Theme
4: Quantitative research
Blog
No. 1
|
||
Author
of the blog:
Jakob Florell
|
Date: 2
december 2013 17:27
|
|
Comment:
Hi!
I
highly agree about the point you made on the weakness of
quantitative methods: "there is no way for the researchers to
verify the information submitted is correct".
I'm
often skeptical to questionnaires as respondents can choose
whatever option to answer the questions and the answers might
depend heavily on the respondents current mood, location, time,..
I feel that a quantitative method usually requires to be followed
up by a qualitative method in order to ensure accuracy of the
results.
I
also agree about your statement that the questions "heavily
affects the quality and accuracy of your data". How the
questions are phrased, in what order, the type of option/s given
to answer the question.. There are so many factors that can affect
the results.
|
||
Blog
No. 2
|
||
Author
of the blog:
Kristoffer Ljung
|
Date: 2
december 2013 18:34
|
|
Comment:
Hi. Good post and chosen paper!
I
wonder though about the reason the study chose students rather
than employees? The research paper is as you stated studying the
online behaviours of employees, so it seems odd to me that they
didn't choose employees but rather students with limited work
experience. Where there any specific student related questions in
the questionnaire?
I
agree with you that the average age in an office is higher that 22
and thus it seems the study isn't very representable. On top of
that I also think that they could have put a minimum number of
years with work experience, like at least 3 years or so to
eliminate those who might have just (recently) started working or
been working too short of a period.
Nonetheless,
the topic of the paper is interesting considering the ever growing
popularity of computer-mediated communication which makes
researches like these important in order to take proper measures.
|
||
Blog
No. 3
|
||
Author
of the blog:
Zahra Alhouaidy
|
||
Comment:
Since the use of social platforms has been increasing, the authors
had pointed out that it's important to pay attention to children
and their effectiveness to use the settings. I think there's an
ongoing general concern for children's safety on the internet and
especially on social platforms where all kinds of people from
different countries, ages, etc, communicate and can reach out to
the children. So yes, they did talk about it being an issue today
which is one reason for their study, to find whether they are
aware of the settings, to what extent, etc. to protect themselves.
The authors had given some examples of recent events where young
people fall victim to different attacks, not only by adults but
also by other young people.
|
||
Blog
No. 4
|
||
Author
of the blog:
Ingrid Larsson
|
Date: 3
december 2013 17:40
|
|
Comment:
Hi! You wrote that the collection of data for the study was done
over a period of three years. But I'm wondering how it was done
exactly during the period? For example in the article we all read
by Fondell about URTI, the questionnaire was sent out by e-mail
about once a month, over a period of four month. So I'm curious on
other types of strategies for the same method and wondering how
they did it in your chosen paper.
I
think the major problem with longitudinal studies is the fact that
usually a lot of the participants get lazy or bored and drop out.
They don't last very long unfortunately. Like in your article, it
went from 2 000 to 708. I wonder if there could be a solution to
this because it's quite disappointing.
|
||
Blog
No. 5
|
||
Author
of the blog:
Gustav Boström
|
Date: 3
december 2013 22:11
|
|
Comment:
Hi. I agree with you about the disadvantage or limitation with
quantitative methods that the respondents are not encouraged to
provide honest answers or they may not even be comfortable to
answer honestly. They could pick the answer that seems expected of
them or so. We don't know the condition of the respondents when
they answered so when analyzing a questionnaire for example, we
only have the option to take for granted that the answers are
honest and represent the thoughts of the respondents. I think we
always have to take into consideration when it comes to such
methods that the results may not be accurate. At least that's the
case when the questions concern thoughts or feelings, that is
personal matters and that vary from day to day.
But
as you also wrote "This problem may occur independent of
physical or non-physical contact". Of course the same problem
may actually occur even with qualitative methods like interviews,
where the respondents may not answer truthfully.
I
guess all methods have their positives and negatives, and we just
have to try to be as objective and considerate of different
circumstances as possible when analyzing.
|
||
Blog
No. 6
|
||
Author
of the blog:
Matteo Campostrini
|
Date: 4
December 2013 00:03
|
|
Comment:
Hi.
