Columbia Statistics Masters for PhD

[NumStats]

So, I got rejected by their PhD, but they offer me the Masters Program. I am currently considering it because I haven't got any PhD offers so far and I think I can finish the master in a year. However, my end goal is still to get my PhD afterwards. I have been looking at their Masters Program and it seems to have not so good review, because they accept so many students at once and the some of the courses are taught by adjunct faculties. Will it be a good idea to go there if I don't get any PhD offers?

[GoPackGo89]

Is Columbia's masters intended to be a terminal degree? I only know that they churn out a ridiculous amount of masters students and it is often referred to as a cash cow. If you plan to finish in one year you only have from now until December to improve your application. In my personal opinion attending Columbia's masters program seems like an expensive/not so helpful bridge for re-applying for fall 2019.

[NumStats]

4 hours ago, GoPackGo89 said:

Is Columbia's masters intended to be a terminal degree? I only know that they churn out a ridiculous amount of masters students and it is often referred to as a cash cow. If you plan to finish in one year you only have from now until December to improve your application. In my personal opinion attending Columbia's masters program seems like an expensive/not so helpful bridge for re-applying for fall 2019.

Thank you for your opinion. I guess I will wait for my other results for now. 

[Bayesian1701]

I agree.  A Columbia Masters isn’t going to improve your profile a enough for reapplying in a year.  You would be better off saving the money and applying again.  You may not even have grades before apps are due (the ones in December).  Your application isn’t going to change much and you probably won’t get much credit towards a PhD.  

[NumStats]

12 hours ago, Bayesian1701 said:

I agree.  A Columbia Masters isn’t going to improve your profile a enough for reapplying in a year.  You would be better off saving the money and applying again.  You may not even have grades before apps are due (the ones in December).  Your application isn’t going to change much and you probably won’t get much credit towards a PhD.  

Thank you for your opinion. It was one of my concern as well. Probably I should take GRE subject instead.

[bayessays]

12 minutes ago, NumStats said:

Thank you for your opinion. It was one of my concern as well. Probably I should take GRE subject instead.

Most stats programs don't look at this, so that sounds like a bad use of your time. Where did you apply? Why not just take a year off from school and reapply to some more realistic options next year? There are like 40 really great programs and if you didn't get into Columbia this year you're honestly almost certainly not going to get in next year. 

[NumStats]

On 3/20/2018 at 2:05 PM, bayessays said:

Most stats programs don't look at this, so that sounds like a bad use of your time. Where did you apply? Why not just take a year off from school and reapply to some more realistic options next year? There are like 40 really great programs and if you didn't get into Columbia this year you're honestly almost certainly not going to get in next year. 

I applied to 13 program this year. So far have 8 rejections:

CMU, Michigan, UCLA, Columbia, PennState, TAMU, UIUC, NYU

Waiting for 5 left:

UCI, UCD, Notre Dame, Rice, and WUSTL

 

I thought I already spread out my applications to different tiers of schools, but no luck yet until now. Is GRE subject really not that important? I heard it would be helpful if you score well. Also, my GRE verbal is really bad (144), but my Quant is almost perfect (169). Do you think I should retake it? Thank you for your advice anyways

[bayessays]

4 hours ago, NumStats said:

I applied to 13 program this year. So far have 8 rejections:

CMU, Michigan, UCLA, Columbia, PennState, TAMU, UIUC, NYU

Waiting for 5 left:

UCI, UCD, Notre Dame, Rice, and WUSTL

 

I thought I already spread out my applications to different tiers of schools, but no luck yet until now. Is GRE subject really not that important? I heard it would be helpful if you score well. Also, my GRE verbal is really bad (144), but my Quant is almost perfect (169). Do you think I should retake it? Thank you for your advice anyways

Improving your verbal would definitely help more. Only a few programs require the GRE subject, and outside of those, it would probably only really help if you had bad grades in math classes. Almost all of your programs are top 20, or in especially competitive small programs because of size and location, or within math department (wustl, ND). Obviously I don't know your entire profile, but this is a pretty high aiming for all but the most elite international student. 

