**Mood:** a-ok

**Now Playing:** All I Have to Do is Dream (The Everly Brothers)

**Topic:** Research questions

So, went to the U500 session yesterday on research questions and design, so got me thinking about my research question, so I slapped down a few questions – they are definitely not perfect but needed to create one for the session: Well a very broad question I developed was:

“How do student learn LP using various types of software?”

This was quite broad but this is what I want to answer essentially, I do need to identify what is the types of software – so far, I have two definitions: one that goes with the box approach but this for solution software: so white-box and black-box, and then I can go to how they teach (software used in the teaching process) such as: problem-based software, example-based software or context-based (such as using problems that are related to their disciplines like agriculture, engineering, business). Well, using the box-approach to the classification of softwares, I developed a somewhat other question and this is based on Jonathan research (and somewhat what James is interested in):

What strategies do students employ when learning LP using a) black-box software and b) white-box software?

I’m not quite sure what I mean by strategy 🙂 and what I hope to get from this. My third research question is alot like the first, so just will state it because I think I will throw it out soon:

How do students learn using software that are example-based, problem-based or context-based?

I was thinking about what James was saying about groups depending on their background may ‘take to’ differing software so perhaps something like this:

How does a student background influence their learning of LP using software?

But here I’m not sure what I mean by background – do I mean their disciplinary background? Their attitudes to computer? Their mathematical ability? Their age? Their sex? Their race? Their culture? All these are background factors now got to ask myself which do I mean! Perhaps, I can go back to what I originally wanted to study such as:

How does employing various types of LP software affect the learning of LP by students?

This is what I’m mainly interested in, and this suggests to me that the learning of LP may also depend on the course, the curriculum, the discipline etc and thus depends on how the software is integrated into the course and to what purpose it is used in the course. I expect in some cases where learning LP is done mostly by hand to learn everything, then employing a software to do the solutions only, wouldn’t really have much of an effect. My problem comes here in what I consider learning of LP to be … that is how do I measure it, and what constitutes that someone has learned LP. The question to ask is does LP consists of someone being taught the formulation of the problem, the solution of the problem (and which solutions these are: the simplex, the graphical, the interior-point?) and the sensitivity analysis (is this just learning the range of the coefficients of objective variables or the range of RHS variables or finding the dual price as well?). The next hard part is deciding when students have learnt LP is that when they are able to put the numbers into the software and get the answer out? Able to interpret the answer? Understand the theory behind it? Able to produce the answer by doing it by hand? Understand the principle? I think I have to soon (but not as yet since I don’t want to limit my thinking)to make strong barriers and definitions for my research, for example, looking at courses teaching the graphical method only or the simplex algorithm only. Or looking at courses where the examination is by hand or something like that – got to think and decide about it. You may have note that my research question has nothing about disciplines in it … and I do want to compare disciplines so somehow got to tuck that in – or let it be some parameter for comparison in my study.

Updated: Wednesday, 16 November 2005 2:07 PM GMT