I
agree with you that quantitative methods need the support of
qualitative methods to give a proper interpretation of the
statistical data and provide accurate causalities. As you said
very well, answers from quantitative methods are mere numbers.
Even if the majority picked a certain answer to one question, is
it really fine for us to draw the conclusion that the majority of
the people have those thoughts? Maybe sometimes, other times maybe
not. But how do we know? I think we need to complete the answers
to the quantitative method with findings from the respondents'
actual thoughts. Some respondents may have thought that two
options of a question were right and didn't know which one to
pick, there could have been misunderstandings, etc..
But
maybe it could also depend on the number of respondents. If it's
high enough, say over 10 000 or so, maybe in that case it's fine
to draw conclusions with some supporting literature/research..
what do you think?
|
||
Blog
No. 7
|
||
Author
of the blog:
Johan Storvall
|
Date: 4
december 2013 00:39
|
|
Comment:
Hey.
Interesting
choice of paper. I think many computer users multitask, that's
usually the case. You argued that they didn't explain though why
or how Facebook would induce multitasking. I agree that there
should be an explanation to his suggestion. But I think the
authors might have thought that's the case because of the reason
that people on Facebook link stuff to each other, chat, message
and comment. So kind of naturally it would induce multitasking?
Clicking on the links brings you to that website and while
messaging or commenting, you may go do something else while
waiting for a reply. I think that's the case for pretty much every
social platform and not exclusively only on Facebook? That'd be my
guess. However as you wrote, the authors would be able to support
their suggestion with qualitative data gathering, which they could
have done after the students' computer tasks.
|
||
Blog
No. 8
|
||
Author
of the blog:
Amanda Andersson Glass
|
Date: 4
december 2013 01:07
|
|
Comment:
Hello!
Your
chosen paper seems to be well done. I liked the idea that they did
both a quantitative and a qualitative method, and even included
previous qualitative research. As you wrote, it sounds like it
would boost the credibility of their theory. As I've read other
people's blog posts and their chosen papers, most of the papers
have only chosen a quantitative method and then presented results
from previous studies/literature, and then combined those to a
final conclusion.
You
mentioned that there were no diagrams of the collected data in the
study. I would have wanted the data to be visualized too to better
comprehend the relationship between different answers and see the
patterns, to complement the descriptions. Since the data would be
only numbers, it might become difficult for the reader to follow
along, like "10% said XX, while 35% said YY, which could be
the reason why 70% answered ZZ...".
|
||
Blog
No. 9
|
||
Author
of the blog:
Sara Långvik
|
Date: 4
december 2013 16:41
|
|
Comment:
Hi Sara!
It's
a very relevant study you've picked. Many people spend so much
time in front of the computer and it ought to cause some visual
harm over the long run. I worry about it sometimes myself.
What
kind of questions was it on the computer usage? Except amount of
time spent etc, since it's a medical study, where there questions
about the distance between eyes and computer screen or number of
breaks, for example? I wonder if such factors actually make any
difference. You wrote that they had a medical check up. I'm
thinking that maybe physical activities and the diet also are
factors that can be affecting. But since the paper were of "age"
and computer use wasn't as broad then, I would guess that such
questions were not thought of.
I
also want to add that you explained it well how the more
respondents you have, the more issues you may get when trying to
analyze the data. Every respondent is an individual with their own
way of thinking, etc.
|
||
Blog
No. 10
|
||
Author
of the blog:
Jenny Sillén
|
Date: 4
december 2013 17:38
|
|
Comment:
Hi Jenny!
The
used method in your chosen paper was an interesting idea that I
don't think I have thought of before, letting the students do the
observation themselves by taking note of a friend. That way it
could eliminate the risk of the observer having an influence on
how the study object behaves. With a friend that they know and are
comfortable with, they would likely be able to behave (somewhat?)
more "normally", as you wrote.
On
the other hand you mentioned that there's no guarantee that the
notes are correct. The students aren't experienced observers. I
would feel doubtful leaving it up to them. They could have
excluded an activity for example.
