For my regular friends and readers, this post will not mean that much to you. In fact, if you attempted to read it, it might bore you to tears, and too much prolonged exposure could cause death from boredom. So, please use caution when reading this post, or anything about statistical analysis. Only those who have suffered through four years of Stats in University will appreciate this post, but the warning still applies to them as well.
About three weeks ago I migrated to my latest “super” computer that I splurged on because I desperately needed an upgrade. This new unit has ten times more power than my previous machine did and the 1.5 terabyte hard drive added everything before I could only dream about as far as storage capacity goes. My last system was only 500 megabytes, so a huge step up from before.
I noticed that right away Ubuntu performed better than before with all this new horsepower under the hood. In fact, many of my favourite pieces of software actually increased by fifty percent, mainly because of the new video card and tons of RAM. Anyway… .
Last night I loaded up a program called “C,” which is the almost, unofficial software for doing statistical analysis around the world. C, is free and is open source and is far more powerful than the next big software package done by a company called SPSS. C still can still do many things that SPSS can, and C is unlimited, no licence fees and contains far more packages and add-ons than SPSS. However, C has one major limitation, it is command-line only, so that just eliminated 90 percent of all the users who wanted to do statistical analysis.
Fear not, I just found a GUI (graphic user interface) for C called “R Commander.” Although it is not up to the standard of SPSS as far as recoding, editing data sets and certain types of analysis (yet), I would say it is about 60 percent there. R Commander does take advantage of some of the rich visual statistical analysis tools that C has to offer, and it does allow the user to explore many more tools than SPSS has in its collection of tools for most types of graphics.
Also, C, though R Commander, now allows you to take the data saved in SPSS, the .sav file format, and import it into C. However, there are a couple of issues I had when I first started working with imported file form SPSS. First, I cold not recode the data. In fact, I could get to the data sets up, showing me the raw data, to recode them. This is a major limitation for now. I’m sure once I spend more time with this I will figure it out. Second, R Commander will only give me limited information about certain type of data sets I needed to know. For example it would not tell me if the data was an interval or binary type, I sort of had to guess if I were going to do a t-statistic, or do a scatterplot graph. This is somewhat critical for my type of work that I do because I just want to look at the program window, choose my sets, and press the calculate button.
But I was able to do basic statical analysis with C Commander using imported data from SPSS. It is a start anyway. But most importantly, this is all free, unlike the $1400.00 price tag that SPSS offers rookie data miners like myself.