dmacfour

If at all possible, stick around for another semester or two to get that GPA up above a 3.0. Taking and nailing a few math classes would help prove that your early math struggles were circumstantial. Also, don't sweat the W's - I had like 10 of them and it was never brought up.

The applied stats program I did was a great start, but I'd like to do something a little heavier on theory if I do a PhD. The probability theory/math stats sequence was easily the most interesting part of the program for me, and I wasn't satisfied with how in depth it was. I've mostly been looking at traditional stats and biostats programs. I guess I'm curious if my profile would be competitive for any traditional stats program, even if it's outside of the top 50.

I'm look at applying to PhD programs (pretty early in the process), and am wondering how high I should aim, realistically. Just a little bit of background: I have a B.S. in Psychology, M.S. in Psychology, and an M.S. in Applied Statistics. GPAs were 3.01, 3.4, and 3.7, respectively. In between the masters programs, I took calc I-III and Linear Algebra and had A's in all of them except calc II, in which I got a B+. After the stats program, I took Abstract Algebra and Intro to Real Analysis and got an A in both. I last took the GRE in 2017 and ended up with ~80th percentile on the quant section (if I remember correctly). I feel pretty good about my transcript and can probably get the GRE quant score up, but I am concerned about the LOR portion of the applications. I've been working in a professional setting for the better part of the last decade. I've worked as a data scientist (my current role), a researcher in multiple labs, and as a product manager, and could get professional recommendations from a number of colleagues/managers. As for relevant academic references, I have no idea what to do. The stats program was accelerated, so I grinded through classes for 2 semesters and completed a capstone project. I had very little interaction with the faculty. Any thoughts where to go from here?

I saw that CMU dropped it for their traditional stats PhD (the joint ones still require it).

To be honest I'm not necessarily even aiming for top 50. I have several publications, but they all involved applying statistics in different contexts (Sleep research, ag research, education research, employment forecasting, and experimental psych).

Gotcha. Now I'm wondering if I should just leave my GRE scores as-is and hope the rest of my application makes up for it. I'm planning on applying to mid-tier programs in the fall.

I've been hearing a lot of talk of programs dropping the requirement, but thus far, I've only found one that doesn't. Is anyone tracking this information? I'm currently looking at Statistics and Biostatistics programs. I've taken the GRE twice in the last 10 years and really hate the idea of throwing more money at ETS. My scores were good enough to get into an applied statistics program a few years ago, but they are not at all competitive for what I want to do now (155V, 156Q). It would be nice not to stress about taking it again and just focus on acing real analysis this summer. That would leave me with straight A's in calc, linear algebra, abstract algebra, and real analysis, which I think is a stronger statement of my abilities than a standardized test with controversial validity.

My undergrad is in psychology, and I took calc I-III and linear algebra prior to completing a terminal masters in applied statistics. I wanted to take real analysis before considering PhD apps, but due to time and money constraints I've only been able to take its prereq (abstract algebra) so far. Unfortunately, I don't know if I'll be able to finish a real analysis class before apps are due for some of the schools I'm interested in next fall. It's likely that I'll be enrolling in it in the fall semester and taking the final the same week some apps are due. As someone without a quant heavy undergrad, Is real analysis going to be critical for me to establish a good quantitative track record?

My biggest concern with PhD applications is going to be who to use as a recommender. I finished a course based masters in statistics a couple of years ago and could tap some of the faculty for a letter, but to be honest I didn't get to know them well at all. Outside of that, my letters would have to come from researchers in sleep or behavioral science. None of them have a formal background in math or stats (closest is a guy who specialized in psychometrics), so they wouldn't be able to say a lot about my quantitative abilities. Mostly just the way I applied statistics to their particular problems. Is this going to be problematic?

That's good to hear! My grades are in good shape (~3.7 GPA), and I'm not overly concerned about rank. I am a bit concerned about LORs, though. Outside of the MAS program and math prereqs, I don't know of anyone with a quantitative background that could give me one. My academic and professional career up until 2017 was in the behavioral sciences.

