Skip Navigation Links
Link to the last RSS article here:
Getting the Most from Your SPSS 17 Output:
Free your research: Open source and other alternatives to cut your costs and improve productivity as a graduate student
By Dr. Mike Clark, Research and Statistical Support Services Consultant
The following are steps and guidelines that will allow a graduate student to do research effectively and at no additional monetary cost. While oriented toward students here at UNT, many of these suggestions would apply to students of other, perhaps most campuses of research institutions. After one semester a student should feel comfortable enough using most of the applications mentioned to have them as a default, allowing them to save money and time in the long run. I will not get into operating systems and other applications that might have some indirect relevance to the research process.
I. Campus utilities
A. The first suggestion is to stay on campus. Your tuition and fees already pay for a vast amount of software that would cost an individual several thousand dollars to own on a single computer. Use it. Furthermore, you are likely to be much more productive on campus than at home.
A. Basic to any education is putting thought to paper (electronic or otherwise) in order to exhibit well reasoned positions born about by evidence.
i. Open Office: Open office is an open source alternative to the MS Office Suite of products and works on Windows, Mac and Linux. With Writer, the MS Word equivalent, you have pretty much the same experience as MS Word (though far less annoying in my opinion), however you can save the files as *.doc if others need it, and has a built in PDF converter. This was initially written using it.
III. Article researching
A. First of all, you do not have to be on campus to use the library. For example, with your UNT ID you can have access to the electronic library for article searches and download in exactly the same fashion. The drawback may be that speeds will be relatively slower compared to on-campus downloads, but time is saved from having to spend a trip to the library itself.
B. The library for a variety of reasons does not have electronic access to all journals, and even some that do have a moving wall (e.g. no access to anything within the past year). However you may still have access to things we don't have here on campus with web searches, e.g. Google scholar, searching the web for that article title with .pdf in the search, and visiting the author's own webpage. The latter should be something you visit for any work of primary relevance to your own.
IV. Data Collection
A. Web-based Surveys
i. RSS currently supports survey construction using Zope and Rich Herrington is the primary source of contact on that. Accounts and usage are free to students.
ii. Many seem to use SurveyMonkey, but its free version is not very useful for typical research by most accounts that I have heard. Expect to pay for flexibility.
iii. There are not a whole lot of options out there but LimeSurvey appears to be the front runner among open source alternatives and worth looking into. We at RSS are looking into it as an alternative at present.
B. Elbow grease
i. If possible, collect and enter data yourself rather than hiring an undergrad. It is often very easy to find volunteers for course credit or future recommendations, and even if not, having more control over your own research process is always a bonus.
V. Data entry
A. If you are on campus use a stat program with a good spreadsheet such as S-Plus or SPSS for data entry.
B. For more cross-program flexibility you might use Excel, and for an open source alternative to that use Open Office Calc. Like Writer, the files can be saved as Excel files and others.
VI. Data compression
A. Got big files? Use 7zip.
VII. Statistical Analysis
A. I have many suggestions, and all begin and end with the letter R. It does more than other common programs SPSS or SAS and is free. Furthermore, to use in an applied fashion doesn't take much with appropriate guidance, though you can go nuts with programming if desired. Given how long a graduate education requires one can be efficient enough by thesis time and an old pro by dissertation. However, there are of course alternatives.
VIII. Quality reports.
i. Using MS Word, while convenient on any campus, does not produce a high quality report, and anyone that's fought it over the course of a paper trying to do simple paragraph, table and figure formatting probably would prefer something else. Again, Writer (mentioned above) is viable for simple text. However, for really great looking scientific documents, one might try Latex even if they aren't expecting to do mathematical writing (e.g. equations), because it allows for fine control over the look and feel of your output without too much hassle, assuming you take a more applied approach. Lyx takes a wysiwyg approach and like Open Office, it runs on Windows, Linux and Mac. If you use R, you can use a package such as xtable (or its Rcmdr gui addon) to convert tabled output to the appropriate Latex code, and then just copy that into the Lyx document where the document can eventually be saved as pdf. For graphics you may insert an awesome R graphic that already looks exactly like you want because you had fine detail control before production. In addition, Lyx also has style templates such as that of APA. You can also use Bibtex editors to keep a running database of citations that can easily be called within any Lyx document. Furthermore, the UNT graduate school is hip to all of this, and provides sample templates, bibliography file etc.
In short you don't need a publisher to produce book quality documents.
i. Many use PowerPoint, which isn't too bad but again, unless you want to shell out the cash you'll be stuck to campus use only. Open Office Impress is an alternative there, but you'll also want to use high quality graphics from a statistical program like R (presentation software graphical capabilities are awful).
In summary, the graduate student conducting research can perform many aspects of it at low to no cost if they don't mind a little discomfort getting used to new software. From the initial literature searches, to collection, analysis and publication of data, the determined student has a plethora of choices and means to be productive on a budget.