Observations
is a good method to use when analyzing behaviours, but I think
that as long as the respondents are aware of them being observed,
the results will be doubtful. I think observations might be best
suited for observations from a distance and "secretly",
for example how the general public behave on the public
transportation, out on the streets, in stores, etc.Though thinking
about the other different methods (interviews, questionnaires,
etc), there's never a guarantee of the accuracy of any delivered
data.
|
||
Blog
No. 11
|
||
Author
of the blog:
Jenny Sillén
|
||
Comment:
Hi Katerina,
I
would like to comment on your comment to my comment. :)
You
pointed out that a secret observation might lead to legal issues,
but I'm wondering if that's really the case? You seem to have been
thinking of applying it into the case of Jenny's research paper,
to secretly observe the students' study behaviors. In that case I
understand of course and I would want to know too (but it would be
very difficult to secretly do an observation there because you
will have to sneak around behind the students while they are
studying at home/school and look at their phones if they are on
Facebook or not.. and without getting caught by them. Might
require some ninja skills.).
But
I was thinking of more general open behaviours of the public, like
out on the streets for example and not as private. You could stand
somewhere far away with a pen and paper and take note of how many
times people choose to cross a certain street on a certain day and
a certain time, or something. The people are random, they are
anonymous and the only notes you take are numbers or how/what they
do with something or whatever the subject is. Another example of
an observation is standing by a street and counting how many red
cars pass by during one hour. I doubt such studies are illegal.
That's why I thought such cases would be best suited for an
observation where only you are aware of the study. That's my
explanation. :)
|
||
Theme
5: Design research
Blog
No. 1
|
||
Author
of the blog:
Felicia Joneby
|
Date: 13
december 2013 13:47
|
|
Comment:
Hi!
I agree with you that the scenario with Pleo was the most
interesting. I think if that concept idea could be implemented to
work in reality, it would really be able to sell in the markets.
There are many toys with sounds and various interactions, but Pleo
would be a little unique and popular I think since it functions
depending on what you do with it (dress it, etc), and not only
pressing buttons or something. I think this concept of actDresses
is great as the toys would seem more "real". :)
|
||
Blog
No. 2
|
||
Author
of the blog:
Kristoffer Ljung
|
Date: 13
december 2013 14:12
|
|
Comment:
Hi!
Prototypes are a great tool for evaluating and testing an idea, a
concept for something. As you wrote, it's a cheap way to research
a complex idea in an early stage if you don't want to invest much
time, money and effort to something that is still doubtful.
However, I agree with you that one drawback with testing with
prototypes is the conclusions from the tests. Sometimes it might
be difficult because the prototypes are not complete and fully
functional. To avoid drawing uncertain conclusions, I think
several tests have to be done to make sure.
|
||
Blog
No. 3
|
||
Author
of the blog:
Hannah Fahd
|
Date: 13
december 2013 14:46
|
|
Comment:
Hi!
Just like you, I also thought it was fascinating how they managed
to draw correlations between programming and their concept of
actDresses. When I think about it, it's quite a simple concept
that doesn't actually need much to be said about to understand, so
it was interesting to read how they connected and argued for it
with various factors like comics, fashion and semiotics. I think
this article for me has taught me ways of seeing things in
new/other perspectives, and also another way of how to write a
paper.
|
||
Blog
No. 4
|
||
Author
of the blog:
Lucy Armelin
|
Date: 13
December 2013 15:39
|
|
Comment:
Hi!
It's
interesting that you mentioned that the difficult part is trying
to get a computer act as a human. The question that I had in my
blog post was related to just that. I had thought it would have
been interesting if the authors had included a scenario with a
human robot with the concept of actDresses. Research is already
ongoing in that area, trying to make human robots act "real"
and human-like with responses, movements, etc. Imagine if it would
react to you putting a jacket on, like the example with Pleo.
Children would have a human-like friend!
|
||
Blog
No. 5
|
||
Author
of the blog:
Johannes Hörnfeldt Nordström
|
Date: 13
december 2013 15:58
|
|
Comment:
Hey.
