Before you start agonizing over your personal/research statement for stat or biostat, read this.

[cyberwulf]

Every admissions season, many students applying to statistics and biostatistics programs are intimidated by the task of writing personal or research statements. Indeed, there is an entire sub-forum on GC dedicated to SoPs (Statements of Purpose) where there is much hand-wringing over how to craft the perfect text.

 

But while I can't speak for disciplines outside of the statistical sciences, I can confidently say that in stat and biostat, the evidence strongly suggests that personal statements have little impact on admissions.

 

I've written several posts about this in the past; here's a summary of why you should stop worrying so much about a 1-2 page essay:

 

1. Mathematical ability is best assessed through academic records and test scores (and to a lesser extent, letters), so it is generally quite easy to order students on this important trait.The pool of students applying to statistics and biostatistics departments isn't particularly deep, so that a major concern of even excellent departments is whether applicants can handle the requisite mathematical coursework and exams.

2. Very, very few applicants have meaningful statistical research experience before starting graduate school. As a result, many students end up working on dissertations in areas entirely different than they were initially interested in... and this is totally OK!

3. Funding in most (but not all) U.S. stat/biostat programs is allocated at the department level to the strongest incoming students, so applicants aren't typically "matched" to potential advisors who agree to fund them*. Rather, the department projects the total number of positions available and then tries to recruit up to that number of students. Once the students are on campus, they are then either assigned to a position or (ideally) have some choices available to them.

 

Given points 2 and 3, declarations in the personal statement such as "I am very interested in studying [X] with Professors [u,V,W]" usually carry little weight. They typically translate to: "[X] is a hot topic which I know very little about but sounds interesting, and I see on your website that Professors [u,V,W] list [X] as a research area." Which, again, is JUST FINE, since that's essentially all most people can credibly write.

4. Research potential *is* important, but the best source of information on this trait is letters of recommendation, not a one-page essay. In some fields, part of showing research potential is demonstrating that you have already thought of a reasonable project that will turn into a dissertation. Since (virtually) no one applying to stat/biostat has a "shovel-ready" dissertation idea, research potential is generally assessed using some combination of mathematical ability, creativity, and perhaps some exposure to lower-level research, all of which are best evaluated using other parts of the application.

 

I don't mean to denigrate the personal statement too much. There are a few key things to avoid (eg. rampant grammatical errors, aimless rambling, saying you have no intention of pursuing an academic career if you are applying to a PhD program) and of course there will be exceptions to every rule, but in general, as long as the PS is competent it probably won't affect your chances of admission significantly. 

 

Edited June 24, 2013 by cyberwulf

[cyberwulf]

As an addendum, the one main exception to the "personal statements aren't important" rule is applicants who have an unusual academic record/background compared to the typical stat/biostat grad applicant. This could include, for example: 

 

- Applicants who have been out of school for several (5+) years.

- Applicants who are changing fields.

- Applicants who lack prerequisites.

- Applicants whose academic performance was affected by serious personal/medical circumstances (e.g., one semester of terrible grades due to death of parent, major illness, etc.)

 

If you fall into one of the first three categories, we want to know why you chose statistics/biostatistics and how you think your background has prepared you for success in grad school. If the last category applies, it's important for us to know since it provides needed context for interpreting your previous academic performance.

[biostat_prof]

I agree with most of this. However, I will say that point 3 is not true for my department. We typically don't know exactly how much funding is available when we admit students, and while we try our hardest to find funding for all admitted students, quite frequently there isn't enough funding for everyone. The admissions committee will try to make sure that funds are available to their highest-priority students, but for many students whether or not they get funding may depend on whether or not a faculty member with a research grant wants to fund them. And that is more likely to happen if you have already expressed interest in working with that faculty member.

 

So I do think it is worth the time to mention the names of a couple prospective faculty advisers, particularly if you have talked to them before you apply. That said, it probably won't make a difference whether or not you get admitted, and it won't affect funding at many schools, either. And you could potentially hurt yourself if you say something that makes it sound like you don't know what you're talking about. But in general I think it is worth the time to say a few words about the research interests of a couple faculty members in your personal statement.

[agent229]

So my feeling is that in stats it's not necessary to name-drop in the statement of purpose, but can't really hurt. I am currently debating whether or not to add names to mine. I already specifically talk about the kind of research I am interested in currently (at what I believe is an appropriate level of specificity) and mention that the university in question has several people I would be interested in working with, but is that too vague?

