Theme 4: Quantitative research

The paper “The Google effect in doctoral theses” by Lav. R Varshney was published in Scientometrics, which holds an impact factor of 2,084. The paper presents a longitudinal study of how citation has been used in doctoral theses at MIT's Department of EECS, between 2004 to 2011. The theory is that since scientific literature is accessed through the internet, the Google effect (increased internal encoding for where to find information, rather than memorizing the information itself) would result in doctoral theses providing more information “by reference”. The hypothesis is that citation behaviour has been measurably affected by the introduction of new information technology.

The benefits of longitudinal studies are the possibility to determine patterns and see long-term changes. The limitations of using longitudinal studies includes the risk of using research conducted by other and the difficulty to accurately date when a technology became available. The latter is mitigated through the author just studying one department were the students have “fairly homogenous technological and social contexts”. To increase the validity of the methodology, the author argues for using doctoral theses, as they do not have length limitations and therefore are free to use as many citations as they wish.

The variables consist of 728 doctoral theses were the references were counted manually. Human errors when counting could affect the research validity in a negative manner. The numbers were then applied to the Kolmogorov-Smirnov test and trend test which showed that the number of references has increased over the years. All years between 2004 and 2011 where studied, and the results were statistically confirmed using Turkey et al's trend test. &nbsp The author looks at two alternative explanations for the increase in number of references besides the Google effect; longer times to get an academic degree or greater efficiency in referencing. Due to privacy reasons the author could not obtain information about the degree time for the theses that were analysed, which forced him to look at the change in time to graduation on a national level. This showed that the median time had decreased. The author also examines if any behaviour shifts has occurred in the sub-disciplines which could impact the findings by looking at the papers style as well as the faculty turnover. By using this approach, the author aims at controlling the effect of other values to ensure that they are not resulting in interpretations of causality. However, there might still be untested variables that impact the results, which is considered a limitation.

The paper ends with the author using a textual analysis method to determine whether the increased referencing is due to an increase in efficiency in production due to technology or if its due to a cognitive change with increased transactive memory use. Ten doctoral theses were randomly selected, and the age of the references were analysed. The result indicates a transactive nature of citation, which suggest that the Google effect have resulted in cognitive changes in doctoral students. Reading this paper I hade learned the importance of looking at different factors that could affect the result, and disqualify that they lack impact.

The paper ”Drumming in Immersive Virtual Reality: The Body Shapes theWay We Play” by Bergström et al. examines how the body ownership illusion works in an immersive virtual reality. When reading the text, I increased my understanding of how meticulous one has to be when conducting research. I was especially surprised with how much detail was put into describing the materials and methods used. I perceive this as a way to increase the reliability of the research as others can recreate the experiment. When reading the paper, I feel that the writers could have emphasised more on why the participants experienced that one of the bodies was perceived as more appropriated for the drumming task. Portraying the two avatars in a way that could cement prejudice is problematic in my opinion, and it would have strengthened the research if the authors would have gone more into depth on the topic of stereotyping based on ethnicity. However, when looking at table one from the article “The nature of of theory of information systems” by Gregor, that illustrates some differing views of theory in information systems, I understand that the authors' aim is not to explain the world, but rather to explain a relationship among constructs.

In the paper, the authors were able to look at the relationship between the constructs of body ownership illusion and the immersive virtual reality. Using a quantitative method could be considerable more objective, since bias from the researcher is reduced. The experiment was highly controlled and due to this, the authors could find a cause (the different avatars) and an effect (differences in how they performed the task). The problem was complex, but the authors managed to find a way to investigate it using defined variables. The result clearly showed precise, statistical significant numerical data. One limitation is that the sample only consisted of Caucasian people – which means that the research result only applies to a Caucasian population and can not be generalized to the population as a whole. When reading about the study, it comes to mind that conducting an experiment like this must be very time-consuming and that it is risky to use that many instruments, as something could go wrong with the collection of data. Once the study begins, the instruments can not be changes, which shows the importance of testing the equipment beforehand. 

The qualitative methods used in the paper is the short semi-structured interview. The benefit is the possibility to get more in depth and detailed on the subject as well as the possibility to direct the interview based on which topics that arise. When the participants are encouraged to expand their responses, new topics of interest can emerge. One limitation could be the difficulty to make comparisons based on the material.