DM2572 - Teori och Metod för Medieteknik

fredag 9 oktober 2015

Theme 6 - Pre Seminar

I have chosen to analyze the research article Social Practices Around iTunes by Amy Voida, Rebecca E. Grinter and Nicholas Ducheneaut from 2006. The article was published in Volume 35 of the series Computer Supported Cooperative Work pp 57-83. The article’s main objective is to investigate the practices surrounding music sharing by looking at the Apple’s music platform iTunes. The method for this investigation is a series of semi-structured interviews with users of the music platform iTunes. The interviews focused on specific examples of social aspects of iTunes use and their attitudes around sharing music through iTunes’ music sharing functions. All the interviewed (13 individuals) were employees of the same corporation. Semi-structured interviews means that the interviewer and respondents engage in a formal interview. The interviewer has a list of questions and topics that need to be covered during the conversation, usually in a particular order. The interviewer follows the list, but is able to follow topical trajectories in the conversation that may stray from the guide when he or she feels this is appropriate. The benefits of semi-structured interviews is that questions can be prepared ahead of time which allows the interviewer to prepared. Semi-structured interviews also allows the conductor of the interview to deviate from the prepared questions based on what the interviewee says. The limitations of semi-structured interviews is that, besides taking a lot of time and effort, it requires skills in the one conducting the interviews and the study is hard for others to replicate.

In my bachelor thesis a part of our study consisted of semi-structured interviews so I already had some experience of the method before. This paper did not unfortunately give much detail on how the interviews were conducted or structured. However, it bases the entire study on these interview which I found a bit odd at first. I would have thought some sort of quantitative study was made before and led to the qualitative study. In this case there was not quantitative study but there was a thorough literary study that led to the semi-structured interviews.

A methodological problem with this particular study is their chosen interviewees. The interviewees had to share the same network in order for this study to work, which is why they choose a group of people who worked together. However, there is no mention on why exactly they were chosen for this study. If general conclusions about iTunes user are meant to be drawn maybe the fact that they probably have a lot in common might affect the outcome.

Eisenhardt has this to say about a case study in his paper Building Theories from Case Study Research. Academy of Management Review, 14(4), 532-550.

“The case study is a research strategy which focuses on understanding the dynamics present within single settings.” - Eisenhardt

A case study is a strategy that combine data collection methods such as interviews, and questionnaires. The evidence may be qualitative, quantitative or both. Case studies typically combine data collection methods such as archives, interviews, questionnaires, and observations. The evidence may be qualitative (e.g., words), quantitative (e.g numbers) or both. The actual case study is a study within a context where all the collected data is relevant to that context.

I have looked at the following paper containing a case study: Feng Gu, Gunilla Widén‐Wulff, (2011) "Scholarly communication and possible changes in the context of social media: A Finnish case study", The Electronic Library, Vol. 29 Iss: 6, pp.762 - 776

The paper investigates the influence of social media on scholarly communication. The aim is to provide an overview of researchers' use of a certain web technique (Web 2.0), and discuss a possible change of information behaviors in the context of scholarly communication.The authors writes the purpose of the paper and defines important concepts and expressions within the topic that is needed to understand the paper. It also has a literature review that presents what is know in the field, both in social media and scholarly communication. The authors select, based on the previous, the Åbo Akademi University as the subject for the study. The developed library system in Finland and the lifelong education system in universities are beneficial to the rapid development of new network technology and a new scholarly environment. The target population for the survey was university researcher and employees. The study consisted of a questionnaire The questions were divided into five groups: the basic characteristics of the participants, the information practices in social media, the application of Web 2.0 tools in research, teaching, and daily life, the attitude about the information quality and collective writing, and the expectations of library services. Based on the conclusion the authors draw, they presents how their findings could be implemented in scholarly communication and examples of future studies.

Sources:
Amy Voida, Rebecca E. Grinter and Nicholas Ducheneaut (2006) Social Practices Around iTunes Computer Supported Cooperative Work, Volume 35 pp 57-83.

