You have probably heard about R by now and that could be the reason why have chosen to read this blog article.  R is a programing language and environment for statistical computing and graphics.  It has gained popularity among analysts since it is open source, free, has powerful graphing functions and is easily extensible. It also has wide and active user community and free support from this community is almost always available.

One thing that new users complain more about is the steep learning curve associated with R. Various reasons are given as to why learning R is difficult relative to other statistical programming languages. Samples of the reasons are that:

  1. R packages have to be installed for particular functions to be available, which makes searching for what one needs time consuming.
  2. Functions from two different packages with different names can give similar results though always at different level of detail.
  3. When running an analysis you most of the times need to use several functions for different parts of the output of say a single regression.  That is, you cannot pick out all the options in one go and wait for comprehensive results. Instead one uses a given function to fit a model and another to summarize the results.
  4. There are different object types and a function can work with particular object type but not others.


As far as the reasons are warranted, the strategy required in learning R is not much different as that required in learning most statistical programming languages.  Some of the guidelines below might prove useful.

  1. Among the first things to know is where to get help! You are new to the environment so quick references will the best companion around you.  In R, typing ‘?’ or ‘??’  followed by the key word/function you seek help for or using the ‘help()’ function will be valuable most of the times. Manuals such as The introduction to R on the comprehensive R archive Network website are very useful for a beginner. You can also subscribe to help forums such as the stackoverflow where R users are happy to assist where possible.
  2. Learn the different types of objects you can create within R and the type of operations you can perform on each of them. Start with the basic ones such as vectors, dataframes, matrices and lists.  These object types are commonly used.
  3. Learn how to read data of different formats into R and how to write the data out in the formats you commonly use. Users can sometimes get frustrated with this process and the earlier you learn how to do this in R the better your experience might be.  Here is a quick tutorial on how to import data into R.
  4. Use the 80/20 rule! Learn 20% of the commands/functions that you will use 80% of the time! This essentially means that you don’t read about function after function in an R manual. When you form a culture of learning how to perform a task when you actually need to execute it you will find that over time you are using a select number of functions and packages frequently.  These functions form the 20%, not literally, being referred to here.
  5. Type your commands in a script file! Why is this important? You are able to reproduce your results quickly, script files act a good reference for yourself, and you can easily share with other R users and let them critic your work/coding.
  6. Keep the momentum! R is fun!


Feel free to write us with questions.

Happy learning R!