# Right forum for stats grad school questions?

I know it isn't stats, but it is academia.

Which forum is most appropriate for stats-centric grad-school related questions?

There is a masters program that I am very interested in attending (remotely) and they want letters of recommendation.

Previously I had asked former professors, and they did most of the structural work in the letter, but when I asked peers, managers, and coworkers in industry they are asking me to provide a draft. It is really hard for me to "toot my own horn" - that is one of my biggest challenges in the annual review process, or in making a resume.

What resources are available for guiding an appropriate letter of recommendation that isn't plagiarism, that evokes the words and perspectives of the recommender?

• I know it isn't stats, but it is academia - why do you ask a question if you know the answer? (I'd say this answer is correct.) – amoeba says Reinstate Monica Mar 1 '17 at 21:40
• its stats-ish. The difference between the practice and the field. A bale of hay is hay. A field of hay is hay. One of them is found in a store and a storekeeper may be asked about it, and for another you need to ask a farmer. Even when they are both hay. – EngrStudent - Reinstate Monica Mar 1 '17 at 21:43
• I am positive about this being on-topic: where else would statistically-minded people be lurking to advise? I'd expect a letter of recommendation to touch as far as possible on strengths in mathematics, statistics, data analysis and computing, including formal qualifications and practical experience. But quite what is expected could be very specific to country and institution. On an international forum, you might need to identify your country or perhaps the kind of institution you are applying for. I'd not want to see specific institutions being named or discussed. – Nick Cox Mar 1 '17 at 22:08
• I think this type of question is a mixture of "too broad" with "opinion-based"; therefore not on-topic. You may take a look in Academia SE (look at their help center, and search for duplicates). – Andre Silva Mar 1 '17 at 22:48
• @NickCox: Well, on Academia SE if they're also interested in answering questions about letters of recommendation & such. Just as if they're also interested in answering questions about R programming they'll be lurking on Stack Overflow. A desire to reach the statistically minded doesn't suffice to make a q. on-topic here. – Scortchi - Reinstate Monica Mar 2 '17 at 9:24
• @Scortchi I can see both sides here. I am reminded of questions on statistical terminology, which some want to see banned as linguistic or semantic. To me the generic advice about recommendation letters is very basic; quite what statistical people prize is germane on CV in my view. – Nick Cox Mar 2 '17 at 9:32
• @NIckCox: Perhaps the two aspects could be separated: asking here about the statistical knowledge & skills relevant to post-graduate study, & on Academia about how to get colleagues to write a recommendation. (I think that, though that'd make the q. germane, it'd still be of the broad, opinion-soliciting type that we don't generally welcome - CW at best.) – Scortchi - Reinstate Monica Mar 2 '17 at 10:08
• @Scortchi - I think that you are asking "what is the soul of statistics". To "beg a metaphor", the question asking "which defines the person: dna (nature), upbringing (nurture) or the exercise of free will (choice)" implies more than one element can comprise an essence of a person, and thereby can comprise the essence of human fields like stats. The DNA of every degreed statistician is in part academia. The parsing into "pure rooms" as you are trying to do is one of the prototypical western philosophical falsehoods. "False either-or" is a logical fallacy. Don't engineer SE around it. – EngrStudent - Reinstate Monica Mar 2 '17 at 12:15
• A question can be either on-topic or off-topic on CrossValidated. This is not a logical fallacy. Any asked question will either stay open or get closed as off-topic. This is just how it is. – amoeba says Reinstate Monica Mar 2 '17 at 13:40
• @EngrStudent: I don't think I am. Deciding which SE site a question's most appropriate for is more akin to librarianship than philosophy. I don't dispute that some questions suit more than one site; but suggest this one might be more usefully split into its generically academical & specifically statistical aspects, & furthermore that not all good statistical questions are good CV questions. – Scortchi - Reinstate Monica Mar 2 '17 at 14:06
• @amoeba - my point is that "false either or" exists link. There are some things that are not "either-or". Boolean algebra is the simplest case: both and neither are also valid comparisons of two binary variables. Human beings are what Godel would call "sufficiently complex". This means that, however infrequent, there are fundamentally meaningful or critical human-things that don't fit cleanly into any bin or set of bins we make. You can argue this topic shouldn't be one if you like. – EngrStudent - Reinstate Monica Mar 2 '17 at 16:18
• @Scortchi - I strongly agree with and will conform to your comment with the following exception: librarianship is the child of philosophy, and wouldn't exist without it. There is no "clustering" without a distance metric and cluster definitions. Librarianship is deciding which point goes into which bin. Which bins to have, and how to apply a separation principle to a point intermediate to two bins is the subject of the philosopher. – EngrStudent - Reinstate Monica Mar 2 '17 at 16:25

I think this type of question is a mixture of "too broad" with "opinion-based"; therefore not on-topic in CV SE.

Which forum is most appropriate for stats-centric grad-school related questions?

Apparently, Academia SE. Their on-topic help page says: