Course Comments

Course comments by Nicklas Holmgren

The following is an ordered list of all my comments throughout the course!

Theme 1

Nicklas Holmgren  2 november 2012 10:06 (In response, on my own blog)

I would say that it is of high quality from what I read of it. I looked at a few articles before selecting the one above - they all appeared to be high quality from what little I read of them.

Also the organizations seems well respected. As I mentioned I was familiar with the conference Siggraph since quite a few years, it is a pretty major event and well known in certain businesses. So while I don't know for a matter of fact that it deserves an IF of 3.5, the circumstances suggest it is at least produced by a well established organization.

Theme 2

Nicklas Holmgren6 november 2012 06:43 (In response, on my own blog)

I wouldn't say that it has in my case, though like in the example I have noticed the stories in the news about security concerns on Facebook from time to time.

Though I have noticed a few of my friends seem to have been affected regarding what and how they post, only because they have mentioned it to me (even specifying the news stories about privacy as sources for why they said they had changed habits).

Nicklas Holmgren9 november 2012 09:59 (On sepehramoorpour.blogspot.se)

I agree that combining the different Types of Theory is the more thorough way to do it. Thought focusing on just one part may be most relevant and less time consuming depending on the aim of a paper.

Nicklas Holmgren9 november 2012 10:05 (On DM2572.blogspot.se)

That sort of ties into the discussion that was going on at the seminar of group B this week. I remember our min-group originally raising a similar point to what Sofie did.

While you may not need theory in a scientific paper, I guess that depends on what its aim is. If the purpose is just to describe observations, or to offer a way to explain them.

The great thing is that with a report of that sort being published it opens up the possibilities of a new researcher or group maybe being able to come up with a theory about it.

Nicklas Holmgren8 november 2012 02:38 (In response, on my own blog)I believe the norms are mostly influenced by and set among closer knit groups, but should in broad strokes be pretty common across the region you are in. As far as I understand it.It may be so that if you have acquaintances that are further disconnected geographically, like if you have a few people from the US on your friends list you may more easily spot some things that indicate their norms differ somewhat.

Theme 3

Lost in space because I forgot to save the links.

Theme 4

Nicklas Holmgren30 november 2012 08:19 (In response, on my own blog)

Having conducted a qualitative survey myself, I know you are very correct. It does take an absurd amount of time depending on how many people you have that are contributing. :)

Theme 5

Nicklas Holmgren30 november 2012 08:13 (In response, on my own blog)

They provided a diagram so it was easy to see in the paper. Anyway, to make it more clear they had a large projector screen at one end of the room to where the students' desks were facing and the screen was divided into 36 grid-squares where each student had their own working space. Strictly speaking it was as if every student had their own screen - but the hardware demand was only one projector and one screen.

Nicklas Holmgren30 november 2012 08:17 (In response, on my own blog)

Good observation! They did not have a baseline where they compared their results to a second class at each location that did not take part in the experiment. Rather they only compared the results internally within each class that used the system. So they could see that the results of the classes improved with each session they had with the system, showing progress. But they could not compare to a status-quo that went on without using the prototype. That may indeed be a bit lacking, and could possibly be the focus of a follow-up experiment.

Nicklas Holmgren30 november 2012 08:21 (In response, on my own blog)

In this case no since they seemed to get all of the information they were after. From the descriptions it seems that their prototype was also very complete in its functionality.

In the text I am speaking more generally, that it may become a problem.

Nicklas Holmgren30 november 2012 08:23 (In response, on my own blog)

They did not use a reference study of the type you mention, which indeed may be a problem. I did not really consider it at the time I was reading the article, but now in hindsight that seems to be a misstep. I do think they could have learned a lot of useful things from that, regarding the validity of their idea.

Theme 5 - After

Reflections on Theme 5

This final theme was about Design Research, a third kind of research next to the more traditional Quantitative and Qualitative kinds.

Out of the kinds and methods we’ve gone through this is pretty unique in that it adds an earlier step. The basis is that the researchers have a new idea or a new product of some kind, and the research is based around this artifact, whatever it may be.

Commonly the artifact is put through hands on practical testing throughout the design research process, as made evident by both the paper that everyone read as well as the one I chose myself. Both are examples of a new idea being presented in different test environments, where the objective is to employ testing to tell whether the artifact is good and improvable in order to see real production.

I was abruptly unable to attend either the seminar and lecture this week, but I have gotten to hear some summations about what went on.

The seminar was similar in shape to the two previous ones, where there were miniature-workshops treating the weeks subject. The class was divided into groups that talked through processes of design research. Material was put together during these sessions and were put into the course wiki.

The guest lecture of course also was about design research. I learned about it after the fact but through the descriptions I got I still feel that I may have understood that the main point of the lecture was about identifying the correct problem out of many - and solving it specifically.

Theme 5 - Before

Turn Your Mobile Into the Ball: Rendering Live Football Game Using Vibration

Réhman, S. et al. (2008). IEEE Transactions on Multimedia.

The paper is a design research where the subject is using a tactile interface to communicate with an end user. The information being communicated is a live feed from a football game. They used a prototype for the evaluation - by connecting an external vibration device to a mobile phone.

1. If you are evaluating an idea of any kind that is not directly comparable to something that exists you may need to build a prototype of it.  In this case the individual parts of the vibrator-technology they research exist but not in the concept they wished to evaluate.

4. Prototypes are inherently quite different from finished products, though they may be at various stages of completion when used. If we look at the prototype used in this example it had all the functions that were relevant to the end-user - so its usage can be said to have been representative of how the finished product would have behaved. I would say that the limitations depend on how rudimentary the prototype is. I think it’s important that the prototype be usable enough to function as a substitute for the finished product, at least regarding the parts of it that are being evaluated. Another limit may be regarding at what stage you use the prototype. Other forms of testing early on may give you information about your artifact that you could take into consideration before building your prototype, thus allowing you to use a more complete and adapted prototype in a later stage of evaluation. Without having to spend time and effort assembling prototypes of different iterations if early results are unpredicted and send you on a different path than you foresaw.

One Mouse per Child: interpersonal computer for individual arithmetic practice

C. Alcoholado et al. Journal of Computer Assited Learning, issue 4 August 2012.

This paper explores a sort of new technology that is mainly software based but hardware reliant. They posit a system to allow multiple students to use the same computer in a classroom environment, by providing them all a mouse to interact with it (through an interface displayed on a common projector screen).

The idea is to use it for mathematical exercise, where normally the school would be limited to students taking turns on one computer - with this system they had up to 36 students doing their exercises at the same time on one computer. This would greatly benefit poorer areas that are unable to normally facilitate enough computers for their students, because the material cost would be very low (estimated at $1 per student/year).

For the data collection they used Qualitative observation and Quantitative analysis. The quantitative part was mainly done by taking tests at various points to show if the students improved in arithmetic by using the system. The results improved a massive amount between the first and second sessions with the system - which suggested that the technology threshold was quickly reached and that the students quickly became accustomed to using the system.

From reading both of these articles I think I reached a basic understanding of how design research is conducted, and I hope to learn more during the following week. Common among these two papers is the implementation of a prototype and its use in testing. I can foresee issues arising if you are not careful with when you implement a prototype, since early data sampling with other methods may give you useful information to consider before constructing your prototype. It therefore seems reasonable to me that design research methods would be most beneficial when used in combination with other methods.

Theme 4 - After

Reflections on Theme 4

This week we discussed qualitative methods. Preparations were as usual to read a bit; a paper by the guest lecturer of the week as well as to find a research paper employing some kind of qualitative method of research.

I have had some trouble finding interesting articles in the past, but this week I found a fitting one pretty quickly. The core of the paper was about finding out what motivates people to start independent podcasting. As it deals with motivation - a very personal aspect - it did focus on qualitative methods.

The authors of my paper used a qualitative survey, which I thought was a pretty common method. They are conducted by using open ended questions so that the persons partaking can provide their own thoughts without having to resort to predetermined options. At the seminar it turned out that I was the only one to have found a paper using a survey of any kind, which surprised me.

I thought that the seminar was the place where I learned the most this week. It was really straightforward; an opportunity to familiarize yourself with a great deal of different qualitative methods. We managed to have an extensive discussion in a group of four where we all presented the methods used in the research papers we had chosen. Some were familiar and others were new. The discussion was focused on finding boons and weaknesses of the methods, so we analyzed them together, and later discussed our findings in the class. The group discussions provided for a good basis of understanding of the methods, like a brainstorming session where you can delve into freely exploring them.

I would like to give credit to the methods wiki-page as well, it is a good resource to keep at hand for the future. It is also good to be able to read about what the other seminar groups discussed, since the content of each seminar can vary quite a bit.

Theme 4 - Before

Doing Radio, making friends, and having fun: Exploring the motivations of independent audio podcasters

By K. Markman. New Media & Society, June 2012.
It is an exploratory paper about the podcast medium, specifically what motivates podcasting.

1. For this paper they used a survey to collect data. A survey can be either qualitative or quantitative depending on the questions used. In this case they used mainly open ended questions. And they do mention early on that they focused on the qualitative part.

I would say that generally the benefit of a qualitative survey like this is that you can get very detailed responses. This is exemplified with the questions that were asked in this case, for example “What made you decide to start podcasting?”. What the researchers are looking for is a personal motivation, and that is something you can get much more accurately with an open ended question like this. A quantitative alternative would have been to provide multiple responses and the persons would check the option that fit most closely. A limitation is that going through and interpreting the answers take much more time than if you provide multiple answers. The researchers had to go through and scrutinize all answers and code the different motives manually.

2. I would say I learned a little by reading their pretty detailed chapter about how they themselves conducted their research. They describe the process and how they went about advertising the survey to collect responses in various steps, so in that way the early parts of the paper can act as a guide to a conduct a similar survey. Some general things I have learned are also that formulating questions is very important, the questions in this case are based on “What...?” and are not leading.

3. A thing that I thought might be a problem is that a part of advertising the survey was that the authors posted about it on Facebook groups for podcasters and public forums. In doing this in a public setting you are not guaranteed to reach your focus group since everyone can read it, and I felt that may have lead to responses from people who were not in fact part of the target group. That and making the submissions anonymous may have at least enabled such responses. Keeping it through private channels, like they did also use (direct email), would have kept it more controlled.

Comics, Robots, Fashion and Programming: Outlining the concept of actDresses

This paper was about actDresses, a way to physically communicate with “consumer robots” of different kinds. Three examples are given through how this can be used to interact with a pet robot dinosaur (Pleo), glowbots and a robotic vacuum cleaner.

The subject was new to me, and the paper was built in a way that I haven’t seen before and differed a great deal from what I am accustomed to seeing in a scientific paper. I am guessing this ties into the fifth theme - design research. The paper introduces a concept and with the three different robots provide examples for how it can be used.

It was interesting reading it, I think it is a nice concept and the reasons for examining it was made most clear to me in the example with Pleo. Since the consumers tended to interact with Pleo in a social way that you would with a pet there must be a better way to control it (and similar robots) than using a computer as mediator. Putting a pyjamas on it to make it sleep, for instance, is not immersion breaking for the person interacting with it. And more fun. :)

Question:

Is there a generally accepted way to know what constitutes low and high response rates to qualitative surveys, or does it depend completely on the information that is sought?

Theme 3 - After

Reflections on Theme 3

The main theme this week was Quantitative methods. We had a lecture, a guest from a US University, and a computer lab. There was unfortunately some computer issues when we were to have the lab, but the program in question was installed on computers we have access to and I was able to briefly check it out a few days later.

We did at least get access to a tutorial powerpoint and had a brief introduction lecture before the lab, so I felt like I did at least get some basic insight into SPSS before trying it on my own.

The guest lecture earlier in the day was really interesting, it seemed like a very fitting person to hold the lecture - given her past and present research on the subject. It was focused on Quantitative methods but also touched upon mixed methods, which tied nicely into the texts we read as preparations before the week.

Looking at SPSS was interesting from a practical point of view. It’s good to learn a bit about how to use some statistical programs that are out there for when you start working on larger research projects (like the looming Master's Thesis). So SPSS is one of those, and appears to be in a leading position on the market in regards to how many companies apparently use it. We were told though that for a lot of applications Excel would suffice to do the work. For our Bachelor’s Thesis we used Google Docs' Excel-equivalent program to compose all the quantitative information we gathered (because the surveys we sent out were made in it - it seemed like the easy solution). Using Docs for this was a very hands on approach though, and took a lot of time because it was hard to automate some steps. Using SPSS for this in the future looks like it may spare a lot of time.

In hindsight I think I have picked up a bit more theoretical knowledge about Quantitative methods this week. And although I did already have some practical experience with it I would say playing with SPSS gave me a bit more practical knowledge as well.

Theme 3 - Before

Mixed Research and Online Learning

This was an interesting read and easy to follow because it kept pretty straight to the point and concise. It was about mixed research, that is mixing qualitative and quantitative kinds of research. I thought what provoked most thought was the early parts of the paper where they briefly discussed the backgrounds of the different schools of research, and how the pragmatic mixed methods started to be explored. With it being a pretty new area they also pointed out that not a lot of researchers are trained in conducting mixed research.

The article also goes through a few steps and points to consider when developing mixed researches, and goes into a bit of detail about the different degrees of mixed research. From partially to fully-mixed. This part relates to what stage you start actually mixing and if you do it on multiple stages of the process.

This subject is something that is very practically applicable for us students who are getting closer to writing our master theses, where we will need to be able to find really good ways to conduct our investigations. At this point we might know when quantitative or qualitative methods are better to use, but I am sure that I could think of several situations where a combined method would produce the best data.

Emotional Presence, Learning and the Online Learning Environment

The article was informative, and to me it feels like an very likely thing that emotion affects learning (both in online and offline environments) so it was engaging to actually read concrete information about it.

After learning a bit about mixed research from the previous article (which I read first) it was interesting to get to go through an actual research paper using a mixed method. Though it became quite technical in the details it was at least partly familiar to me how they had used the two aspects - quantitative and qualitative sources - to conduct the research.

For our bachelors thesis me and my thesis partner used what might be classified as a partially-mixed research approach, though we did not know much about it at the time. The qualitative part of our research was based on pretty lengthy interviews, only a few, and the quantitative part was a multiple choice survey that got around 100 responses. The data from these were joined at the presentation stage of the thesis, and we built our conclusions on the combined data, but the methods did not affect each other earlier than that. With that thesis work fresh in mind it was especially interesting to read about how these authors had done theirs. 

General reflections

I think this theme and the next will probably provide a lot of practical information that will be really useful for future writing. Over the years we’ve come to know quite a lot about both qualitative and quantitative methods of various kinds through practical experience. Though I feel a bit has been missing in theoretical knowledge about them. As it is I think most of us know how to conduct, for instance, interviews and survey collection by having tried it a few time. But learning more about the subject, and ways to develop methods for this with more forethought, will no doubt lead to a better understanding on how to develop our methods to improve a great deal.

Theme 2 - After

Reflections on Theme 2

I think I learned a bit more during this theme than the previous one, partly because the texts were a bit easier to comprehend.

The seminar was the really engaging part of this week though. We started out discussing “What Is Theory” in smaller groups. It was nice to vent your thoughts on this with others, and exchange ideas. 

Our contribution to the course wiki was:

“A theory is a coherent reasoning for explaining phenomena, until it is proven invalid by another theory. When a theory is accepted by a majority of people, it is commonly regarded as a fact.”

Which was then added upon in the general discussion afterwards. For instance we were given perspective on how theory does not need to be completely invalidated when proven wrong in a specific context. It can instead become contextualize and might still be accurate in a few contexts, but inaccurate in others. A theory also doesn't have to be replaced with a new theory when it's invalidated, sometimes new data may prove it wrong but a new theory does not necessarily need to take its place at the same time. Maybe forming a new theory on the phenomenon will take a long time. There might also be totally different theories at the same time - even drawing on the same data. The example someone in the seminar gave was how there were at least three major current theories explaining how the universe is constructed.

Earlier we discussed the difference between hypothesis and theory - which tied into how the term “theory” in common talk is actually more like an hypothesis and different from the scientific definition of theory we use in this course. Also how data relates to theory - that theory needs to be built on a foundation of some kind of data or observation. Another interesting topic was where to draw the border between an idea and a theory. The example was Copernicus who formed a theory that the planets orbited the sun instead of the earth, and if this was considered a theory at the time even though few others believed it.

I think I learned a lot during this week that will be useful to me in the future when I am writing my thesis. I want to point to the second text especially for the tips it provides in identifying what is not theory, which I will save for critically analyzing what I write in the future.

