During the length of this course we have discussed several different research methods, but it still feels like we have only scraped the surface. There are still a thousand variants of the ones we know, and add to that the several methods I’m sure we haven’t even touched upon. Even such a simple thing as objectivity and truth turns out to be vastly complicated. There are arguments on whether there exists such a thing as truth, and if objectivity is ever possible or even desired. In fact, the question of objectivity is an interesting one. The first week was spent on whether we should strive to be as objective as possible (Plato), or to embrace the fact that we cannot be objective and work from there (Kant). I am of the opinion, and I believe that a lot of modern thinkers agree with me, that it is better to admit our faults in neutrality and instead spend time on analyzing how those faults affected our conclusions during research. This ties in to the choice of research method we use, since all methods have different ways for our preconceptions to influence our results.
Before you even get started on the actual research, it’s important to be sure of what you are studying. A lot of effort should be placed on defining the problems you are trying to solve or research, a part of the process often ignored or skimped on. In fact, if you define the same basic problem differently there will most likely be a world of difference in the solutions that someone would come up with.
Depending on the type of research question you want answered the method most appropriate will vary. A lot of research done in topics of social studies require a variety of qualitative research methods, for example, while many physics studies takes a quantitative approach. During this course I have discovered how many of my fellow student seem to be of the opinion that quantitative data is by its very nature objective and therefore more true and better all around. This is what is known as scientism, the belief that the empirical nature of the natural sciences means that they are always superior to all other kinds of science. Since quantitative research is more empirical in nature, it is easy to use the same logic on that to reach the conclusion that it is always better. However, that is not always the case.
So, what about qualitative methods? Well, while quantitatives are useful in, for example, measuring and analyzing things, qualitative methods are very useful in research where the results can’t be clearly measured. In the social sciences, and for that matter in a lot of research relating to our subjects such as interaction design, often you want to analyze things like people's opinion, or their preferences. These subjects can’t always be quantified and measured but must instead be reached through questions and other methods that fall under qualitative. Most of these methods, to my not all that complete knowledge, involve asking the subjects themselves to evaluate their reactions and responses to various objects and stimuli. This introduces the risk that the researcher will influence the subjects, or that their own bias will negatively affect the results. This bias can also be present in quantitative methods but there it takes other forms, such as confirmation bias, or ignoring and excusing data points that doesn’t fit the results desired.
We can from this draw the conclusion that there is no one superior research method. Well that sucks, things would be so much easier if there was. What method you should use is totally dependent on what question you are trying to answer. In turn, the questions you formulate are wholly dependant on the topic of your research.
If you are doing an exploratory study into a new area of research, a case study is often the most appropriate. Case studies allow for a more dynamic process, where you can gather data and formulate hypotheses and theory simultaneously. They are also, in my opinion, very useful for deep and focused research into a very specific case of something. Media technology research is often user centered, in fact the course we had earlier in human computer interaction was wholly focused on user centered research. When working with users you need to develop prototypes that they can test and evaluate, something which is easily done with design research where the changes done to the prototype during testing is the empirical data gathered.
So, in conclusion. Complex research questions can have very different requirements. In fact, research sometimes require a multitude of studies on the same topic to be complete. You might start with an exploratory case study, to generate enough theory that hypotheses can be constructed. After that, perhaps a small pilot study using quantitative data gathering to figure out the broader application of the initial findings. With that data from a broader group of people you can start building prototypes and evaluating them, in a design research process. After numerous iterations you can do a qualitative study on the impact your product had, or how it was received. This was just an example of how I imagine that a research process within media technology research could look, but it is an example of how the combination of various methods work together to make a fairly complete and adequate picture of the field you are investigating.
Inga kommentarer:
Skicka en kommentar