Theme 6: Reflection

This week was about qualitative research methods and case studies, and we had to choose two different articles, each using one of these methods.

The article I chose conducting a qualitative research method only ended up using ten people in their study. I personally think that it has to be nearly impossible to get any conclusive or reliable results from such a small study. One can hardly make any generalization with that amount of research subjects, and how valid is the research then? I’ve learnt that this method perhaps isn’t suitable for certain types of research. If they want to study how a technical device/system is being used, it’s perhaps better suited to work with a more quantitative data gathering, and then perhaps finalizing it with qualitative methods. If, for an instance, one is designing a new system, I can see the real use of qualitative research when defining how the system should be designed and perhaps evaluating the product after designing.

The article conducting a case-study I thought carried out a much more suitable research method. They examined a case, used multiple ways to gather data, conflicted with the brought up literature, and in the end – according to me – got more conclusive results than the article mentioned above. I guess I’ve come to realize that case-studies can be really useful when you want to make a new hypothesis or defining a new theory, but it’s more or less something that usually needs to be completed with further research. It’s hard to make any “definite” generalizations outside the studied case, since the population or situation is (or should be) limited to just that case in question. But, in general, I learnt a lot about case-studies, and a lot from Eisenhardt's text. I had a very vague idea about what a case-study was, and now I really see the use of them, especially when it’s concerning new research areas.

Theme 6: Qualitative and Case Study Research

Mobile Geotagging: Reexamining Our Interactions with Urban Space

The first article I chose is called Mobile Geotagging: Reexamining Our Interactions with Urban Space, written by Lee Humphreys and Tony Liaowhich and was publish 2011 in the Journal of Computer-Mediated Communication, which has an impact factor of 1.778. In this paper they focus on a mobile geotagging service called Socialight and how people interact with it in an urban environment. They do this by exploring Socialight’s ability to leave “Sticky” notes, where you tag a note of some sort of note to a specific geographical location, and later on you or another person can read it when being at that location.

• Research method

In this paper they contacted active Socialight users, and in the end got 10 active users with which they performed in-depth interviews. They asked the users question like how and why they use Socialight, what made them start using it, how they integrated it into their everyday lives, and so on. The authors also used the service themselves for two years to become familiarized with it. They used an iterative approach to the collection of data – they continuously collected and analyzed the data they obtained.

One weakness when doing it like this, which the authors pointed out as well, is that new mobile services like Socialight tend to change and sometimes switch focus, and when the study stretches over two years it might very well be something else than they started out with. Another weakness is that they could not make any generalizations, since they only had 10 subjects and those subjects couldn’t represent all the Socialight users, as well as also only Socialight users, not any other kinds of geotagging services. It is also harder to analyze this sort of data, since the interviews might differ so much, which in turn can lead to a lacking conclusion.

The tweets that killed a university: A case study investigating the use of traditional and social media in the closure of a state university

The second article had to be a case study, and I found the article The tweets that killed a university: A case study investigating the use of traditional and social media in the closure of a state university, written by Nicholas J. Kelling, Angela S. Kelling and John F. Lennon and was publish 2013 in the journal Computers in Human Behavior, which has an impact factor of 2.067. This article is a pretty straightforward case study, where they investigate the case where social media (mostly Twitter) led to the ultimate closure of a university.

• Briefly explain to a first year university student what a case study is.

A case study is, just as the name implies, a study of a certain case. The case can be a certain person, group, country, current or historical event or decision, pretty much anything that can be analyzed holistically. It can be one case or multiple, as long as they can be analyzed as a whole. Case studies are generally good in new research areas, for creating hypotheses and theories, and in some cases test and describe theories.

• Use the "Process of Building Theory from Case Study Research" (Eisenhardt, summarized in Table 1) to analyze the strengths and weaknesses of your selected paper.

The article use multiple ways to collect data - a questionnaire sent out to students asking things like their media habits and their part in the closure of the university, and they also data-mined the different media and social media sources that were active during this event. According to Eisenhardt this can strengthen grounding of theory by triangulation of evidence.

It also has a specified population. It focuses on the students attending the university and the media covering the event at the time. This limits the unnecessary variation and sharpens the external validity.

The article starts by introducing different literature concerning e.g. how students get news, social media and campus communities, and how social media is utilized to document events, all of them containing their own theories. The result of this article was however unexpected. It conflicted with the previous literature, and indicated that students was not the ones utilizing social media in this case, it was mainly the media – the journalists and the politicians. According to Eisenhardt, when comparing with conflicting literature, the theory can be strengthened because it builds internal validity, raises theoretical level of the article, and sharpens concept definitions.

A weak point of the theory this article produce is of course that it only represents a certain population, which makes generalizations weaker. It is also mainly just a quantitative gathering of data – it doesn’t really combine with qualitative methods and therefore doesn’t go very deep into why these specific results were obtained.

Theme 5: Reflection

The theme this week was design research which I think is an interesting research method. It’s more in line with the kind of research I’m interested in, where you actually create something or at least propose the creation of a product or system.

However, we did not have any seminars this week. Instead we had two lectures, one held by Ylva Ferneaus and one held by Haibo Li. Both of the lecturers were also part in the writing of the two texts we read for this week’s theme - Comics, Robots, Fashion and Programming: outlining the concept of actDresses and Turn Your Mobile Into the Ball: Rendering Live Football Game Using Vibration. Both of the texts were design oriented research, albeit not exactly the same type of design research. It was nevertheless interesting to see how design research can be approached in different ways and aimed at different outcomes.

