After this weeks reflections I
think it has become clear how quantitative methods can be used in various ways
in order to produce valid results.
I believe the most important
factor in the process of quantitative research is to carefully control the
conditions during the study. Since quantitative research is handling a large
amount of data, there has to be a structure that assures that all of the data
is collected within the same framework. The reason for having this kind of
structure is to make the data generalizable – without generalisation there is
no possibility to reach a result. Another problem without a controlled
structure/conditions would be that an irregularity would be perceived as a
standard and therefore cause a deceiving result.
Quantitative studies aim to get
an overview of and describe circumstances/phenomena as they are. Therefore,
they cannot take irregularities into account. Instead, quantitative research
does often lay the foundation for further studies, e.g. qualitative research
that can study object in-depth. This way it is possible to get a causal
understanding of phenomena, and also highlight abnormalities that could not be
included in the quantitative research.
With this in mind, I believe I
presented a reasonable analysis/discussion of the essay I read related to the
topic (Henning Rode’s “To share or
not to share, the effects of extrinsic and intrinsic motivations on
knowledge-sharing in enterprise and social media platforms” published in Journal of Information Technology). In hindsight I believe that Rode did manage to
target the right people for his study. I mentioned in my previous post how the
aim of the essay was to clarify why workers did not apply Enterprise Social Media Platforms (ESMP) into their daily use of
social media. However, as Rode did not control who answered the survey, the
research did not manage to reach the passive users that were the main object in
the study (and in case some of them answered, he would not know, as he did not
control who of the workers actually used the ESMP and who did not). In other
words: at it’s best, Rode’s result could at it’s best only be partially accurate.
Rode, Henning “To share or not to share, the effects of extrinsic and intrinsic motivations on knowledge-sharing in enterprise and social media platforms”, Journal of Information Technology, volume 31, June 2016, accessed 22.09.2016
Thank you for a good post. You seem to have a good understanding of quantitative research, and I found your writing about controlled structure/conditions and generalisation quite interesting. I agree with you, and think it's very important that all data is collected within the same framework for the results to be valid. If not the results would as you also say, be irregular and the research would give a deceiving result. The paper you chose i think is a good example of how important it is to really know how to obtain the right information, and how important it is to know who takes part in the research.
SvaraRaderaHi! Nice post and reflection - you thoroughly go through what you've learnt during the week, and it's clear that you have gained some insight into the method of quantitative research. When you write that results per definition have to be generalizable, I slightly disagree. What if the study is set out to only investigate a certain sample, without the amibition to generalize?
SvaraRaderaYour reflection on the environment quantitive research is performed in, that has to follow a structure and be carefully obtained because it would otherwise falsify the results is very interesting and I agree. Without proper planning, testing and execution, the data obtained during the research will not be usable.
SvaraRaderaThat being said I think in most of the cases the qualitative research comes before the quantitative. Using qualitative methods one gets an insight on a certain topic and formulates a hypothesis. After forming that, it is being tested with quantitive methods. Nevertheless it is also possible to be the other way round, just not as common.
Although I agree that it's important to have a structure and conditions, I disagree with that this should be done during the study. I think it should be done before the research, in the planning stage.
SvaraRaderaI particularly find your reflection about generalizability to be of interest. I think this article regarding scientist trying to reproduce experiments within the fields of chemistry, biology, physics and engineering, medicine, earth and environment and other research fields is really thought-provoking. 51 percent of the 1 576 researchers that partook in the study, believes that there is a significant reproducibility crisis. The main factor contributing to irreproducible research is according to this article the selective reporting from researchers. The main factor put forward as a way for boosting reproducibility is increasing the understanding of the statistics. Interesting reading for sure: http://www.nature.com/news/1-500-scientists-lift-the-lid-on-reproducibility-1.19970
SvaraRaderaAllo,
SvaraRaderaI totally agree on the fact that the study needs to be totally under control. this is a point that has been repeated over and over during the seminar, that we need to test the study. If we do a part of the study differently each time then the answer doesn't make any sense in a well made study. You're making such a clear difference between qualitative and quantitative, this was very nice to read and I couldn't agree more on that. This is the foundation of qualitative, makes things clear because one abnormality makes the total study wrong or incorrect and might be hard to go through the approving phase.
Thank for your reflection