Choosing a program: better to go with MS or straight to PhD?

[aluc]

I'm trying to decide between a PhD offer at U. Missouri and a funded 2-year MS at Oregon State. I am interested in Bayesian ecological and spatial statistics, so I think Missouri would be a really good research fit. However, one big hesitation I have is that, due to the pandemic, I didn't get to visit Missouri and get a feel for the department and location. I've spoken to faculty and students, but there's definitely no substitute for an in-person visit before moving somewhere for 5 years. I've also seen people caution on this forum that research interests are likely to change over the course of a program so it can be risky to choose a smaller, specialized department based on what I expect to be interested in.

Taking the MS offer seems to have the benefit of letting me explore my research interests before re-applying to PhDs with more certainty and with a stronger profile. I also was able to visit OSU and liked the location. Since a PhD is my ultimate goal, though, maybe it's better to just take the offer I have in hand rather than gamble on the admissions process again (the next admissions cycle could be more competitive, getting straight A's in the MS will be easier said than done, etc.). Also, if my interests stay fixed, then I would have just wasted time not going straight to Missouri since I'm not aware of too many other programs with such a strong focus on the areas I mentioned.

I'm not sure how much weight to give each of these considerations, so I would really appreciate some outside perspective from the more experienced members here. I'd appreciate any general thoughts others can provide on these two departments, as well. Thanks in advanc!

[Sigaba]

No matter which path you choose, you will be wasting time if you work as hard as you can to do your best.

Given the gloomy economic forecasts for post-COVID-19 America, I would recommend that you consider the possibility that a couple of years from now, spots in graduate programs will be especially competitive. If you look back through the years of this BB, you may notice that there was more activity during the Great Recession than the last couple of application seasons.

While some of the decreased activity is likely due to the change of ownership, the availability of other resources, and more lurking, it's my hunch that ten years ago, some people went to graduate school as a way to weather the storm.

A small point. As a student at Missouri, you will be required to fulfill the requirements for a master's degree. http://catalog.missouri.edu/undergraduategraduate/collegeofartsandscience/mathematics/phd-mathematics/ If you get to the point where you want to go elsewhere or do other things, you can do so with a master's degree in hand.

[statsday]

I think Sigaba is giving solid advice. I thought about reapplying next year but it’s likely to just be more competitive and I have an offer from a program I think I would like. For bayesian and spatial stats (not applied to ecology) TAMU has some faculty. Matthias Katfuss comes to mind, so maybe after getting your masters you can go to college station. I will likely be there once this virus finally stops and I can say they do have quite a few people doing spatial stats and bayesian stats. 

 

Sigaba what did you mean when you said, “No matter which path you choose, you will be wasting time if you work as hard as you can to do your best”? 

[bayessays]

Did you apply to the other PhD programs in your original profile eval?  I remember your initial profile you posted and I'm pretty surprised Mizzou is the best school you got into, but programs are getting more and more competitive - I went to a visit day for a program ranked around 40 and the applicants mostly had MS degrees from top 20 programs or were from Ivy League schools.  Have you talked to the professors at Mizzou?  If you can work with people like Wikle and Holan, Mizzou might not be a bad option for you depending on your goals.  Getting an MS and establish a record of high grades will help you a little bit, but honestly not sure it's going to help a ton given that the classes you take in Oregon's MS likely won't be much more impressive than the math/stat courses you took at the directional state U.  I'd ask how much you would prefer living in Oregon for 2 years over the midwest (Columbia is supposed to be a pretty decent college town), whether you'd be ok with having to settle for a program ranked lower than Mizzou if things don't go well, and how sure you are that you want a PhD.  I think the MS could make sense if you're on the fence, if you don't want an academic career (and thus would be ok going to a lower ranked PhD program later), or if you're generally not in a big rush and want to live in Oregon and aren't that enthusiastic about Mizzou. But I think if you know you can work with some of the bigger names at Mizzou, then it's probably a pretty solid place to go for your research interests and you should strongly consider it.  The top 40 programs are getting extremely competitive for PhD admissions.

[aluc]

@bayessays, I didn't apply to that full list, but I did cover quite a range. I am still waiting on replies from a handful of PhD programs like Ohio, but assume the chances of those coming through are not huge this late in the process. I have to wonder how big of a factor my subpar GRE Q might've been or if I didn't get the SOP quite right--both things I could hopefully correct before next time. I also had it in mind to try and get a paper at least submitted while at Oregon to boost my chances, though I know accomplishing that wouldn't a given. Apparently they have sent a few students to top 50 PhDs recently, but since I can't know what the rest of their profiles looked like, I'm inclined to believe you if you think the MS wouldn't do much for me.

