Getting Help in SAS and R - Syntelli

Getting help was my first real discomfort working with R. Having worked with SAS for most of my career, when I had a questions, I followed the steps below:

  1. Ask a coworker.
  2. Go to the printed or online manual.
  3. Call the SAS helpdesk.

Because you pay quite a bit for SAS, the last two options are very reliable. If something has been updated or needs to be fixed, the documentation and help desk are almost always up to date. Not to mention, helping users is all these people do for a living. So very few questions you ask get a blank stare.

Support for R, on the other hand, doesn’t provide options 2 and 3. Other than third-party purchased manuals, almost all help for R is found in online forums. Truth be told, I found that these groups provide a wealth of information. And they are very responsive if you ask a question.

However, there are some best practices to be observed. Personally, I’ve found the following process to be most useful.

  1. Search the archives first. There’s a good chance someone has asked the same question before. If you ask a question that is well-worn ground, you will suffer the ire of the community.
  2. Google is your friend. I thought it would be difficult to search for answers using this general tool, because the key component to the search is a single letter of the alphabet. However, I was happy to be proven wrong. Even a search string such as “Missing observations in cov/cor” returns R-oriented results as its top choices.
  3. Ask a question in the forums. The collective skills and experience of R users are impressive and genuinely helpful. So don’t hesitate, ask your question. Just make sure it’s a new one.

Step #3 is where the best practices come into play. There are plenty of people at the ready to answer your question rather quickly. But you will only get a couple of shots before you are ignored. Here’s the best way to ask a question and get a prompt response:

  • Create a header that lists your version of R, function, operating system and the error you are getting. For example “R 3.1.1 lm() Windows 8.1 – seg fault on large data frame.”
  • Describe the goal, not the step. You may be approaching the issue from the wrong direction. Don’t try to lead your answer.
  • Be explicit about your question. Here detail is very useful. You want the reader to know exactly what you are asking.
  • Be brief and to the point. Don’t be personal and don’t list all the things that didn’t work.
  • Follow up with the solution if you find one before you are answered. This will add to the forum’s pool of knowledge and perhaps save others from having to ask your question again.
  • Append code only if needed. It may be useful to copy a block of code at the bottom of your post, but be careful. You do not want to get the dreaded ‘tldr’ (too long, didn’t read).

There is a lot of help out there for R users. Hopefully, these tips will help you find answers to your questions. As always, you can always contact us:


Carter MillsCarter Mills
Principal Consultant
About Carter:
Carter has had a passion of solving problems with data for over 20 years. He enjoys spending his days understanding all levels of an environment and designing tailored solutions which lead to measurable results. Carter’s professional experience includes developing strategic solutions in the financial, healthcare, insurance, retail, energy and telecommunication industries. His level of expertise encompasses specialties such as marketing analytics, data integration & enhancement, client management training, database modeling & segmentation, custom CRM systems, SAS, SPSS, R and SQL.