R-Studio and statistical analysis

R-Studio is a free statistical analysis program. It is a graphical front-end to R.
Which is a good thing, because R itself runs through the command-line interface (CLI), which is like trying to do programming in the late 1970s. One overwhelming problem with command-line computing is that if you didn’t happen to know the syntax for a command, you were out of luck. The second, really irritating problem is that if you did know how to enter a command, but you make a typo, the operation would not work. Both of these are really good reasons for graphical user interfaces (GUIs) with drop-down menus that give you clear options (and preformatted commands out the back end). So thank you Xerox PARC, and thank you Steve Jobs.

However since 1995 a few things have made command-line environments a lot easier to deal with, and they still do have several advantages. First: if you don’t know how to do something, you can look it up on the internet. Second: if you find an answer on a web page, it usually includes a sample of the actual, typed-out command. You can then copy it from the web-page and paste it into your CLI terminal-window and run the command. Or better yet, you can copy the command from the web, paste it into a text file, save it for now and for future reference, and then you can copy it from your own file and paste it onto the command-line and run it. Thank you, Al Gore, for converting DARPANET into the internet.

But even if command-line computing in the 21st century is easer, why should we do it?
Generally, I do not encourage command-line computing. But there are still a few instances where it is advantageous. Sooner or later we all need to do specialized things in computing. You might want to disable the CapsLock key on your Windows system, or make it work like a typewriter (no accidental CAPS–it should only engage when you are also holding down the [SHIFT] key). For that, you need to do a little command-line work.
Likewise, if you want to do full-blown statistical analysis without paying an arm and a leg for commercial packages (that only run on Windows), R is your best option for now (2014). Furthermore, since the language is open-source, it means that people who know a little programming can add functions to R that you can use. Here is an example: on these pages I explain how to calculate the indices of Dissimilarity and Isolation in R. However, in July of 2014, Seong-Yun Hong wrote a package called seg which includes the functions for the “Five measures of segregation” explained by Massey & Denton in 1988. Since 2010, a large number of natural scientists have been adopting R, and that means there is a large user-base. And that means there are a lot of web-pages out there where people have explained solutions to problems in getting R to work, and a number of users are adding functionality.

So! With all that said, the following pages are essentially a series of tutorials on using R-Studio for the class Sociology 393: Quantitative Analysis of Social Information. If you want to follow along and use these pages, I suggest that you find and assemble some Census data, using the instructions under the adjacent column.

Scroll to Top