Theme #4: Quantitative research

My selected paper

As for the last seminar I chose the Journal of Computer-Mediated Communication (Impact factor of 1.778) since I found it to highly appropriate for Media Technology with it's in-depth analysis of modern communication mediums like social media and the likes. For this theme though I wanted a more quantitative research paper since the last one about the use of internet and social media in homeless youth doesn't seem to fit that nice. However, I found (in the same journal) a paper about Channeling Science Information Seekers' Attention? A Content Analysis of Top-Ranked vs. Lower-Ranked Sites in Google.

In this research, the authors wanted to analyze the search results in google when searching for "nanotechnology" and if these top results vs. the lower-ranking results differed in terms of "theme". In this paper they concluded that top-ranked results tended to show the technical-, environmental-, and risk-related aspects of nanotechnology whereas lower-ranked results feature totally different themes on the same subject.

Which quantitative method or methods are used in the paper? Which are the benefits and limitations of using these methods?

By sampling search results (links to web pages)  for a total of 60 weeks, ultimately resulting in 9,120 parent links and 224,987 child links. The methods used to quantify their data was a self-developed program that basically gathered the URL, the domain adress, date of retrieval which automatically entered into an Excel spreadsheet. In the spreadsheet the results was sorted according to the sequence of appearance in the search results on Google. From this point, their program then mined the first ten results and further analysed the actual site and captured its profile (theme) and the contents of all the hyperlinks that were associated with that specific page (the child links).

In the end, the content from the body tags of a page was extracted of keywords based on 10 categories covering the different aspects of nanotechnology. This way, it was easy to determine the theme of a specific page.

The benefits of using this type of analysis is that the results is quite accurate. You can take a bunch of sites and perform a keyword-aided search on each of these websites content and quite easily get the most prominent theme (based on predefined categories). Therefore this method yields good results on these types of analyses.

One of the limitations of this method is that the categories was predefined, based on assumptions of what themes that are prevalent in this field. For a more generic application, the categories can't be "hardcoded".

What did you learn about quantitative methods from reading the paper?

In this specific paper I learned about one of many ways of analysis big amounts of data and how to do this quite effectively. The method used in this research, I think, is highly specific for this application. In another setting where the results is not from a specific field of research, maybe the data that is to be analyzed is harder to predict, ultimately making it hard to use and draw conclusions from.

Which are the main methodological problems of the study? How could the use of the quantitative method or methods have been improved?

One of the main problems in this study is the actual implementation of the computer-assisted sampling. The authors clearly stated that, although the sampling worked well in the end, they had a lot of bugs that they had to fix during the data gathering-process. One of the bugs was so severe that they actually lost 2 whole months of data. That is really bad. For this method to be preferable (even if I really don't see an alternative to computer-assisted sampling with this massive amount of links to scan) it has to well implemented and tested before an actual study can take place.

They also stated that their software would need to be independent of which search engine is being used since much more conclusions can be drawn if compared to other search engines result of the same topic is to be done.

Paper by Olle Bälter and colleagues

This paper is about how to potentially reduce Upper Respiratory Tract Infection (URTI). The research is based on the fact that this is case for the main part of why people seek primary care in many countries, and therefore, a need for this type of study is highly relevant (in opinion). The research were performed by investigating the relationship between physical activity level, perceived stress and cases of self-reported URTI. 1509 people (men and women) participated in this study, a web questionnaire was also carried out.

Which are the benefits and limitations of using quantitative methods?

In my opinion, the main benefit of using this type of method (population-based) you can quite easily get a good sample of the population without having to assess "that many" people in the study. In this research a total of 1509 people were participating, aged 20-60 yr (men and women). The limitations of using a population-based study is that it may not be appropriate for certain types of research where the population mean is actually skewed compared to the mean of a specific group of people (maybe more relevant to the research).Which are the benefits and limitations of using qualitative methods?

By using a qualitative research it is easier to get a more in-depth look and the spectrum of data that can be gathered is much more diverse. The limitations of a qualitative method is that it is hard to do with a lot of data/participants since it would be too time consuming and/or pricy. In many cases, practically impossible if you would like to do 1hr interviews with each participant.

References:

1. Fondell, E., Lagerros, Y. T., Sundberg, C. J., Lekander, M., Bälter, O., Rothman, K., & Bälter, K. (2010). Physical activity, stress, and self-reported upper respiratory tract infection. Med Sci Sports Exerc, 43(2), 272-279.
2. Li, N., Anderson, A. A., Brossard, D. and Scheufele, D. A. (2013), Channeling Science Information Seekers' Attention? A Content Analysis of Top-Ranked vs. Lower-Ranked Sites in Google. Journal of Computer-Mediated Communication. doi: 10.1111/jcc4.12043