Should I learn R for a project? I am in charge of an academic research project, building predictive models based on clinical data from multiple longitudinal cohorts of diseased patients. The total number of patients and controls (over all cohorts) is about 1500. Each patient has multiple visits. In total there are around 100-400 items for each patient. I have 8-15 months to complete the project.
I have no experience with statistics or R. I have a PhD in another area of applied math and know Matlab. 
The question: is this a good opportunity to learn R? Will learning R help me get through this project or will it be a source of frustration?
Some considerations:


*

*I know I will enjoy learning R  

*It's a skill that may/may not benefit me in the future

*It does not matter what I leave behind in terms of code etc

*The output of the project is a publication

*It is fine if progress is slow to begin with

 A: Coming onto CrossValidated and asking if you should learn R is a bit like going to a football match and asking if anyone likes beer.  Nevertheless, with this little bit of selection bias out of the way, here is my view on the matter:

R is a widely used statistical programming language in academia and some parts of industry, so learning this skill is likely to be useful for any subsequent statistical analysis you perform.  Most data analysis and data science work is performed in either R, Python or SAS these days, with the first two becoming highly popular in recent years.  Like anything, devoting time to learning a new skill is a trade-off, and so you should look at opportunity cost.  If you decide not to learn R, what will you use that time for instead?  How do the alternative uses of that time compare to the value of learning this statistical language?  My experience inside and outside of academia is that having goods skills in R programming is highly valued, and there are a lot of people in the clinical/bioscience area that use this language as their primary tool.
If you do decide to learn R then I cannot recommend anything more highly than subscribing to DataCamp.  (They have a special offer running at the moment to subscribe at 50% off normal price.)  This will allow you to do a range of courses focused on each aspect of data analysis and statistical programming, and it is much easier than trying to learn from self-study.  In six months you should be able to learn a substantial amount.  (I recommend learning the dplyr and ggplot2 packages as top priority, as well as all the basic programming.)  Personally, I learned R during university while doing some projects, and was mostly self-taught.  I later did the accreditation program at DataCamp and it filled in a lot of gaps and was very simple and helpful.  You are very lucky - I wish this stuff was around when I was first learning the language!
A: Learning R will benefit your career. It’s widely used for analysis in many fields. It’s a great addition to your resumé and shows that you can learn new skills as they’re needed for a project. It will give you more experience with performing different analyses and more insight into work different programming languages work so you can choose the best tool for each job.
You should take any opportunity to learn new transferable skills. It is rare to have sufficient time to invest in your professional development. You will be able to use these skills future project and may even seek projects that require more technical skills once you are able to do them. If you go on to mentor students then experience outside your comfort zone, learning a new language will help you understand what they’re going through when they need to use new tools and techniques in their research.
R is an open source language so you will be able to use it, even in working environments where you can not access MATLAB. It’s far easier to release you code (as a stand-alone research output) or share it with collaborators. You’ll also have access to a wide array of R packages. With these you can perform analyses specific to different fields without having to implement these methods yourself. The plotting capabilities available in R will also benefit your future projects and make it easier to communicate your research results.
While you may have flexibility on this project to produce the results by any means, it’s still worth considering the opportunity to update the tools in your workflow. Using R is encouraged to perform reproducible research. You can document the analysis that you’ve performed and export reports with Rmarkdown so that collaborators can see exactly what you’ve done. This will also be beneficial to you if you need to run it again on new data or after correcting a mistake. When you come to write up the project, a good record of the analysis that you’ve performed and figures you’ve generated will also be useful.
