Archive for April, 2012

Theses and Dissertations About Software Testing

Tuesday, April 17th, 2012

This week, Florida Tech’s Center for Software Testing Education & Research (my lab) published a bibliography of dissertations and theses focused on software testing (http://www.zotero.org/groups/cster_dissertations). To use the bibliography, you need an open source (free) tool called Zotero (http://www.zotero.org/)

What are these documents?

Theses are research reports written by graduate students as a final requirement for graduation. Doctoral theses are also called dissertations.

The typical thesis includes:

  • a literature review that describes a significant set of related research published by others
  • an idea (for example, an idea about how to improve testing, or how to create and assess a testing tool, or how to study how testing is really done, or how to teach it more effectively, or how to prove that some other idea about testing is wrong)
  • a description of the thesis methodology and technology (for example, how a test tool was designed, implemented, and studied)
  • a description of the results of the study.

Theses are evaluated by professors who are experts in the discipline and at least one who is not. For example, the supervisory committee for a doctoral student in Computer Science might include three professors of Computer Science, one professor of Biology and one of Business Administration. Of the three computer scientists, one (or two) would typically be expert in the subject matter of the thesis (e.g. software testing) and the others would probably be experts in other areas of computing that the work depends on. for example, a dissertation focused on testing databases might be supervised by an expert in testing, an expert in databases and an expert in research design.

Why I recommend them

Theses are designed to be read by someone who is not an expert in the field. Therefore, a thesis will typically organize a testing problem–including the most relevant research papers–in a way that a student or a mid-level testing practitioner can understand.

Of course, theses vary in quality. Some are written poorly. Some are researched poorly. Many present half-baked ideas (this is student work, not the work of an experienced practitioner or a professional researcher). But overall, I have found them good starting points when I start working in a new area or when I assign a student to an area that is new to her.

What’s in the bibliography

We have over 700 references, most before 2007.

Each reference includes the basic bibliographic information (author, title, etc.). It also includes:

  • a URL.
    • If we found a free copy of the thesis online, we point to that.
    • If not, then if the thesis is listed in WorldCat, we point to that. WorldCat indexes many of the world’s public libraries. If your public or university library is on the Interlibrary Loan system, WorldCat will tell your reference librarian what library has a copy of the thesis, so you can borrow it. Interlibrary Loans are often free to the borrower. It’s not as convenient as free-on-the-web, but it’s still free.
    • If it’s not listed on WorldCat, we point to ProQuest (we often point to ProQuest in the notes as well). You might know this branch of ProQuest as University Microfilms. You can order dissertations from ProQuest but this is not always free (PQDT Open and ProQuest with Google Scholar publish some dissertations for free, possibly including some that we thought were available only commercially). Prices vary. I think $39 is a typical number. Because theses are of variable quality, I strongly suggest that you preview as much as you can (you can often download a chapter for free from ProQuest) or read an article that summarizes the thesis (see next section) before paying $39.
  • a related reference
  • an abstract (a short summary of the thesis)
    • We chose not to copy abstracts from Dissertations Abstracts (ProQuest) because we don’t want to risk a copyright fight. If we found a copy of a thesis online or if an author posted a copy of their thesis abstract online, we copied that abstract into the bibliographic record for the thesis.

Digging up this extra information takes a lot of time and painstaking work. We’re continuing to add more recent work, and expect to grow the collection significantly over the summer.

Acknowledgments

This bibliography was created primarily by Karishma Bhatia, Casey Doran, Pat McGee, Kasey Powers, Andy Tinkham, and Patricia Terol Tolsa.

This bibliography is a product of research that was supported by NSF Grants EIA-0113539 ITR/SY+PE: “Improving the Education of Software Testers” and CCLI-0717613 “Adaptation & Implementation of an Activity-Based Online or Hybrid Course in Software Testing.” Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.