 

If I should put the names, do you think it would be bad to mention professors in the statement of purpose without contacting them first? I have not really contacted professors because I've gotten advice that it's generally mildly annoying at best. If that was bad advice, would it look really bad to suddenly contact them right before the deadline? 

[runningincircles]

As someone who is changing fields (I have a masters in gender studies, and I now plan on applying to ms programs in statistics/biostatistics,) I feel that it is more incumbent upon a student like myself to have a deeper knowledge of what I'm getting myself into, as opposed to those with a STEM background. Is this correct? Like a lot of other people, I don't have a background in statistics or statistical research that I can refer to, so I'm going on the impression I've gained of the field through my own research.

I've written below a very rough draft as to my reasoning for pursuing an MS in statistics. Any opinions would be very welcome, but please be kind.

My reason for going towards stats, with the possibility of maybe pursuing a Ph.D at some point, is the fact that when I've looked into research careers, I've come across the need for research associates and assistants to have the ability to process and analyze statistics. I have always wanted to get into research, and due to my rejection to a Ph.D program in my prior field, I was forced to go back to school to improve my quantitative capabilities (I got a very low score on the quant side of the GRE.) As a result, I found out that I was naturally more suited to the study of mathematics than the social sciences, with mathematics being less emotionally taxing than the field of gender violence and the awful things that I studied previously. All in all, I feel that my personality is suited to research, and preferably the research of data that I don't have to connect with as rigorously on an emotional level. As someone who is extremely analytical, and has a mind that works over time, I feel that the past few years studying math have been the most pleasurable for me personally, and I truly believe that this is where I belong.

I don't want to discount my prior education, as I know it definitely enhances the education I am receiving at present. However, given the chance, I would probably go back and do an undergrad in mathematics or statistics instead of political science and Japanese.

Is the above a good starting point? I can also highlight the fact that as a female, I was subjected to the social stereotype that often prevents women from following careers within a quantitative field, and additionally had an upbringing that didn't foster good practices with regards to studying math (my parents liked to move a lot, and I ended up going to five different high schools and dropped out at 15).

My GPA is currently 3.8 in all math classes I've taken, I'm in calc 1b as of now.

Finally, I'm a Brit who immigrated to the U.S., and all of my undergrad and graduate education was obtained in the U.K. I'm a perm resident, so I wouldn't be classified as a foreign student.

Thanks for reading!

[Stat Assistant Professor]

runningincircles,

 

I think it is definitely possible to go the statistics route. I was also a non-traditional student, with an undergraduate degree in a qualitative social science. However, after taking several math classes, I decided I enjoyed it so I took several more math classes and applied to Applied Math MS programs. Now two years after finishing the Masters, I've applied to PhD programs in statistics for fall 2014. Your math background sounds a bit light as of now, but if you take multivariable Calculus, linear algebra, Calc-based probability, and Calc-based statistics, you should at least be in the discussion for admission to Masters programs in statistics. If you want to go the PhD route, you should definitely also take real analysis (and preferably some additional proof-based courses) as well.

 

You will definitely want to explain your motivation for changing to math in your statement of purpose (or elsewhere in your applications), although you probably should not mention that you found your former field of study too emotionally taxing or that you were rejected from other PhD programs in your former field. You can frame it like, "I decided that in order to better prepare for myself for a PhD in social sciences, I had to improve my math background. While taking Calculus, I discovered that I had an aptitude and affinity for mathematics and now wish to pursue a Masters in statistics." I think that would play better.

 

You will also need to retake your GREs to obtain a higher Q score.

 

Best of luck.

Edited March 6, 2014 by Applied Math to Stat

[runningincircles]

Thanks for your response. I'm going to take linear algebra, multivariable calc, diff equations, and two courses in probability theory, which make up 2 of the core courses for UC Davis MS in Statistics. I'm also planning on taking the GRE, which I know will be a lot better on the quant side now!!

Thanks for the advice about the personal statement too. I definitely think that's the correct route to take.

[rlatngus]

 

On 6/24/2013 at 11:20 PM, cyberwulf said:

Every admissions season, many students applying to statistics and biostatistics programs are intimidated by the task of writing personal or research statements. Indeed, there is an entire sub-forum on GC dedicated to SoPs (Statements of Purpose) where there is much hand-wringing over how to craft the perfect text.

 

But while I can't speak for disciplines outside of the statistical sciences, I can confidently say that in stat and biostat, the evidence strongly suggests that personal statements have little impact on admissions.

