MediaTechnologyatKTH

2015 m. spalio 10 d., šeštadienis

Comments on Theme 4

Reflection: what I learnt during week 41

Before starting week 41, I did not have any knowledge about designing research. I have never implied this research type. Therefore, this week required additional efforts to understand what a designing research is. I would summarize the main points what I have learnt about designing research.

Lesson 1: Define the problem

To understand the designing research, we started from ideas and problems. Ideas are tightly connected with problems. In order to conduct a solid research, the definitions of the problems have to be made accurately. Clarification of problem brings possible solutions i.e. ideas. I learnt that ideas are validated under the building of a prototype. In general, prototype could be described as a proof of concepts. Also, the prototypes itself have to be evaluated. It involves the questions what possible impact would be made, if there is any commercial, business, consumers interests. These questions are used like filters to concretize the research field.

Lesson 2: Prototype will not solve the problem

Going deeper I found some difficulties to understand what the prototype is. Prototype is a proof of concepts used to test and evaluate the idea. The main purpose of prototype is to gain knowledge. As it was said during the seminar, only using of prototype will not lead to knowledge directly. Prototype only provokes the research but will not give the answer. Problem and answers depend on the context. Therefore, it is important to articulate the context in which the problem is occurred. For instance, Anders Lundström started Differentiated Driving Range research conducting discourse analysis in order to understand the attitude and obstacles of current situation and formulate the intentions of research. The context is important to shape the prototype, to narrow down the research field and reach research goals, find the answers. As it was said during the seminar designing starts from the arising of problem until the realization.

Lesson 3: Designing is an artificial scenario

Many years ago the researches were based on empirical observations but more and more difficult questions were risen and scholars began to create artificial conditions to collect data and get the answers. Designing is also not an exception. Designing research is a biased scenario to create alternative reality to collect data. I wrote down: “Designing reconfigures unpreferred reality to preferred one.” Designing is a systematically framing what you want to do. Designing creates the condition to collect data but itself will not bring the answer. The method will not make a research, method is a way to look at data, but knowledge are gained through analysis, asking yourself what is under empirical data. As it was said during the lecture, research through design contains more than designing research.

Lesson 4: Designing research and experiment

Before the Friday’s lecture, I thought that designing research is based on experiment. That is the reason why the prototype is created and implied. But this my presumption was denied. For example, Anders Lundström (Differentiated Driving Range) did not run experiment (according to scientific common definition of experiment) because the scholar did not formulate hypothesis, did not define variables and measures. Nevertheless, A. Lundström did practical designing test which can be called as designing research trying but it is not the same as an experiment despite the fact that it brought outcomes.

2015 m. spalio 8 d., ketvirtadienis

Theme 6: Qualitative and Case Study Research

Part 1: Qualitative Research

I analyzed article Doing Radio, Making Friends, and Having Fun: Exploring the Motivations of Independent Audio Podcasters by K. Markamn (New Media & Society, 2012 vol. 14 no. 4 547-565 (impact factor 1, 39). Research paper analyzed the profiles of podcasters: why did they choose to produce podcast, what kind of people make audio records, what kind of satisfaction is made, what are the minus of podcasting, according to podcasters.

1. In order to describe the profile of podcasters exploratory qualitative method was applied. Scholar carried out a qualitative, web-based questionnaire which contained opened and closed questions. Questionnaire was based on four research questions: who are the podcasters, why they started and still keeping podcasting, what their motives, what are the pros and cons of producing podcasts. (Markman, 2011, p. 551). The answers were categorized into 6 groups and described in terms of content, personal and interpersonal motives. Researcher in order to reach the target group implied two sampling techniques: snowball sampling and purposive sampling it means that questionnaires were announced in the main podcasting websites and groups willing to get the answers and spread the questionnaire. Nevertheless, the sampling methods did not bring statistically significant data. It was the lack of answers and therefore the collection of data was conducted the second time. It is one drawback of qualitative methods, especially when the research is based on interviews with people, there is a risk not to collect statistically significant data. Another drawback of this qualitative method is biased and narrowed exploratory since the research concentrated on one podcasters group settled in the North America. The benefit of this research could be described as novel research because podcasting as a phenomenon is barely investigated

2. I learnt about sampling techniques and its limitations. This research paper gave me an example how to code and categorize answers. It is important to note that scholar did quantitative research calculate the percentage of activity, what features of podcasting are used. Bus these data were used only as background to go deeper exploring the profile of podcasters. Analyzing this paper, I learnt how to use quantitative data to carry out qualitative research. Also, I learnt how to draw conclusions linking and comparing with literature review and in that way enhance validity of results.

3. One of methodological problem of chosen research paper might be the piloting of questionnaires. There is no information how the scholar tested and revised the questions and as it was learnt a week ago, piloting is the compulsory beginning of every research. Moreover, scholar pointed out that it is worth doing a content analysis of podcasting because now the profiles of podcasters are described from their own personal perspective and what content is produced is not taken into account while content analysis is one of qualitative research field. Finally, as the author of research paper acknowledged the sampling methods have to be adjusted in order to get valid data and avoid situations when data collecting process has to be repeated.

Part 2: Case Study

1. A case study is one of research ways based on examination of variables with particular circumstances. Case study is grounded by various data (qualitative, quantitative, questionnaires, interviews) which usually are combined in order to describe or test the phenomena, or formulate new theory (Eisenhardt, 1989, p. 534). The process of case study is similar to other researches strategies. Firstly, research question is raised, constructs are defined, but it is very important to narrow down the research field avoiding consideration and influence of existing theories “<…> because preordained theoretical perspectives or prepositions may bias and limit the findings.” As a result, solid case studies build novel theories or at least refine the current ones. Data are collected analyzing and sampling cases. “The case selection is typically a system of action rather than an individual or group of individuals. Case studies tend to be selective, focusing on one or two issues that are fundamental to understanding the system being examined.” (Tellis, 1997). It is common that interpretation of data brings new themes and adjustments which can and must be made during the research (Eisenhardt, 1989, p. 539). Case studies require interpretation of data from various perspectives but in order to not be lost it is recommended to categorize data and then find differences and similarities. Case analyzes confirm or deny the hypothesis and relationships but in order to get grounded validation, constructs should be accurately defined and measured. Finally, as in all researches conclusions have to be compared with literature that is the point why literature review is summarized in the beginning of research paper.

2. I analyzed the research paper Emergency Knowledge Management and Social Media technologies: A Case Study of the 2010 Haitian Earthquake by Daves Yates and Scott Paquette (International Journal of Information Management, 2011, Vol. 31(1), p. 6-13, impact factor: 1.55). The case study analyzed social media power controlling knowledge management during Haiti earthquake crisis. According to the authors, it is the first study that “examines the role of social media applications and technologies used by organizations responding to disasters” (Yates, Paquette, 2010, p. 2). The case analysis was based on action research. Scholars were collecting data directly observing Air Force Crisis Action Team actions.
Regarding weakness of research paper, the definitions of observing constructs are not clear. Authors clarified which organization and when was observed but the research object i.e. social media is not determined. The same could be said about research questions. Scholars did not clearly stated research questions. Usually, questions are pointed out separately one by one. In this research paper question researches are formulated as the patterns of observations (Yates, Paquette, 2010, p. 2). Case studies based on multiple data collection but the chosen research paper did not explain how exactly the observation was conducted what methodology was applied and what kind of evidences support conclusions. Talking about conclusions, authors did not make any comparison of their findings with literature and previous studies. Finally, a solid case study confirms or improve the theory but in the chosen research paper conclusions were drawn evaluating the advantages of social media technologies in Air Force Crisis Action Team, but it is hard to understand if researches formulated or revised any theory, not only summarized the observations of particular case. On the other hand, this research paper is a good example how to categorize data through particular framework. Scholars used 3T framework in order to categorize observations and look at data from different perspectives (i. e. syntactic, semantic, pragmatic).