After finishing my Applied Statistics program last year, I was left interested in going even further with the subject. It was made abundantly clear that the MAS program isn't suitable preparation for a PhD in Statistics, so I imagine that I'm not in much better of a position than an undergrad looking to go straight into a PhD. I'm currently looking for programs or professors that are doing research on applied topics. Any advice for someone in my position? I've taken Calc I-III, linear algebra, and the probability theory/math stats sequence offered by my applied stats program. I plan on taking more math classes in the meantime (real analysis and its prereqs, at the very least). Thanks!

My advice is to look carefully at the curriculum of each program. Some of them are glorified analytics programs that focus on coding, while others overlap significantly with regular MS programs. You'll find anything and everything in between that. I'm in CSU's Applied Statistics program and while calc I-III and linear algebra are the only math prereqs, I think I would have benefited from a few more advanced math classes. There were plenty of confusing proofs scattered throughout the program even though it wasn't a "theory heavy" program.

I have a masters in psych and went in to statistics. Same reasons for you. You'll have to take the entire calc series and linear algebra to get in to most programs (even the applied statistics program I went in to). Even after that, I felt barely prepared for the theory sequence in my program. If you want to a theory heavy masters (which you'll want to do if you're interested in a PhD), then you'll probably want to take a lot more math than that. By the way, what's your endgame for getting a PhD in stats? What academic statisticians do can be quite a bit different than what you're doing as an analyst and how you use stats in research. I chose the applied stats route because it better lined up with my career goals.

I finished my first masters at 24, and I'm finishing up a second this year in applied statistics at 30. I'm researching PhD programs in Quantitative Psych and Applied Statistics, but I want to spend several years working as a statistician or data scientist before applying.

We all were, but we weren't getting many call backs. This was in 2013 so it's possible that things have changed a bit since then.

Man, that would have been nice. Only one person in my cohort had a job lined up (at a place he was already working). The rest of us took 1-2 years to land jobs. I think location had something to do with it - the closest major city was 6 hours away, and it had a small market for UX and HF.

Do you know of any resources on placement rates in HF? Only about half of my cohort managed to break into HF or UX after graduating, and I'm wondering if that's part of a larger trend. The ones who broke into the field have awesome jobs, but they had to fight tooth and nail to get them.

What do you mean by "pretty damn high"? In my current masters program that means that you're doing a lot of mathematical proofs, but in my previous masters program that meant that you used calculus and linear algebra. I'm wondering if I need to take more proof based classes (like real analysis) to excel in a quant psych program.

As of this summer I'll have masters degrees in both applied statistics and experimental psychology, and the idea of bringing those two domains together for a PhD is very appealing. That being said, I don't know much about the field yet and was wondering if anyone here could answer a few questions: 1. What does the profile of a competitive applicant look like? 2. What are some hot areas of research to consider? 3. What level of mathematical ability is recommended to excel in a quant psych program?

For what it's worth, I have at least 7 W's on my transcript and I didn't stop me from being admitted to grad school for psychology or for statistics. In my opinion, there's a lot more room for justifying a W than there is for a grade less than a C.

I installed Linux over winter break and haven't had a reason to boot Windows back up. The biggest annoyance I've encountered is that Adobe Acrobat and MS Office have no Linux support. Another annoyance is that Ubuntu doesn't have a tablet mode like Windows does. Not a problem if you're okay with open source alternatives. R seems to run a hell of a lot faster on Linux than it did on Windows. I would explore virtualizing Windows if you need it for a particular task.

I think people seriously overestimate how bad W's look. I have several and I've never been asked about it.

Have you taken multivariable calc? That would be calc III at most schools in the US, not sure how it's split up in Canada though. If you haven't I'd see if you could take it to bolster your math grades. the B, B-, and C+ in your math classes might cause you problems.

Non-tradition kind of loses its meaning when talking about grad school. I'd guess that about two thirds of the students in my program are fresh out of undergrad and one third are between 25 and 40 years old. I'll be 30 in a couple of months (holy crap!).

One thing that influenced my decision: Even though statistics and data science are the hot thing right now, do I really want to wait 5+ years to enter the field? I've already experienced starting school when a field was hot and graduating (with no work experience) after it was saturated. I'm not willing to go through that again so I'm going to work for a few years after getting my masters and then evaluate where I am at.