Yes, I agree with you that many devices or toys today usually
don't have a display with feedback, but there's also usually no
feedback with physical languages, like putting on accessories or
clothes. Such feedback is instead relied on buttons most of the
time in order to cue some action from the toy. By using languages
that are more natural for us humans to bring feedback (rather than
buttons which are robotic), it would be more enjoyable to play
with the toys I think? It's easier to understand how to play with
it, rather than having to read through instruction manuals as you
mentioned.
|
||
Blog
No. 6
|
||
Author
of the blog:
Ekaterina Sakharova
|
Date: 13
December 2013 16:29
|
|
Comment:
Hi.
I think it was a good point that you made there about design
research, that it is much closer to practice than to theory. Most
of the other research methods that we discussed so far have been
heavily theory focused, but in the design research, it's much more
practical and object-focused. Of course there is theory , but it
seems the main point is coming across with your research about the
design and convincing about the advantages with your design,
concept, object, etc.
|
||
Blog
No. 7
|
||
Author
of the blog:
Aron Janarv
|
Date: 13/12/13
16:58
|
|
Comment:
Hi.
I think it's interesting how you pointed out that the world is
becoming more and more artificial and simulated, and that's why
prototypes will become more common and necessary. Do you think
that development is positive or negative? In my blog post, I
suggested that the authors of the concept actDresses could have
included a similar scenario to Pleo, that instead of a dinosaur
robot using a human robot. We will most likely be able to create
very realistic human-like robots in the future, as research in
that field is already ongoing. It seems that the concept of
actDresses was thinking of going into that direction, to make the
robots realistic to our own natural behaviors.
|
||
Blog
No. 8
|
||
Author
of the blog:
Marit Aldén
|
Date: 13
december 2013 17:20
|
|
Comment:
Hi.
The thought of having a great but yet realistic idea can make one
feel locked up and stuck, so I really agree with you that it
blocks people from getting into action and realize their ideas.
The ideas might feel "too boring", "too
unrealistic", "too similar", etc. So an advice
should be to screw such criticism and "just do it", as
Haibo Li also said, and see where that leads you! Also as Matteo
said in his comment, Facebook is an example. Although there were
similar social media platforms out there already, somehow it still
managed to appeal the masses.
Another
thing I want to say is about what Li said, regarding 90% vs 10%.
You wrote that 90% is for defining the problem while 10% is for
solving it. I don't know if it's just me misunderstanding him
then, because as I understood he actually said the opposite?
|
||
Blog
No. 9
|
||
Author
of the blog:
Alexandros Sombo
|
Date: 13
december 2013 17:47
|
|
Comment:
Hi.
I agree that prototypes are a good help in exploration and
experimenting new areas or scenarios. Especially if we have a new
idea that has not or very little been practiced before, then it's
wise to use prototypes in order to explore this new area more in
depth and gain a better understanding. Since the area would be
still doubtful, then it obviously needs a lot of testing before
producing the actual thing. Maybe if it's a well known design
area, with a lot of research and practice done before, then some
things can be taken for granted I guess. In those cases, maybe
it's not necessary using prototypes and it's just enough to refer
to the previous research.
|
||
Blog
No. 10
|
||
Author
of the blog:
Olle Bergendahl
|
Date: 13
december 2013 18:59
|
|
Comment:
Hey.
I think it's interesting how you wrote that since technology is
growing increasingly complex, it's also growing as increasingly
difficult for users to imagine a product or service. It's a good
point because as you mentioned, "consumers don't know what
they want until they see it" which is true. It's really funny
too how you cited Henry Ford about faster horses! It's a clever
thing to say because sometimes I think people get stuck only
thinking about what other people need or want, instead of focusing
on the idea. Prototypes are great to explore the eventual
behaviors of people with the thought product/service and how well
it would work, but sometimes I think there could be a risk that
the prototypes will be altered and adjusted too much based on the
opinions of the test persons. I think there is a need to also have
that in mind and take into consideration that you want to create
something new and so what people say about a prototype, which is
an unfinished product, may turn to the opposite when they finally
see the final product.
