2

Some background, for my bachelor of science degree I have to do a practical research (thesis) for a company that develops software.

At the end of my Introduction chapter I arived at the following central question:

Which testing strategy should company X implement to decrease the time and cost that goes into testing software Y?

After finishing my introduction chapter, I'm stuck with the literature review/theorthical framework. In my mind I already have the practical solution to the problem and find it easy to write the steps it would take to solve this problem but I'm not sure in what direction to approach this theortically and what type of literature I would need to use.

-update-

Perhaps I'm focussing to much on the testing strategy part of the central question? Should I rather focus on decreasing time and cost that goes into testing? And write down the possible solutions to accomplish this?

So for example a theoritical perspective could be:

  • Test automation?
  • 3
    This is a good question for a mentor or advisor. Generally, even if you know what approach you'd take, I'd think of it argumentatively - for skeptics who might wonder why you didn't employ strategy Y or strategy Z, how can you point out some limitations in those approaches and some benefits to your own approach? Finding the place to start researching those strategies in your field may require some content knowledge - and that's where your advisor/mentor/supervisor comes in. – TaliesinMerlin Nov 13 '19 at 15:32
2

In general, the literature review means discussing the ways that other people have tried to solve similar problems in the past, and then giving some comparison/contrast between what has been done previously and what you are about to. i.e. put your work in context of existing work. You might see it worded something like this:

"Historically, two main strategies have been used by companies to reduce testing time and cost. Strategy 1 was first described in Article #1 and consists of using trained monkeys to do the testing. Strategy 2 is discussed in several articles including article #2 and article #3 and involves just assuming everything works fine and going out to lunch early without doing any actual testing. Strategy 1 works well, but does not scale to large projects due to a shortage of properly trained monkeys, as described in book #4. Strategy 2 scales very well, but can often miss subtle errors, as mentioned by author #5. In this thesis, we will present a new strategy, which is more scalable than strategy 1, but does not suffer from the quality problems of strategy 2."

Depending on the length of your thesis, you might go into a lot of detail here. E.g. for a Ph. D. disseration you might have a couple of pages on strategy #1, a couple of pages on strategy #2, etc. But for a short article a sentence or two is usually enough.

2

Just go look for some textbooks or general review papers on software testing and/or efficiency of software testing. The idea of the Intro is to give a little bit of the big picture. Of course your problem will be it's own tiny brick in the wall. But it's helpful to have some general idea of the field. '

Since this is a bachelor's thesis, you don't need to go crazy. But do a basic Google, Google Scholar, and library search. (And when you find something skim it and check its references. Hopefully this leads to a handful of relevant review articles and/or textbooks.

By the way, I am far from a computer person, but if I were reading about your particular situation, some general "lay of the land" would help me to get into it. Presumably, there are schools of thought on software testing methods. And there are a few conflicting paramaters to optimize (at a minimum time/money/efficacy). Perhaps also different types of software (or even business needs) drive the use of particular testing strategies.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.