Hi. I had very little (almost no) coding experience, but fairly good stats knowledge, so I did these free data camp courses: https://www.datacamp.com/courses/free-introduction-to-r. Then I did the following "Statistics with R" specialisation on Coursera: https://www.coursera.org/specializations/statistics. I really rate the Coursera courses, I chose to pay for them to force me to do it, which was well worth it in the end.

I would look for introductions to the packages that you're into rather than looking for a general introduction into R. I often use lavaan, lme4 and tam, for example, and all of them have a great support and tutorials on their website.

Still working at learning. I started with Sarah's approach. Data Camp, Coursera, edx, and others. I now have a few books" "R in Action", "R for Data Science", and "Biostatistical analysis in R." There are also web resources like stackexchange to get help, or the online documentation from cran-r-project.org. Sometimes just reading some of the help questions that I did not ask is useful.

The biggest problem is that I find it hard to remember unless I have a specific project that needs some method. I have also gotten into a habit of copy-paste code bits to make a program rather than typing it out each time. I think this may make it a bit harder to remember the code, and I risk misplacing pieces of code.

Two main reasons I dislike R. The documentation is not written to make it easy -- or at least I don't think so. I think the documentation might be very useful if I already know 90% of the answer and just need a quick reminder. I also don't like packages that disappear when the original programmer stops supporting the package. The Basic program that I wrote 30 years ago still runs. The SAS program that I wrote 25 years ago still runs. I don't have an equivalent in R, but I know people that have had their favorite package die, and their program then dies too.

Past experience can make using R easier and more difficult. My background: I took several semesters programming Fortran. Also had classes in C, Pascal, Basic, Machine language, ADA, and Lisp. I have a minor in statistics as part of my B.S. degree. I have used SAS off and on since about 1986. In the off periods I used Statistica, BMDP, SPSS, CricketGraph, Minitab, Design Expert, and others. The non-SAS programs were used when SAS was not available or when I needed better graphics. SAS Graph could make some lovely graphs if you had lots of spare time to work out all the strange codes for your specific printer.

Hi Jessica,

I found this really useful to start with: http://swirlstats.com/

It runs interactively through the R console which means you get to practice everything as it goes along.

Good luck!

Helen

Hi everyone,

Thanks for your feedback. I am writing a course on how to learn R and stats together, and the feedback has been really helpful to make it (hopefully) successful. I definitely get the impression, that practical learning and "doing" rather than listening to lots of lectures is the way forward. If anyone has any additional comments or ideas, all are very much appreciated. Thank you!