|
||
Theme 6: Qualitative and case study research
Blog
No. 1
|
||
Author
of the blog:
Zahra Alhouaidy
|
Date: 14
december 2013 16:20
|
|
Comment:
Hi Filip. Yes, I'm like you that have mostly read papers that
usually used one or two methods to answer their research
questions. So it indeed was interesting to read a paper that
combined three methods and learn how they manage to argue for them
and draw conclusions. I agree that the used methods need a purpose
and be truly relevant. However it depends on the research
questions. It doesn't mean that it's better to have multiple
methods. It doesn't always strengthen the paper more than if there
were "only" two or even one method. Obviously it would
depend on how well the methods used are conducted. I think that in
the end that's what it all comes down to; the approach taken to
the method/s.
|
||
Blog
No. 2
|
||
Author
of the blog:
Jakob Florell
|
Date: 16
december 2013 16:37
|
|
Comment:
Hi. You wrote that new user behaviors were discovered by
researchers that they hadn't thought about before during their
pilot interviews. Because of that discovery, they could adjust
their study. That sounds interesting to me as the pilot affected
their study to that degree. Without that adjustment, I wonder if
the conclusions would have been different. Was it mentioned in the
paper what kind of user behaviors were discovered or what
adjustments were made? That's something that we can learn from,
reading their motivations and reasonings.
|
||
Blog
No. 3
|
||
Author
of the blog:
Jakob Florell
|
||
Comment:
Hi Carl. You're asking about the differences between the methods
of semi-structured interviews and focus groups. I'd like to give
my answer to that. As far as I have understood it, semi-structured
interviews are usually done one-on-one, as Filip also mentioned
briefly in his comment below. While focus groups are, as the name
suggests, done in groups focusing on openly discussing a subject
(a relevant topic to the research question/s for example).
The
semi-structured interviews have a set of questions made up
beforehand to be asked to the interviewees, but they are not
locked up to those questions only. They are also open for bringing
up other relevant questions and follow-up questions of interest
during the interviews.
This
is what I think mainly separates them. The things they have in
common in my opinion would be the possibility to bring up a
discussion with both methods, but it's a little more difficult
with the interview as they would be discussing with only you and
you might unintentionally lead their answers to a certain
direction if you're too involved.
|
||
Blog
No. 4
|
||
Author
of the blog:
Ragnar Schön
|
Date: 16
december 2013 17:28
|
|
Comment:
Hi. You wrote that your paper used focus groups with 3-5
participants in each group, which I think is a good idea. As you
mentioned that they argued, it allows for deeper discussions with
the participants. I think it also gives more space for every
participant to take part in the discussion and better reflect upon
their own opinions as they are given enough time to express them.
This could lead to more genuine answers I guess.
I
also agree with your idea that the focus groups could have been
followed up with a short questionnaire to make sure that all
possible valuable opinions or thoughts of the participants are
picked up. It's also a great suggestion to do this iteratively.
That could strengthen the study even more as it would assure the
results, that the analysis of the data is agreed upon by the
participants too so the matter of subjectivity is minimized.
|
||
Blog
No. 5
|
||
Author
of the blog:
Havva Göcmenoglu
|
Date: 16
december 2013 18:13
|
|
Comment:
Hi. Your paper used a focus group with 80 middle and high school
students. But I suppose that they further divided them into
smaller groups, right?
As
you mentioned, I also think that using the method focus groups on
children or youths can be very difficult since they are at that
age where they are easily affected by another, feel embarrassed,
too shy, even uncertain of their own opinions but get led on to an
opinion by others, etc. You wrote that the authors considered this
difficulty and so tried to create a more relaxed and comfortable
setting. Was it mentioned in what way they tried doing that? It
would be interesting to know the details. We discussed the topic
briefly in my seminar group, doing research with children as the
group being studied, and I started reflecting on that. I don't
have personal experience in research with children so it awakened
my curiosity.
|
||
Blog
No. 6
|
||
Author
of the blog:
Johannes Hörnfeldt Nordström
|
Date: 16
december 2013 18:44
|
|
Comment:
Hi.