Feng Gu, Gunilla Widén‐Wulff, (2011) "Scholarly communication and possible changes in the context of social media: A Finnish case study", The Electronic Library, Vol. 29 Iss: 6, pp.762 - 776

Eisenhardt, K. M. (1989). theme 6 Building Theories from Case Study Research. Academy of Management Review, 14(4), 532–550. http://doi.org/10.5465/AMR.1989.4308385

måndag 5 oktober 2015

My Comments - Theme 3

Theme 4 - Post Seminar

I have read two text for this week’s theme. Ilias Bergström and colleagues Drumming in Immersive Virtual Reality and an paper I have chosen to analyze: “Social Networking Sites: Their Users and Social Implications — A Longitudinal Study ” by Petter Bae Brandtzæg. The article was published in the Journal of Computer-Mediated Communication, volume 18, issue 4. The paper investigates the rapid adoption of social networking sites (SNSs) by looking at the social implications of their usage. I have also participated in a lecture held by Ilias himself that was mostly about his his paper about virtual reality and about the two main data collection methods.

We started the seminar by discussing within our small group about our experiences with qualitative and quantitative data before this course and our view of them. Most of us had done our bachelor thesis very recently so we had some understanding of the subject even before this theme. We could all agree that our understanding of the different methods of conducting research has since then expanded.

During the seminar we discussed and compared the two main forms of studies: qualitative studies and quantitative studies. In our group we started by listing the situations where one type of study form was better than the other. We concluded that a lot depends on the question one wants to answer and what type of answer you are after. In Ilias Bergström’s Drumming in Immersive Virtual Reality they wanted to find out how their test persons behaved and acted subconsciously. Their study could be based on both qualitative and quantitative data. What form of study to conduct must therefore be answered by what answer they want from the study. Ilias explained during the seminar that they wanted the subconscious reason why a stronger body illusion has an effect on the performance. Since the test persons wouldn’t know the subconscious reasons behind the illusion not much would be gained by explaining it through a qualitative research. However, if the answer they were after was different, a qualitative research method could have been preferred. There are different advantages and disadvantages with qualitative and quantitative research methods. With a quantitative research method it is easier to generalize and compare your findings. There are statistical and analytical tools to utilize to present your findings in an understandable way. The drawback is that you don’t take everything into account. The underlying reasons can be lost since a quantitative research doesn’t care about the before or the after. Qualitative methods can take a lot more into account (reasons behind the whatever you’re are examining or consequences) but it takes more work, is not as replicable and it’s hard to avoid being bias.

During the seminar we also discussed wicked problems - problems are so complex that can only be answer qualitative because there are too many factors in play. It would be impossible to answer a wicked problem purely qualitatively, it would be too much to try to squeeze in, in a quantitative stud. These types of problems are often seen in fields like humanistic or psychology.

fredag 2 oktober 2015

Theme 5 - Pre Lectures

I have read the paper Turn Your Mobile Into the Ball: Rendering Live Football Game Using Vibration. It’s about using the vibration technique to communicate what is happening in a live football game. I found this paper very interesting and appealing since I am very fond of football and because I have never heard of this way of communicating sports before. Imaging sitting in a lecture and knowing what is happening on a football field by the vibrations from your phone.

How can media technologies be evaluated?
Media technologies is, in one way or the other, about different forms of communications through time and/or space. The evaluation of media technology can of course differ from case to case but as far as I know media technology is often evaluated by observing how individuals without prior knowledge of the concept behave or perceive whatever is being tested. You chose these people because you want the evaluation to be from an outside perspective. It’s also important to successfully identify your target group (if you have one) and let someone from that group evaluate it. The actual evaluation often looks at user-friendliness, design or efficiency.