I definitely want to keep the academic career options open if possible and I'm fairly certain I could work with the bigger names at Mizzou. Given that, it sounds like spending the time at Oregon would only be dragging things out at best and counterproductive at worst with the increased competition every year.

[Bayequentist]

Something to consider: given your strong background in math and stat (for a program like Oregon State), it should be no problem for you to get straight A’s in Casella & Berger.

[Stat Assistant Professor]

7 hours ago, bayessays said:

Did you apply to the other PhD programs in your original profile eval?  I remember your initial profile you posted and I'm pretty surprised Mizzou is the best school you got into, but programs are getting more and more competitive - I went to a visit day for a program ranked around 40 and the applicants mostly had MS degrees from top 20 programs or were from Ivy League schools.  Have you talked to the professors at Mizzou?  If you can work with people like Wikle and Holan, Mizzou might not be a bad option for you depending on your goals.  Getting an MS and establish a record of high grades will help you a little bit, but honestly not sure it's going to help a ton given that the classes you take in Oregon's MS likely won't be much more impressive than the math/stat courses you took at the directional state U.  I'd ask how much you would prefer living in Oregon for 2 years over the midwest (Columbia is supposed to be a pretty decent college town), whether you'd be ok with having to settle for a program ranked lower than Mizzou if things don't go well, and how sure you are that you want a PhD.  I think the MS could make sense if you're on the fence, if you don't want an academic career (and thus would be ok going to a lower ranked PhD program later), or if you're generally not in a big rush and want to live in Oregon and aren't that enthusiastic about Mizzou. But I think if you know you can work with some of the bigger names at Mizzou, then it's probably a pretty solid place to go for your research interests and you should strongly consider it.  The top 40 programs are getting extremely competitive for PhD admissions.

Agreed that if spatial statistics is your interest, then Mizzou would be a good choice if you can work with Wikle and Holan. I know of a few Assistant Professors at decent schools (Florida State, Colorado School of Mines, etc.) who were postdocs at Mizzou and were supervised by these two. I actually think that some mid/lower-ranked statistics programs may provide more opportunities to work on spatial and environmental statistics. For example, Colorado State, Ohio State, and Missouri seem to have more faculty working on these areas than some of the "top" schools. This might be because spatial stats is a research area more concentrated in a Biostat rather than Stat dept at "elite" schools, though.

Edited March 30, 2020 by Stat Postdoc Soon Faculty

[aluc]

13 hours ago, Stat Postdoc Soon Faculty said:

Agreed that if spatial statistics is your interest, then Mizzou would be a good choice if you can work with Wikle and Holan. I know of a few Assistant Professors at decent schools (Florida State, Colorado School of Mines, etc.) who were postdocs at Mizzou and were supervised by these two. I actually think that some mid/lower-ranked statistics programs may provide more opportunities to work on spatial and environmental statistics. For example, Colorado State, Ohio State, and Missouri seem to have more faculty working on these areas than some of the "top" schools. This might be because spatial stats is a research area more concentrated in a Biostat rather than Stat dept at "elite" schools, though.

That is good to hear. Like I said, I'm very interested in spatial stats, but I was hoping to get to explore other areas as well. I'm sure the coursework will largely satisfy that curiosity, but it does feel daunting to commit to a specific area of research so early on. Is this a valid concern to have, or am I overthinking it? Do you also think going for the MS would not meaningfully bolster my profile?

[bayessays]

Mizzou has some other solid people like Tony Sun working on non-spatial things, and they have people doing other things like genetics too.  I don't think you'll be "locked in" there, though they do have a strong environmental focus.  Oregon State, though not as good of a department, also has a more environmental bent.  If you're interested in environmental stuff now, I don't think you're going to just have an epiphany in the next 2 years that you hate environmental stuff and can't figure out something to work on.  You're not committing to research this topic for life, just for a couple years, and Mizzou has people working on different things anyways - even people like Wikle are doing plenty of non-spatial stuff.

[Stat Assistant Professor]

25 minutes ago, bayessays said:

You're not committing to research this topic for life, just for a couple years, and Mizzou has people working on different things anyways - even people like Wikle are doing plenty of non-spatial stuff.

Bingo. For my postdoc, I started working on semiparametric and time-varying models, which I had never touched at all in my PhD dissertation research. You can always switch your research focus later after your PhD (if you decide to stay in academia). Another one of my PhD advisor's former students did small area estimation for her dissertation, but then she changed her area as a postdoc and is now a world-renowned expert in the area of record linkage (at a top school, I might add). The PhD is more about learning *how* to do research than it is about a specific topic. Once you master that and figure out the best system that works for you personally, you can teach yourself most things.

[ICGEDelhi]

I would recommend you to wait as the situation is worst due to COVID19 and would like to suggest to go for Phd as this is your dream. Make sure to choose the university very careful.

Good Luck