cyberwulf

This is right on. PhD programs are money losers, and Masters programs are money makers. So departments have an incentive to increase the size of the Masters program even if it means decreasing the quality of each individual's experience. Of course, a slightly sub-optimal experience at a very reputable program might still set you up well for the future.

Berkeley and Michigan are, by quite a ways, the most prestigious departments on your list. Like virtually all top-tier stat departments, they are more known for their contributions in theory/methods than in application, but they are both quite strong in machine learning. Unless there is something particularly attractive about the specific programs (or faculty) at the other three places, I would choose between those two.

Yeah, this shouldn't be a problem. Provided you choose suitably rigorous coursework as a Masters student (many programs allow or require Masters students to take the same first-year math stat sequence as PhD students), you'll be better prepared for a PhD program in statistics than the vast majority of undergraduate students applying. Also, you may have the opportunity to participate in some "real" statistical research and get letters from professors who are well known and respected in the statistics community.

Of the programs you mentioned, UNC is clearly the best one. If you end up getting funding from them, that program would offer the best opportunities on either an academic or industrial path. Of the rest, while Vanderbilt is a pretty new program I think the overall quality of its faculty give it a slight edge over MUSC and FSU. If you are thinking of pursuing an academic career, the name of your dissertation advisor matters, and there are more "name" advisors at Vanderbilt. As far as industry goes, MUSC sends most of its graduates to industry so they presumably have established connections with employers. Vanderbilt, being new to the PhD graduate game, may have less of a pipeline in place. I don't know as much about FSU, but would rank it below Vandy and MUSC.

Berkeley is the better-regarded program, but not by a lot. I think you're right in your assessment that Berkeley provides a better springboard to an academic career while Emory is better-connected with government and industry. But since they're both high-quality programs, you would not be closing the door on one career path or the other no matter which you chose.

Oh, and it gets better! Numbers for Santa Barbara:

Is $22k even livable in LA? From a cost of living calculator I've used before (see pic, available at http://www.bestplaces.net/cost-of-living/) it looks like that $28k offer from FSU corresponds to a stipend of almost $50k (!!) from UCLA.

You've got another problem here beyond your grades, which is that if you are currently taking Calc 3 and linear algebra you will have minimal math background when you apply in the Fall. Though your list of target schools is ambiguous (UC = U of Colorado? U of Cincinnati? UC Berkeley?), you will have trouble getting admitted to a decent PhD program in statistics if both your math background and grades are relatively weak. Have you considered applying to Masters programs instead, and using those to improve your prep/profile before applying to PhD programs?

Take the stat course taught out of Casella & Berger. This is the core math stat material in most MS programs, so having been exposed to it before will enable you to learn it even better when you're doing your Masters.

Many independent study classes show up on transcripts under some generic name and give no indication of what was actually done. So, it's unlikely you'll get "credit" from an adcom for taking them. Of course, having exposure to measure theory prior to embarking on a PhD can only be helpful, as the grad school version will be much easier if you're seeing things for the second time. But, as @PhDStats mentioned, the main value of an independent study could be that you have the opportunity to impress the professor you're working with, which will lead to a stronger recommendation.

I would dress pretty much the same: professional wear, but no suit But you can get away with slightly more casual attire.

Well, there's a couple of "cons" right there. On the plus side, Penn State is a good school and their statistics department is well-regarded so I could see them developing a solid biostat program down the road.

Recruiting visits usually look exactly like interview visits, except that you aren't being evaluated. You'll meet students and faculty, and learn about the department.

There isn't a lot of time series analysis in biostatistics, actually. Also, I wouldn't judge the depth of a course by its syllabus; you can do a shallow treatment of a whole laundry list of topics just as easily as you can do a very rigorous presentation of a small number of topics. The reality is that neither of these courses will make or break your grad school application. If it were me, I'd take the one from the better teacher as you will probably learn more.

Inquiring has ZERO (I repeat, ZERO) impact on the admissions decision, for two reasons: 1) Asking when decisions will be sent out is a COMPLETELY REASONABLE thing to ask, especially when a program is lagging behind its usual timeline and other peer schools have already sent out acceptances. Obviously you can go overboard by emailing (say) every few days for several weeks, but politely inquiring once is totally fine. 2) Typically, the personal handling correspondence with students is an administrator who has no role in making admissions decisions. Even if they think you're an obnoxious pest (and for reason #1 this is unlikely), this opinion won't make it to the committee of faculty members (and possibly a student or two) reviewing applications.

Obviously this is going to vary by department but generally the probabilities range from about 0.7 to 0.95.

I'm also curious as to the reason the forum has been quieter this year. At least at our institution, applicant numbers have been about the same as last year so it isn't from lack of interest in statistics/biostatistics.

I wouldn't take DiffEq unless you don't have other good options. It's just not a widely used technique in biostatistics; it's much more of an applied math thing. And, as @2016biostat mentioned, definitely make sure to take multivariable calculus.

1. Taking pre-reqs at a community college is fine. Of course, the bar for what constitutes "good" performance at a CC is higher; if you don't get A's, that may raise some questions. Also, no biostat program that I'm aware of requires differential equations as a pre-requisite. 2. Research experience doesn't really factor into Masters' admissions, so the fact that you don't have any won't hurt you.

Just write what happened (family issues = one awful semester). It's a relatively commonplace occurrence, and I don't think that one "off note" will hurt your chances much.

I think you're being fairly realistic about reasonable target schools, but I would consider "reach" schools you're applying to is probably too large as I think your chances of getting into any of them As far as Biostat PhD programs go, you might also consider applying to places like Iowa, Pitt, and UTexas.

A 'C' is roughly what I'm imputing for an otherwise good student who takes a 'W' in a course, so I would view those two as essentially equivalent outcomes. Not everyone will feel the same way I do, though. Also, if you "have a history of doing very poorly on exams even when I know the material due to high performance anxiety," then you have to ask yourself whether you will be happy pursuing graduate studies, where exams are more challenging and often more high stakes than in undergrad. Most Masters programs in statistics have only a small research component, and success depends largely on a student's ability to get good grades in courses (which, in turn, will often depend on the ability to do well on exams).

A GMAT score, are you sure? That's not a typical standardized test for someone planning on pursuing graduate studies in math. Perhaps you mean the Mathematics subject GRE score? If so, a 75th percentile is fairly respectable and will not kill his chances of getting into a good math PhD program. Another confusing point: he's applying in Spring? Virtually all math PhD programs have deadlines in December-February.

When I see a 'W' (withdrawal) on a transcript in a more challenging mathematics course, my assumption is that the student was finding the course too difficult and dropped it because of that. So, withdrawing doesn't really "hide" a poor grade, and in fact might leave an even worse impression than a 'B' would. I would stick with the course; if many students are struggling, final letter grades may be higher than the numerical grades would suggest.

It would have zero impact on your eligibility. People turn down offers all the time. At most places, they probably wouldn't even notice that you were applying a second time.