What role will prototypes play in research?
The reason for having a prototype is to have a mock-version of your intended product (or what ever you are designing) to evaluate its capabilities while you are still developing it. A prototype is a great way to make your research more realistic and more connected with what you are trying to develop and helps you figure out how to improve your concept.

Why could it be necessary to develop a proof of concept prototype?
A proof-of-content is evidence that the idea for a product, for example, is doable and realistic. A situation where this could be necessary is when trying to convince others that your vision of a product, innovation, etc is even possible.

What are characteristics and limitations of prototypes?
The first characteristic of a prototype that pops into my head is that a prototype is often very basic and “low-tech”. Of course this doesn’t have to be the case, a prototype can be an almost finished version of the final product. A software-prototype, like an app for instance, can appear to be a finished product but in reality it’s just a mock-version of the app. This is a great way of testing a prototype since you can come very close to a “real” version very easy. The limitations depends of course how good of a prototype you have. I would think that in the field of industrial design it’s quite hard to come close to what you imagine with a prototype.

I have also read Finding design qualities in a tangible programming space - Fernaeus & Tholander and Differentiated Driving Range - Lundström.

What is the 'empirical data' in these two papers?
The empirical data mainly comes from previous research, qualitative analyses through interviews with experts, early adopters etc and evaluations of prototypes. The quality of the prototypes is both early “low-fi” versions and highly functioning ones.

Can practical design work in itself be considered a 'knowledge contribution?
I believe that the process from an vision to a final product generates knowledge. One can never fully predict how something will behave or be received in the real world and by making the journey between theory and practice, new findings and discoveries will come along the way.

Are there any differences in design intentions within a research project, compared to design in general?
I think that design in a research project does a more thorough investigation that takes more aspects into consideration and the goal is to come to some sort of conclusion that (hopefully) will contribute to whatever field of science the project is about. Design in general focuses on making the design as good as possible based on the end-user and certain conditions.

Is research in tech domains such as these ever replicable? How may we account for aspects such as time/historical setting, skills of the designers, available tools, etc?
I think that research is never entirely replicable, especially in the tech domain like these papers. There are too many factors in play like the exact technology, and their chosen test persons. Aspects like time/historical setting, skills of the designers and available tools are things that are constantly under change.

Are there any important differences with design driven research compared to other research practices?
The main difference between design driven research to other research is that in design research there is always someone or something you are designing for. There is a clear objective with certain needs and conditions that need to be taken into account, when designing in the best possible way.

måndag 28 september 2015

Theme 3 - Post Seminar

For the third theme’s seminar I have read the following articles: Gregor, S. (2006). The Nature of Theory in Information Systems and Sutton, R. I. & Staw, B. M. (1995). What Theory is Not. I have also chosen a paper in the field of Media Technology and evaluated it critically.

The seminar for theme three was about one topic: theory. During the group discussions we agreed on that theory is about explaining why something occurs. It’s about somehow observing something and trying to grasp it, to understand it. We all felt that theory is something related to the previous themes; nominalism, enlightenment and knowledge. Theories describes the world which can lead to knowledge. We started discussing theories in relation to truth and came to the conclusion that it is dangerous to mention truth and theory in the same sentence. A theory doesn’t have to be “true” to be a theory. During the lecture we were taught that when a theory is tested and accepted by a majority of experts, it can be regarded as true. There are multiple theories or models today that we know are not entirely true but we still use them. Newton’s laws of motion is one example. They explain motion in a great way when velocities are within what we can relate to and observe in our world. But when velocities starts to reach the velocity of light, things depart and the theory can’t explain what is happening. So why do we use Newton’s laws if we know they are not entirely true? Because they are a simplification of something very complex. They help us explain the world around us. Even though we know the law is not universal with respect of velocity.

We also discussed that a theory doesn’t have to come from an observation or similar, it can also be based on another theory (which is almost always the case according to our seminar leader). In our group we talked about quantum physics and how there can be theories in that field when there is so little empiric data. Our hypothesis was that there must be theories build on entirely on other theories, which is rather interesting.

