Theme 6: Qualitative and case study research

Paper I

The first paper I have chosen, including qualitative methods is called Usability of geographic information - Factors identified from qualitative analysis of task-focused user interviews (Harding, 2013) and was published in Applied Ergonomics –Human Factors in Technology and Society, which has an impact factor of 1.728 (2013).

This paper introduces a research approach based on semi-structured interviews with people who work with geographic information on a day-to-day basis. By interviewing people in different fields they were able to find nine key categories that determine important factors when designing a geographic information system.

They find that there is an existing problem in the way current geographic information systems are created and that the usability could be improved. I liked the fact that they thoroughly went trough the interviewees and evaluated aspects of the users work, the organizations objectives and in particular noticed the geographic context in which the user was situated. Sources included web pages and documents published by the organization. I feel that by having such a good pre-study on the participants, the rest of the research are built upon solid grounds and it also makes it easier to draw conclusions of the answers (something that sometimes is difficult when analyzing interview results).

Another good thing with this study was that they took the time to do a pilot study with volunteers to establish the effectiveness of the interviews later on, so that they would be in the timeframe of 1-2 hours. By doing so, possible errors could be corrected before the actual interviews, thus saving time and making the interviews more effective.

The semi-structured interviews was carried out in order to profile characteristics of the user group, their tasks and the environment in which the task is carried out, in order to determine how geographic information could help the user’s task. One of the strengths with using semi-structured interviews is that new ideas could be introduced by the interviewee, thereby making problems previously unnoticed to come forward (Wikipedia, 2013b). One example they gave was the problem with local names on places and the fact that they sometimes weren’t the same as the official names.

They interviewed 55 participants and when conducting semi-structured interviews, you should conduct as many as you need to get the information sought (Cohen, Manion & Morrison, 2000:278). Disadvantages with semi-structured interviews would be that it could be difficult to analyze and compare answers and that the flexibility could affect the reliability of the data (something I remember from writing the bachelor thesis).

They briefly introduce the Theme Based Content Analysis (TBCA), but actually never use it, which I thought was a bit weird. It’s apparently a flexible method for just evaluation of virtual environments (Neale & Nichols, 2001) so that is definitely something I think they should have used.

A case study explores the underlying principles of a topic. It is usually conducted in the social sciences as a way to explain the cause of some events (could be one, could be many)(Wikipedia, 2013a. Eisenhardt, 1989). 

Paper II

The second paper I’ve chosen, containing a case study, is called “The tweets that killed a university: A case study investigating the use of traditional and social media in the closure of a state university” (Neale & Nichols, 2001) and was published in Computers in Human Behavior, a journal with an impact factor of 2.067 (2013).  

To collect data they used online surveys and data mining to determine the usage of social media among different groups. As we discussed the previous week the use of online surveys often lack in the response rate and here, only 19% of the university students responded. Also, you can’t be 100% sure of the answers you get (i.e. “On the internet nobody knows you’re a dog").

There is a clear distinction in what target groups they have chosen and their research question is clearly stated. The use of both surveys and data mining also correlates with Neale & Nichols process for a case study. I feel that they use all the steps presented in Neale and Nichols study. They use several approaches to analyze the data (Kendall tau correlations, chi squared analysis, Mann-Whitney U’s and the Radian6 platform) that seems valid but I think it’s a big drawback that they only rely on 19% of the students. Especially when they are examining people’s social engagement; it feels like the ones answering already have bigger involvement (since they answer). By getting feedback from the ones who didn’t reply to the mail (i.e. people with low social engagement (?)), the answer could be very different. 

However, I feel that the biggest weakness with this study is the name of the paper. In my opinion, if you label a paper “The tweets that killed a university”, the tweets analyzed in the paper should have something to do with the closure of it. But in reality, the university didn’t close, it just changed from public to private (and changed name) and the tweets didn’t have anything to do with that. They only analyzed the tweets that occurred during that transition, and for me, that’s pretty misleading.

Resources

Eisenhardt, K. M. (1989). Building Theories from Case Study Research. Academy of Management Review, 14(4), 532-550.

Harding, J. (2013). Usability of geographic information–Factors identified from qualitative analysis of task-focused user interviews. Applied ergonomics.

Neale, H., & Nichols, S. (2001). Theme-based content analysis: a flexible method for virtual environment evaluation. International Journal of Human-Computer Studies, 55(2), 167-189.

Wikipedia (2013a). Case study.

URL: http://en.wikipedia.org/wiki/Case_study (2013-12-13)

Wikipedia (2013b) Semi-structured interviews.

URL: http://en.wikipedia.org/wiki/Semi-structured_interview (2013-12-13)