Additional literature:

Tellis, W. Application of a Case Study Methodology (1997) The Qualitative Report, 3 (3) [On-line serial], 3(3). Available: http://www.nova.edu/ssss/QR/QR3-3/tellis2.html

Snowball Sampling In Explorable.com. Available: https://explorable.com/snowball-sampling

Purposive Sampling, In The SAGE Dictionary of Social Research Methods. Available: http://srmo.sagepub.com/view/the-sage-dictionary-of-social-research-methods/n162.xml

References:

Eisenhardt, K. M. (1989). Building Theories from Case Study Research. Academy of Management Review, 14(4), 532-550.

Yates, D., Paquette, S. (2011). Emergency Knowledge Management and Social Media technologies: A Case Study of the 2010 Haitian Earthquake International Journal of Information Management, Vol. 31(1), p. 6-13.

Markamn, K. M. (2012). Doing Radio, Making Friends, and Having Fun: Exploring the Motivations of Independent Audio Podcasters New Media & Society vol. 14 no. 4 547-565

2015 m. spalio 5 d., pirmadienis

Reflection: what I learnt during week 40

Before this week I thought that I had already known how to conduct quantitative research, but after this week I learnt that there are very important and personally never heart points which have to be kept in mind in order to make a solid quantitative research.

Designing research

It was new for me that firstly you have design and piloting questionnaires, experiment or other measurable way which will give you data. It is very important because data must be interpreted in a logical way and this argument framework is homework before doing quantitative research because only data is not a theory. It is important to run a pilot test in order to validate a research way before actually carrying out it. Piloting includes not only testing and getting answers but participants’ reflection and evaluation are even more important. I revised again my selected research paper Using social media applications for educational outcomes in college teaching: A structural equation analysis (Yingxia Cao and Paul C Hong) and noticed that scholars did not mention anything about their piloting and refining process. It is only written that questionnaires are based on qualitative literature review and interviews with teachers. Talking about questionnaires I learnt this week that is very important to formulate the questions properly. For instance, avoid leading and assumption questions and it is very important to raise the question: Do these questions bring you data? All in all, designing and refinement of quantitative research in one way is a qualitative research which is very important part of process in order to get not only data but draw logical conclusions.

Quantitative vs Qualitative

During the seminar our group mainly discussed about the definitions of quantitative and qualitative researches. It was agreed that quantitative method is based on measurement and data explanation. While qualitative research is more about the content analysis and interpretation of complex description. In the group we assumed that quantitative research will not give the answer why particular phenomena occur but this our preliminary conclusion was denied. Ilias gave a good example. Drumming in Immersive Virtual Reality quantitative experiment explained why virtual body causes behavioral changes. As it was explained in the seminar, what kind of method will be applied depends on research fields. There are some topics which require more than logic argumentation of data. Usually, these research objects belong to humanities and social sciences. For instance, if I want to analyze refugees’ language only data how many people speak in how many languages will not give any additional value. And in other way around, if I want to measure sea level changes then quantitative research method would be the main tool. As Ilias said, it is very important to ask the question if the chosen method will let you gain and interpreter interesting results. There is no point to calculate what percentage of homeless people sleep in the station (as it would be doing quantitative research), but it is much more interesting to analyze their behavioral changes and habits (qualitative research).

p.s. I was always wondering how to be sure that you collect enough data to draw statistically significant conclusions. For instance, how many participants have to take part in the experiment? During this week I got the answer. It turned out that there is a special program which will calculate and esteem significantly required number of participants.