During the large discussion in the seminar we discussed what makes a theory good or bad. If you are not an expert in the field then it can be very hard to identify this, especially if the theory is “hidden” within big amounts of data and conclusion. In other words; it’s hard to define a good theory. A good logical argumentation might be one thing that makes a good theory, or that the area where the theory applies is well-defined might also be something to strive for.

My Comments - Theme 2

fredag 25 september 2015

Theme 4 - Pre Seminar

The quantitative paper i have chosen is “Social Networking Sites: Their Users and Social Implications — A Longitudinal Study ” by Petter Bae Brandtzæg. The article was published in the Journal of Computer-Mediated Communication, volume 18, issue 4. The paper investigates the rapid adoption of social networking sites (SNSs) by looking at the social implications of their usage.

Which quantitative method or methods are used in the paper? Which are the benefits and limitations of using these methods?
The quantitative method the authors use is by using an online survey to determine how nonusers and SNS users differ in different social capital dimensions. The survey was sent to representatives of the online population in Norway in the 15- to 75-year age group, gender, and education. The survey was sent out using various means (including telephone, websites, newsletters, and face-to-face surveys). The study consisted of three “survey waves”, one per year. Meaning the same respondents were to participate once per year for three consecutive years. They started out with 2000 respondents, but the number that responded to all three surveys was only 35% of the total number of people. Apparently such a big amounts of drop-outs is expected is such surveys. Since they sent the surveys to representatives of the online population in Norway’s they could check if the drop-outs belonged to a certain group or if it could be regarded as a coincidence through statistics. Their conclusion is that it most likely doesn’t affect the results. The benefits of a online survey is that it’s easy to reach out to the respondents and it can collect results pretty quick (if the people taking the survey answers as soon as they can. However there are of course limitations. One can not be entirely sure that the person claiming to answering the survey actually is. Also the survey type could be more suitable to some groups than others.

What did you learn about quantitative methods from reading the paper?
From reading this paper I have learned that online surveys can have a substantial dropout rate but that even though the dropout rate is rather big, you can still draw conclusions by doing statistical analysis and cross-referencing with other studies.

Which are the main methodological problems of the study? How could the use of the quantitative method or methods have been improved?
I think the biggest problem, which I mention earlier, is that this study is suppose to cover Norway’s total demographic, but its method (online surveys that are sent through e-mail) of achieving this suits some groups better than others. This would mean that some groups become over represented. I doubt that the dropouts are completely random and that some groups are more likely to drop out than others. One could approach different groups differently and in a way that suits that particular group in a better way.

Read the following paper written by Ilias Bergström and colleagues. Reflect on the key points and what you learnt by reading the text. Also, briefly discuss the questions below.
I found the paper by Ilias and his colleagues very interesting. I found a bit odd actually that this paper is almost entirely based on quantitative data when the objective is to investigate behaviors and how we can perceive ourselves in a Virtual Reality - an investigation I would spontaneous think would be based on qualitative data. The paper does mention that they conducted semi-structured interviews after each test, but they are not used to support their discussion or conclusion. Instead focus lays on the quantitative data, such as the movement data and the post-experiment questionnaire. I understand if the authors made this choice to have as objective and measurable data as possible, but since they are investigating behavioral change it feels like there could have been more focus on qualitative data. Otherwise there is a chance that some of what the test persons experienced was lost in the quantitative data.

Which are the benefits and limitations of using quantitative/qualitative methods?
Quantitative methods are good to use when performing research were the number correspondents needed to investigate something is rather high. The output data is more measurable and can be compared much easier. One can also use statistical methods and see if there are patterns. The negative part is that important details can be left out.

Qualitative methods are good to use when the question one hopes to answer is complex. Methods can give you a deeper understanding and lead to interesting side tracks. However, qualitative methods requires more time and funds. It’s also much easier for the one conducting the experiment/test to stay objective during qualitative methods.