[NumStats]

3 hours ago, bayessays said:

Improving your verbal would definitely help more. Only a few programs require the GRE subject, and outside of those, it would probably only really help if you had bad grades in math classes. Almost all of your programs are top 20, or in especially competitive small programs because of size and location, or within math department (wustl, ND). Obviously I don't know your entire profile, but this is a pretty high aiming for all but the most elite international student. 

I would reconsider re-taking the GRE general then rather than taking the subject test. I have 3 majors (Pure Math - Honors, Engineering, Q Econ - Honors), 2 minors (Stats & CS). My Overall GPA is not that high comparably to other applicants (3.57), but my Stats, Math, and CS classes are mostly As with a few Bs. So I thought it would cover the low GPA, but apparently it's not. I am mentioning CS here because one of my main interest is Machine Learning, so I thought having a good CS grades would help, but yeah maybe I'm aiming to high this year. Thank you for the recommendation! It really helps

[bayessays]

Yes, to be blunt, you are aiming far too high with that GPA and GRE score. 

[statfan]

The admission process is very complicated and no single factor plays a decisive role. GPA is only meaningful given certain context. A lower gpa may be more impressive at a prestigious instituition than a perfect gpa from an unknown institution. People from lower-ranked instituition will have a much tougher battle. Also, your math/stat gpa matters much more than the overall gpa. What is your math background? How well did you do in each course? For example, if your B's are in real analysis/linear algebra/math stat, that may be unfavourable. Your recommendation letters play an even more important role than the GPA. If you get strong letters from eminent faculty members highlighting your research potentials, that would really help you. I agree with bayessays that your verbal score is so low that it would raise concerns about your ability to communicate in English. Though verbal section of the GRE is one of the less important aspects of the whole application but still you do not want to stand out in a bad way. I would say at least aim to get 50th percentile in order not to be screened out. 

Edited March 22, 2018 by statfan

[NumStats]

17 hours ago, bayessays said:

Yes, to be blunt, you are aiming far too high with that GPA and GRE score. 

Yeah I realized that.. Thank you for being honest, I know I was dreaming too high given the efforts I put in my first few years as an undergrad.

13 hours ago, statfan said:

The admission process is very complicated and no single factor plays a decisive role. GPA is only meaningful given certain context. A lower gpa may be more impressive at a prestigious instituition than a perfect gpa from an unknown institution. People from lower-ranked instituition will have a much tougher battle. Also, your math/stat gpa matters much more than the overall gpa. What is your math background? How well did you do in each course? For example, if your B's are in real analysis/linear algebra/math stat, that may be unfavourable. Your recommendation letters play an even more important role than the GPA. If you get strong letters from eminent faculty members highlighting your research potentials, that would really help you. I agree with bayessays that your verbal score is so low that it would raise concerns about your ability to communicate in English. Though verbal section of the GRE is one of the less important aspects of the whole application but still you do not want to stand out in a bad way. I would say at least aim to get 50th percentile in order not to be screened out. 

I believe my institution is around top 50 in the US, but I am not so sure how the other schools rank it.

Here is my relevant coursework list:

Probability and Stochastic Processes Series : A's

Numerical Analysis : A

Bayesian Statistics (Grad Course) : A - 

Linear Algebra (I II) : B+ A

Real Analyisis (I II) : B A

Statistical Methods (Grad Course) : A

Linear Regression (Grad Course) : A

Machine Learning : A

Econometrics : A+

Data Analysis in R : A-

Python Series : A's

Discrete Math : A

Math Finance : A

Risk & Uncertainty : A+

 

For my recommendation letters, 1 of them comes from my honors' adviser (research mentor as well), 1 of them from independent study supervisor, 1 of them from my Bayesian professor.

For my Verbal, I realized that it was low, but did not have another time to take another one (I was applying pretty late). I believe my communication skill is acceptable (at least my honors' adviser said so). If I don't get anything this year, I'll definitely prepare better for next year.

 

[statfan]

Your grades should be fine for the schools you applied but your math background is relatively light compared to strong applicants. You would want to have at least one mathematical statistics in your transcript, perferrably two. Maybe the problem lies in your letters. Generic letters talking about how well you did in their classes won't help for phd application, instead you would want to have letters talking about your research. Also, having one or two amazing letters from top faculties could dramatically improve your chance. 

Edited March 23, 2018 by statfan