiving about this if you went to a top school and your GPA in math/stat classes is good. If I recall correctly, one poster on this board was admitted to Berkeley with a 3.6ish GPA from University of Chicago (but a 3.8ish GPA in math classes). You should be in good shape to get into a good PhD program, assuming the rest of your application is equally strong.

I suspect you're talking about me with this comment, and for the sake of correcting misconceptions, I want to point out that my math GPA was in fact lower than my overall GPA of 3.6ish. However, I did receive a 900 on the GRE Math subject test, so that probably played a large role in alleviating any concerns that adcoms may have had about my mathematical abilities. Being a domestic student also probably contributed a lot to my acceptance.

[Stat Assistant Professor]

4 hours ago, Radon-Nikodym said:

I suspect you're talking about me with this comment, and for the sake of correcting misconceptions, I want to point out that my math GPA was in fact lower than my overall GPA of 3.6ish. However, I did receive a 900 on the GRE Math subject test, so that probably played a large role in alleviating any concerns that adcoms may have had about my mathematical abilities. Being a domestic student also probably contributed a lot to my acceptance.

Thanks for the clarification. University of Chicago is well-known for grade deflation and rigor, so even something in the range of 3.5 from UChicago looks good (if it were below a 3.0, that would be a different story though). The same applies to schools like MIT and Caltech. Adcoms know these schools are very tough but also much more intellectually rigorous than average. I have no doubt that you were well prepared for your PhD program if you studied math at Uchicago and had around a 3.5. So no need to deflate your accomplishment!

Edited January 31, 2019 by Stat PhD Now Postdoc