[This article was first published on W. Andrew Barr's Paleoecology Blog, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.
JGR Logo |
This post introduces how to use the Java Gui for R (JGR, pronounced Jaguar) along with the Deducer package (manual here) to get a fairly full featured graphical user interface for R.
Installation
Note: Be sure you are logged into an account with administrative privileges, or some stuff won’t won’t install correctly but will give only cryptic error messages.
- Make sure you have the most recent version of R (currently 2.15.0) installed.
- Install JGR. On Windows, download and run the jgr-1_62.exe file or the jgr-1_62-x64.exe file for 64-bit systems. On Mac OSX, download and run JGR-1.6-SL.dmg. If you are a Linux/Unix user you can figure this out yourself.
- Open JGR and make sure you don’t get any error messages. If you do I can only suggest the google. Make sure you are an administrator!
- Now, you can install the Deducer package from within JGR using the package manager. Find the menu item Packages & Data > Package Installer. If you are prompted to pick a “CRAN Mirror” select one that is geographically close to you. When a long list of packages appears, choose “Deducer”, and click Install. A lot of text will whiz by as R installs this package from the interwebs (a series of tubes).
- Once it quits downloading, you need to load the package into active memory. Close the package installer, and from the same Packages & Data menu, choose the Package Manager. You will see a list of available packages. Check the box next to “Loaded” to load the Deducer package.
The Data Viewer should pop-up with options for you. You are ready to graphically use this interface to R!
Notice that in addition to opening this new window, loading the Deducer package changed the available menu items for JGR. Now you have things like “Analysis” and “Plots” as options.
Loading Data
Click on “Open Data” to import your data. JGR/Deducer can open many types of files. For best results, I would try using a comma or tab delimited text file. Also, double check that each column of data only has one type of data (e.g. a number versus text). Once you get your data loaded, the pop-up window turns into something like a spreadsheet that shows you your data.
Plotting
On the menu, click “Plots > Plots Builder”. This feature alone will be worth the price of admission for a lot of people. If I continue to find this useful, there will be a future post on how this works. For now….experiment!
Analysis
So now you are ready for some analysis. Ummm…just look at all those new menu items and do whatever analysis you want. The interface is pretty similar to what I have seen of pre-packaged stats programs (like SPSS), but I am not at all fluent in the Menu driven options in either JGR/Deducer or any of those other programs. For much of what you will want to do though, you will do linear models.
Linear Models – The Workhorse of R
What do multiple regression, ANOVA, ANCOVA, MANOVA and Plain-Old Least Squares Regression have in common? They are all linear models. This simplifies your life because in R, you only have to use this one tool to do a lot of different analyses. Simply find the menu item “Analysis > Linear Model”. This will launch the below window.
Choose your outcome variable, as well as any numeric or categorical (factor) variables. R automatically knows which type of analysis to do based on the input variables. Assuming your outcome is always continous: put in a single Factor, it will do an ANOVA. Put in two numeric variables, it does multiple regression. Etc.
To leave a comment for the author, please follow the link and comment on their blog: W. Andrew Barr's Paleoecology Blog.
R-bloggers.com offers daily e-mail updates about R news and tutorials about learning R and many other topics. Click here if you're looking to post or find an R/data-science job.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.