 

I've written several posts about this in the past; here's a summary of why you should stop worrying so much about a 1-2 page essay:

 

1. Mathematical ability is best assessed through academic records and test scores (and to a lesser extent, letters), so it is generally quite easy to order students on this important trait.The pool of students applying to statistics and biostatistics departments isn't particularly deep, so that a major concern of even excellent departments is whether applicants can handle the requisite mathematical coursework and exams.

2. Very, very few applicants have meaningful statistical research experience before starting graduate school. As a result, many students end up working on dissertations in areas entirely different than they were initially interested in... and this is totally OK!

3. Funding in most (but not all) U.S. stat/biostat programs is allocated at the department level to the strongest incoming students, so applicants aren't typically "matched" to potential advisors who agree to fund them*. Rather, the department projects the total number of positions available and then tries to recruit up to that number of students. Once the students are on campus, they are then either assigned to a position or (ideally) have some choices available to them.

 

Given points 2 and 3, declarations in the personal statement such as "I am very interested in studying [X] with Professors [u,V,W]" usually carry little weight. They typically translate to: "[X] is a hot topic which I know very little about but sounds interesting, and I see on your website that Professors [u,V,W] list [X] as a research area." Which, again, is JUST FINE, since that's essentially all most people can credibly write.

4. Research potential *is* important, but the best source of information on this trait is letters of recommendation, not a one-page essay. In some fields, part of showing research potential is demonstrating that you have already thought of a reasonable project that will turn into a dissertation. Since (virtually) no one applying to stat/biostat has a "shovel-ready" dissertation idea, research potential is generally assessed using some combination of mathematical ability, creativity, and perhaps some exposure to lower-level research, all of which are best evaluated using other parts of the application.

 

I don't mean to denigrate the personal statement too much. There are a few key things to avoid (eg. rampant grammatical errors, aimless rambling, saying you have no intention of pursuing an academic career if you are applying to a PhD program) and of course there will be exceptions to every rule, but in general, as long as the PS is competent it probably won't affect your chances of admission significantly. 

 

Hi,

You wrote this a couple of years ago. Would you say that it still holds that the SOP is not a huge factor in admissions?

[cyberwulf]

On 11/8/2016 at 1:34 AM, rlatngus said:

You wrote this a couple of years ago. Would you say that it still holds that the SOP is not a huge factor in admissions?

Yes. Research experience is becoming a bit more common as stat/biostat filters down to the undergraduate level, but admissions continue to be driven largely by the three core components: grades, letters, and test scores.

[OuterEdge]

On ‎11‎/‎10‎/‎2016 at 1:42 AM, cyberwulf said:

Yes. Research experience is becoming a bit more common as stat/biostat filters down to the undergraduate level, but admissions continue to be driven largely by the three core components: grades, letters, and test scores.

Hello! Another year and a half later...

I am finishing up applications now for biostat programs. I've been thinking my GRE scores are acceptable (V: 160 84th percentile, Q: 167 92nd percentile) and my letters of recommendation should all be strong (I've been given one to read and it is very, very strong). However my GPA isn't the best. My undergraduate degree is in statistics and while my statistics courses had no issues with grades, a few math courses were a "miss" (not calculus or linear algebra, though).

In this case, would work experience in a data & research office and/or SoP help outweigh the GPA being a weaker part of my application?

Other things to maybe consider: 1) I've taken a few courses after earning my degree, some of them graduate level statistics courses, and have done well. 2) About a year after graduating with my bachelors degree I was diagnosed with dysthymia and its suspected I have mild anxiety. While all of that is well under control now, it certainly played a negative roll in the end of my undergraduate career. Should this be briefed in my SoP?

[Schweinchen]

On 6/24/2013 at 2:20 PM, cyberwulf said:

But while I can't speak for disciplines outside of the statistical sciences, I can confidently say that in stat and biostat, the evidence strongly suggests that personal statements have little impact on admissions.

That's what I heard too from a pretty reliable source. But she also said that the SOP becomes important when you are on the border line. If they are deciding between (1) admitting you or someone else, (2) moving you or someone else off the waiting list, (3) awarding a prestigious fellowship to you or someone else, etc, your SOP begins to play a critical role because they have already compared you with that someone else using other yardsticks, the only thing they haven't used are your SOPs. Also, my undergraduate advisor told me that your SOP stays in your file even after you're admitted and they consult it before deciding whether to nominate you for NSF, DOE, Hertz fellowships. So it is best to take it seriously; I definitely am.