I'm
wondering whether they used participants from one particular
dating website or used participants from different websites? It
sounds like it was from one site. Did they want to specifically
analyse that website? If not and they wanted more general results
for their questions, I think they should have used users from
different sites. What do you think?
You
wrote that they asked questions about how they shaped their
identity on the site. I'm thinking whether those questions aren't
considered private. Maybe they could have asked the more sensitive
questions in a survey or something. In a paper I chose in a
previous theme, they had done interviews with youths and then let
them answer the more sensitive questions that they might not be
able to say out loud in a survey.
I
would also like to ask regarding this "snowball method".
I don't think I recognize it. What kind of method is it?
|
||
Blog
No. 7
|
||
Author
of the blog:
Christoffer Carlsson
|
Date: 16
december 2013 19:09
|
|
Comment:
Hi!
Just
as "Brainy" commented above, I was also surprised by the
time each interview lasted for, up to two hours!(!!) I think
that's just too much, isn't it? There should definitely be a time
limit or some other strategy done if there are many questions that
researchers want answered. The interviewees wouldn't be as
optimistic in their answers after maybe 20-30 minutes already.. At
least for me personally, I think I would quickly get bored and
tired. Interviews require quite a lot from you as an interviewee
(at least if you're genuine and want to answer properly). Was the
time not mentioned in the paper by the authors, if they did any
consideration in any way or why the interviews lasted so long?
|
||
Blog
No. 8
|
||
Author
of the blog:
Alexandros Sombo
|
Date: 17
december 2013 017:35
|
|
Comment:
Hey. It's interesting that they used a program for analyzing their
data from the interviews. Does that mean that their conclusions
are based on the results they got from the program? Did they give
any explanation on how the program works, in what ways it was
helpful? Through this course, I have found that there are plenty
of such programs available to help researchers. I've been
wondering about the disadvantages & advantages of using such
programs. Do you think it's really a good idea to rely on them? I
think that even if it requires a lot of effort to analyze a huge
amount of data, it's still better to do it yourself in order to
have full control of exactly how the data is analyzed, etc. Maybe
there would be other results if they analyzed the data themselves?
I don't know anything about those programs though. Maybe there is
research examining the reliability of such programs.
Also,
I think it's an interesting suggestion you had there about making
this study into a documentary! Actually I think every paper could
be made into a documentary. They would be a little more
interesting that way rather than just plain text.
|
||
Blog
No. 9
|
||
Author
of the blog:
Isabella Arningsmark
|
Date: 17
december 2013 19:19
|
|
Comment:
Hey. Your paper seems to be quite mysterious and questionable!
It's really strange if they didn't explain how they conducted
their interviews or given examples of what type of questions they
asked. Also, they did observations but didn't say what it was they
observed? Hmm, really odd! Then what kind of results did they
present from the observations, because that description must have
given some clues on what they actually observed?
I'd
also like to comment on what you wrote about semi-structured
interviews, that you usually obtain relevant info since the
participants are carefully chosen. I agree with that, and I'd like
to add that I think you're also able to have random participants
for interviews with the exception that they have some common
qualities that are required for them to be able to answer the
questions and thus get relevant info. In that sense, it's quite
similar to questionnaires where your target usually have some
specific qualities. But I understand your point, that with
interviews the participants are usually especially hand-picked.
|
||
Blog
No. 10
|
||
Author
of the blog:
Maria Strömbäck
|
Date: 17
december 2013 18:03
|
|
Comment:
Hey. Very interesting that your paper used interviews and that
they conducted them in different ways, because that was exactly
the case with my paper too. In my paper, they interviewed
face-to-face, by phone and by chat. Just like you, I also
suggested that they should have conducted them all in one way
considering the different conditions the different ways have which
might influence the results. It's interesting that you also
thought so!
Because
I was thinking whether it's a good suggestion or not, maybe it
doesn't really matter that much? We still (should) have the same
opinions even if we are interviewed in a different way. I wouldn't
have one opinion if I was interviewed by phone and another opinion
if I was interviewed by chat. I guess the difference would be not
the opinions but just the way the opinions are expressed,
explained, etc? So in the end, you might get the same data but the
data would be described differently which could affect the
results.
|
||