2015 m. spalio 3 d., šeštadienis

Comments on Theme 3

Comments on Theme 3

2015 m. spalio 2 d., penktadienis

Theme 5: Design Research 2

Part 2

1. In the Finding Design Qualities in a Tangible Programming Space (Fernaeus & Tholander) and Differentiated Driving Range (Lundström) empirical data is gathered using prototypes. In Differentiated Driving Range case, range and speed are measured variables. In Finding Design Qualities in a Tangible Programming Space case, children activities, interaction with different resources, programming mode are measured to get data.

2. Practical design work is a way to get knowledge. Experiment built on a design work richer understanding of the problem and arise possible obstacles which broaden knowledge about the research field.

3. Research designing is based on experiment when the prototypes are tested in defined and very concrete circumstances. Design in general is a tool to visualize the matter but in research project that tool is implied according to certain criterions which are measured. Also, creation of research designing is based on pre-studies and analysis of current problem.

4. It is possible to replicate mentioned researches but it order to repeat the experiment the same prototype should be used. Nevertheless if the participants vary, then there is a risk to collect different data. Also, researches’ background influences the process of investigation. For instance, how they construct the design of the research is one way personal and biased choice. What data they collect also a matter of personal choice.

5. Design research is built on the creation and test of prototype. Also, design research does not draw solid conclusions only the possible solutions and bring obstacles and highlight the possible way of coping with current researching problems, and point out what should be taken into the consideration to future researches. While the goal of quantitative or qualitative analysis is to draw the conclusions, indicate causalities, and correlations, and deny or confirm the hypothesis.

Theme 5: Design Research 1

Theme 5: Design Research part 1

1. Media technologies can be evaluated according these criterions:

Users-friendly; media technologies have to be easily used. Also, it includes simple learning process how to use media technologies.
Applicable; media technologies have to be wanted to use by clients. It means that it has to provide entertainment, knowledge, information or any other desirable implication or improvement.
Recent and novel; media technologies belong to constantly shifting field. Nevertheless, media technologies have to enter the market offering new functions.

These criterions can be measured according to the number of followers, users, subscribers. In general, media technology can be evaluated on willingness to buy and use. Taking into the consideration technical and applied evaluation of new media gadget, in that case the prototype is used.

2. Prototype is created to run the test which let deepen knowledge about the research field and particular object. Prototype gives a widen perspective and help to esteem if a possible solution will solve the problem. Usually, prototype shows even the potential drawbacks of implying system and let researches improve their module. For instance, several design sketches were used to build few prototypes to test the best alternative of “guess-o-meter”. At the same time, usability issues were arisen and in that case the prototype shows what should be taken in to the consideration.

3. It is necessary to develop a proof of concept to test if the possible solution is implied and what characteristics should be taken into consideration. Also, a proof of concept prototype deepens knowledge about the problem and arise problems which might not be brought by pre-studies. Proof of the concept let investigate the feasibility of the possible solution.

4. Prototype is used conducting the experiment therefore prototype should be measurable or in other way testable gadget. Prototype should be users-friendly especially if participants use prototype during the experiment. Prototype is created on analysis of current issues therefore prototype should be based on new solutions and meet requirements to carry out an experiment. On the other hand, prototype is designed before the experiment based on the current problem analysis, but doing the experiment prototype could limit the research field due to the lack of features. For instance, prototype of driving rage did not measure the climate change factors which influence the speed and driving range. Regardless the fact that prototype could be used to test application of the object but it does not ensure that the market will react positively to a new gadget. As far as I understood, it is common that not only one prototype is used to carry out investigation, therefore creating and testing prototypes requires time and intellectual capacity which will not give the final result but point out the future investigation way.

5. Design research can be presented using defined coding system and making correlations and drawing diagrams (i.e. Turn Your Mobile Into the Ball: Rendering Live Football Game Using Vibration). Also, design research is covered with video material (i.e. Drumming in Immersive Virtual Reality) and publishing design sketches (i.e. Differentiated Driving Range experiment) even a test app could be used to present the prototype.