[theduckster]

@cyberwulf Your insights here are thoroughly appreciated by both myself and many others.

I am currently writing the statement of purpose and find that there is no getting around making a reference to a health issue and the impact it had on me over parts of the last year. I have, however, fully recovered in that time.

Is there a preferred format for including such a point in my SOP? For example, should I spend ~100 words to demonstrate to the adcoms that I've recovered and won't be an academic liability as a grad student (for MS programs)? Or should I keep the whole thing extremely brief?

Thanks, as always.

Edited November 25, 2018 by theduckster

[cyberwulf]

20 hours ago, theduckster said:

Is there a preferred format for including such a point in my SOP? For example, should I spend ~100 words to demonstrate to the adcoms that I've recovered and won't be an academic liability as a grad student (for MS programs)? Or should I keep the whole thing extremely brief?

It's fine to mention that this is not an ongoing issue, but you don't need to say much more than that.

[theduckster]

1 hour ago, cyberwulf said:

It's fine to mention that this is not an ongoing issue, but you don't need to say much more than that.

Would it be better to just not mention it at all? This is the advice I'm getting from some people, since the courses I performed poorly in were non-math/stats courses (and a part-time status is common enough that apparently many adcom members don't bat an eye when they see it).

Mainly, what I'm hearing is that any mention of health potentially sets off red flags in committee members' eyes if extremely strong evidence of recovery is not present (in my case the issue was only resolved ~6 months ago, so I only have my "summer months" as proof - not exactly 100% compelling on paper).

Edited November 26, 2018 by theduckster

[Gauss2017]

The ADA makes it very illegal to discrimate against anyone with a health issue. If necessary special procedures need to be in place to insure compliance during the admissions process.  It would be very informative if @cyberwulf could discuss what procedures are in place at his university to insure compliance with the ADA during the admissions process. I think that would help address ducksters concerns

[baabaablacksheep]

Hi, @cyberwulf, just wondering if this was still the case/whether people have been getting more research experience lately?

[cyberwulf]

On 9/15/2020 at 9:24 PM, baabaablacksheep said:

Hi, @cyberwulf, just wondering if this was still the case/whether people have been getting more research experience lately?

There's a lot more stuff that I would call "research lite", for example REUs, summer internships with professors, senior research projects, etc. However, since many applicants now have this, and the fact that it's almost impossible to gauge how much a student has really contributed to any research outputs they list, it doesn't really move the needle that much. Also, these opportunities are much more readily available to students at larger and more elite institutions, and it doesn't make sense to penalize those at smaller colleges who might have excelled if given these "research" opportunities but weren't able to access them.

[fujigala]

For MS programs with no research requirement, such as Stanford, do you think it is necessary to look into the research at the program and mention in the SOP which professors one would be most interested in doing research for? I am having trouble with this, since my interests are more applied, in biostatistics, finance, etc., and a lot of the research in statistics is heavily theoretical, i.e., developing new methodology, etc. Would it just be acceptable to state these areas of interest? My ultimate goal is just to work after I get my MS, maybe do some research over the summer. Any insight you may have into this would be very helpful. Thanks.

[bayessays]

On 10/31/2020 at 7:49 PM, fujigala said:

For MS programs with no research requirement, such as Stanford, do you think it is necessary to look into the research at the program and mention in the SOP which professors one would be most interested in doing research for? I am having trouble with this, since my interests are more applied, in biostatistics, finance, etc., and a lot of the research in statistics is heavily theoretical, i.e., developing new methodology, etc. Would it just be acceptable to state these areas of interest? My ultimate goal is just to work after I get my MS, maybe do some research over the summer. Any insight you may have into this would be very helpful. Thanks.

You can just mention some broad areas like that (biost/finance) and it's fine.  Most people getting an MS in stats are going to be interested in applied topics, and lots of Stanford PhD students are doing "applied" work that fits in those topics but also involves a lot of math.  It won't be a big deal

[tc1234]

Hello! I am currently preparing for fall 2023 biostats and stats phd applications.

I took honors (graduate) level real analysis 1 and dropped the class halfway (with notation on transcript). Then the next two quarters, I took the regular real analysis 1 (A-) and then 2 (B+). I am wondering if I should mention this in my SOP? Will this hurt when they look at my transcript? Thank you!