I thought Haibo Li’s lecture brought up some interesting things that I believe that I, and perhaps many other students and researchers, often overlook. Especially the part about how we approach a problem caught my attention. Haibo Li mentioned that researchers can often be quite hasty when defining the problem, and then quickly move onto the solution. The difference between good and bad research can be the result of how you distribute the work over time and which parts that get the most focus. In general, Haibo Li thought that more time and effort should be put into the defining of the problem, which, after hearing it, sounds very logical to me. How can one solve a problem when every part of the problem isn’t yet revealed? And what if you start designing the solution, and notice that the problem is askew from the solution given by the design? I really think this is something I’ll think about when conducting research on my own, or just designing a product/system. Jumping into the designing right away could possibly make all the work ineffective or just downright not very good.

Theme 5: Design Research

This week we’re discussing design oriented research, i.e. research that proposes solutions in the form of designed systems and products. There were two articles provided for this theme, and the first one was Comics, Robots, Fashion and Programming: outlining the concept of actDresses, which is written by Ylva Fernaeus and Mattias Jacobsson. This article propose a “[…] design of physical languages for controlling and programming robotic consumer products”, and it does this by exploring some basic semiotic theories that exist in comics and the world of fashion. The design is presented in the form of three examples; one where the robotic dinosaur Pleo acts differently by wearing different clothes; one example where the smaller GlowBots are provided with pins that define their movement and behavior; and one last example where an automated vacuum is provided with comic book-styled stickers that affects its behavior. The article does suggest technology that can be used to implement this, e.g. RFID in the items that tells the product which programming code to use, but the study does not put this into action itself.

Personally I was a bit disappointed that the study didn’t include actual implementation of the technology to make working prototypes, but I still think it’s quite an interesting and creative subject. If robots are becoming more and more common in our everyday life, perhaps it would be interesting to be able to interact with the robots in a less “controlling” way. When a robotic device lacks the normal forms of interaction, like screens or buttons, perhaps this kind of interaction can provide an effective way personalize the robots. I wonder though, in the time since this article was written, have there been any attempts to really implement this? And, since I figure this concept might entail things like additional work, more programming and processing power, larger expenses and so on, would this then really be a feasible concept for the more practical robots like the vacuum?

The other article to read was Turn Your Mobile into the Ball: Rendering Live Football Game Using Vibration, written by Shafiq ur Réhman, Jiong Sun, Li Liu, and Haibo Li. While the last article was more of a suggestion and research for further research and implementation, this article takes a more hands-on approach and develops a product that is then tested and evaluated. The article is, as the title suggest, and attempt to render a live football game on your mobile phone using vibration. The main idea is the make the phone vibrate in different ways and in sync with important events that happen to the ball in the actual football field.

We were asked to reflect on a couple of questions after reading is article and I chose the following:

• What role will prototypes play in research?

Prototypes can play quite an important role in research, but it depends highly on the sort of research that is conducted. In the article Turn Your Mobile into the Ball: Rendering Live Football Game Using Vibration they made a prototype with the needed features to be able to test the concept and without said prototype the study would’ve been quite hard to perform and evaluate. In general a prototype is an early stage of a product or a system and can be used to test new designs and hypotheses. It is also generally very interesting to create a prototype when one is conducting design oriented research.

• Why could it be necessary to develop a proof of concept prototype?

In the early stages of a study it can be really useful to develop a proof-of-concept prototype, before you continue making the prototype on which the theory will be based. The proof-of-concept provides a way to test different features, before actual implementation, remove the features that doesn’t work and where new designs and/or development might be needed. It can reduce the cost and the time needed for the study and with an improved functional prototype it can improve the general results in the end.

Theme 4: Reflection

This week’s theme was Quantitative Research, which is actually a subject that I (and many of my classmates) have already brushed upon while writing for the bachelor degree and also during the course Media, Technology and Culture. For me this theme served more as a reminder of what quantitative methods are and why and to what cause they can be used. But, apart from for a reminder, this week’s theme also gave at least some deeper insight into quantitative methods, and perhaps mainly produced a sharper definition as to where the quantitative and qualitative methods differ from each other.

During Wednesday’s seminar we played “Boggle” - or at least a variation of it; instead of creating words we listed definitions of the two different research methods. I personally thought the seminar was a very constructive way of learning, and it really suited this kind of subject. Discussions were raised and we had a knowledgeable judge to end them. The only troublesome part, according to me, is that competitiveness can cause bickering and excessively prolong the discussions.

Just like last week’s theme we chose an article for this theme, with the exception that the main criteria of this chosen article was that it had make use of quantitative methods in its research. The article I ended up choosing used pretty much the same research method as the article we were supplied with. Both of them used surveys (both with a rather substantial amount of test subjects) to draw data upon and then predicted and generalized to makes some conclusions from their research. In the end the conclusion are therefore only predictions and generalizations, and they never explain why and how the conclusions really came to be, only that they do. If there’s something this course has taught me, it is that research will never be exact or definite. Perhaps one can evolve a study that’s using quantitative research methods, by the use of qualitative methods (or vice versa or even further use of the same method), and provide deeper insights and therefore further defining the theory. But, however far one evolves the theory, it will not provide a definite truth. So what then is the point of research? Well, perhaps one could accept that the definite truth may not be within human grasp, and hope that the research at least touches parts of the truth and is still beneficial.