Scared I'm not good enough for biostat because of programming

[pizzle]

Hi Everyone,

 

I want to do a biostat masters, but I'm scared I'm not good enough of a programmer to do well enough. Do you have to be really good at programming to do well in the industry and in the masters program? Do you have to be really good with data structures and algorithms? I've taken a few Java courses (I'm a finance major), but I REALLY struggled. Is this a bad sign for me? I don't want my programming skills to hinder me from getting a job/doing well in a biostat master program.

 

Thanks so much!

[svent]

Um, I doubt it. The people I knew in grad school who did biostats probably didn't know anything about Java or C or even Python. They knew some R or SAS or whatever things stats people use. I seriously doubt any of them would have known anything about, say, depth-first search.

[pizzle]

Thanks for your response. What about the whole SAS programming deal? I heard SAS is just a variant of C....

[svent]

I've never used it. I always assumed it was something easy and high-level, sort of like Mathematica. C is a pretty low-level language. Based on this page http://www.stat.ufl.edu/~aa/cda/Sas_web.pdf

 

it doesn't seem like you'd be doing anything low-level like manipulating pointers or embedding Assembly code. But I could be wrong.

 

(The code looks nothing like C.)

Edited August 12, 2015 by svent

[cyberwulf]

What about the whole SAS programming deal? I heard SAS is just a variant of C....

 

SAS is nothing like C. It's a very high-level language which allows you to perform data manipulations and analyses with a small number of relatively simple commands.

[pizzle]

SO I'm confused....what is the extent of programming for biostaticians? When I look on job boards for master students, most says that you need to have strong programming knowledge in R/STATA/SPSS/SAS. What's the most complicated things you do with these programming languages?

[trigga]

Those languages are very high level.  This might be a somewhat misguided analagoy, but languages like Java, C, etc can be used to design software.  To be a good programmer in those languages, you need a very good understanding of data structures, algorithms, etc.  Stata/SAS/R are statistical packages that are used to analyze data.  You won't be developing custom software with them.  You need to know programming techniques (defining variables, data types, looping, logical thinking in general, etc) but the process is not nearly as rigorous as programming in Java/C.  SAS, R and those other packages are custom built to analyze data.  You'll be doing things like grabbing data from databases, creating tables, cleaning up datasets, performing statistical analysis, etc and the packages generally make these tasks fairly straightforward.  If you get advanced with something like R, I believe you can create your own functions that requires more n depth prgraming knowledge, but I don't believe ts anyhtin you should really be concerned with, especially at the master's level. 

 

Summary - while all of the above is considered 'programming', using the statistical packages you mentioned are not nearlyas rigorous as developing custom applications with Java, C and other languages like that.    

[Sharonyash]

Programming in Graduate School varies from one college to another.

It also depends on the professor. I had NO CLUE of Matlab or Python when I went into Graduate school but I knew some SAS and SPSS and Minitab.

But now, I am really good in Matlab, Python and SAS. It is all about self-teaching and commitment. Professors won't teach you programming but expect you to learn. I was shocked when the professor (in numerical analysis) said, that we will use Matlab! I decided to learn using online tutorials, youtube and practice them.

As for SAS, it is a powerful tool. It is nothing like C+ or Matlab. You don't write code like in C+ but precisely for Statistical Analysis and research.

It takes time to learn but you can do it. Don't be scared to try --- you have to teach yourself to succeed.

hope this helps

Sharon

[Biostat_Assistant_Prof]

If you're intelligent enough to handle the math that a masters in stats/biostats requires, you can certainly learn SAS. Seriously, it takes maybe a month or two of regular SAS use to become a reasonably competent SAS programmer, but you'll start getting the hang of it after only a few days. 

R on the other hand has a steeper learning curve, but is lower level than SAS, more akin to 'traditional' programming. Even though it's primarily built as a statistical analysis software, I'd put the syntax/language of R closer to something like Python. It takes longer to become proficient in R as it requires a more traditional understanding of loops, building functions etc., but can be used for much more in depth analysis. 

Don't worry though, basic analysis is easy enough in both languages. 

[Clarinetist]

My first impressions: I'm no expert, but first impression of SAS I had was that SAS is like a strange hybrid of SQL and Python. Neither SQL nor Python take long to learn.

Edited August 18, 2015 by Clarinetist

[missleung]

I really don't think you should worry about the programming at all. Rather, I think you should worry about the statistical theories/concepts behind the Biostat degree...

Edited April 21, 2016 by missleung