Theme 2 - Before

What theory is and isn’t

Theory is to be used to present a logical flow of arguments underlying an authors hypotheses about a given subject. Many do this inadequately in their papers. Common mistakes are said to be focusing on just presenting data, diagrams and references. These by themselves do not provide theory - but used as aides may indeed help to explain or clarify theories.
The focus on writing theories should be in laying forth logical arguments for an idea or a statement. Data needs to be used moderately and most importantly relevantly in theory. Diagrams are useful to illustrate and to make your arguments clearer. References too are useful but a reader will most probably not be familiar with every work you cite in your research, you can therefore not simply refer to other articles in lieu of presenting your arguments. The main thing to answer with theory is the why.

So as a conclusion theory is an implement to explain phenomena, and to predict future events.

Paper

“Norm evolution and violation on Facebook” (C. McLaughlin, J. Vitak) from the journal “New Media & Society” (IF: 1.394) issue March 2012.
(http://nms.sagepub.com.focus.lib.kth.se/content/14/2/299.full)

The norms in the subject refers to the mostly implicit (not explicit) terms under which users engage each other on social network services, in this particular instance the focus environment has been Facebook. This involves the material and content that users post, and what happens when someone violates the existing norms in some way. For example if someone posts something inappropriate or offensive, or if their posting habits differ from what is considered “normative”. The paper is about how norms like these evolved. They make a distinction between violations of expectations (which are more likely to happen within certain sub-groups) and general norm violations. The latter are presented above, but to explain the former they are when someone steps out of line with what a user expectes. Participants of the focus groups used were said to have learned the norms by observing how others behaved, as well as knowing the norms of everyday life. The actual norms listed in the article are far too many to list here, but it can at least be interesting in knowing the conclusion about how they were formed. A theory is also presented that the media affects how the norms evolve - participants claimed to have become more concerned about privacy after this became prominently mentioned in media. This also affected the content posted. Example of a norm violation was said to be engaging in "fights" or heated discussions publicly on Facebook, a pretty common violation.

The arguments in the paper are causal but does not really provide predictions beyond what can be showed through their method to have already happened during their examination. Because of this I think that the theory used fits most well into “II. Explanation”. It can be said to be attempting to explain the subject matter in terms of why - to put it abstractly they go through events and what followed them, explaining and analysing but they are not generally aiming to predict as much.

Benefits and limitations

The authors write that the norms evolve with time so the theories in this paper may not be current in a few years. Especially with regards to something of such a volatile nature as social networks. It does provide, when performed well, the benefit of a concept of understanding certain established events. I think theory is used well in this paper because it relates to sort of an implicit rule system, which benefits from explanation.

Theme 1 - Posteriori

Research

The first seminar was about the journal/paper part of the preparation. It was interesting to me mainly because it was very free-form. We started out by gathering in smaller groups to pick the best journal and paper among those we had gathered. After a period of internal discussion we presented these to the other groups. This was a great way to find out more about the interesting material that others had found. These presentations lead to pretty long discussions on the subjects of their papers and related things, which was pretty interesting. Overall I was overwhelmed with how much information is out there in the form of research journals and papers. It was much harder to find relevant information than I’d have thought before starting this theme. In the groups we also discussed the relevance of impact factor and cites in regards to picking research papers. Someone pointed out that new material will be less cited even though the quality may be superb. Something to keep in mind when searching for material.

Epistemology

The second seminar was about the book “The Problems of Philosophy”, which we read as preparation. I was absent due to illness but have gotten acquainted with what happened at the seminars afterwards. In similar form to the first seminar, they split into smaller groups to discuss the material - then finishing off with a larger discussion. They also went through some background for the book, and why it was written. They said it’s a good book for an introduction into philosophy and epistemology, which I can agree on since it was pretty easy to read and follow (though convoluted at points but I attribute that to its age). Because of my absence I can’t say much more about the seminar itself, but I can expand a little on the source material. Overall on this theme I think the most interesting thing was getting to understand Epistemology a little bit better. I was a little familiar with, through reading independently, but until now was not aware of any practical implementations.

I understand a bit more about its importance in applications of analytical thinking, and where being aware can help in practical situations. This applies to research very much, but also businesses of law, medicine and others. It seems like a broad foundation to build experience upon, to better understand how knowledge gathering is better performed.