Final Project Progress Update
Before updating my final project progress, I would briefly explain the background of my study, research methodologies, and research questions. E-learning has always been my most passionate personal and professional interest. From my own experience of an online learner, I recognized that though learning activities in online environment may vary course by course, yet discussion forum is always an inevitable part of e-learning and it has become the locus of many research in the area of computer-mediated learning. Therefore, I have decided to conduct a pilot study using three different methodologies - data analysis, social network analysis, and content analysis to examine learner-learner interactions on the discussion forum of an entirely online asynchronous course at Teachers College.
Inspired by the study of Lowes and colleagues (2007), my study aims to explore the outcome of each methodology: (1) What insights concerning the structure and functioning of these discussion forums could be acquired from these different methodologies? (2) Compared to Lowes and colleague’s study, would the data analysis conducted in Canvas be more insightful than that in Blackboard? (3) Would the results of three methodologies be complementary, repetitive, mutually exclusive, or conflicting? Additionally, using social network analysis and content analysis, I would like to gain insights into patterns of interaction and factors of an interactive discussion forum: (1) What the interaction patterns in the discussion forum could be? and (2) What factors or indicators would provide useful insights into an interactive and engaging discussion forum?
After the class session last week when we had the social network activity, I am even more interested in using the social network analysis for my study. Rethinking about my initial research plan, I am considering to remove the last methodology of analysis - content analysis and more focused on the social network analysis since I am positive that this methodology would potentially reveal many interesting insights into how learners interact to each others on discussion forum and implications with regard to designing an interactive and engaging discussion forum for online learning.
Dataset is the crucial part of my study, so I have started thinking about how I would collect the data effectively and efficiently as data collecting might be very time-consuming and tedious if the dataset is too large. There are two primary sources of data that I use in this study: (1) the pre-program survey created in Google Form and (2) the forum discussion posts on the Canvas course site. Regarding the survey data, I do not worry too much about this as Google has done a very good job in consolidating and visualizing the data that makes it ready for analysing. However, collecting the discussion posts is a really cumbersome task though I have confined the data set to only 5 modules (3 to 7) out of 15 modules of the entire course. In the Lowes and colleagues (2007)’s study, they did not describe how they collect the data, so I assume that they might have a graduate student or a research assistant conducting the job manually. In my case, considering the facts that there are nearly 200 learners participating in the course studied and the number of discussion posts for each module is about 400 posts, collecting these posts manually is absolutely not a wise, not to say impossible thing to do for me. In addition, in our last class session, one of the classmates with whom I discussed about the project told me that she was initially interested in exploring the discussion forum on Coursera, but she did not think that she would be able to collect the data, so she had changed her research topic.
The issue of data collecting had stuck in my head as I want to find an efficient solution which not only works for me, but also can be used by other researchers who are interested in studying online discussion forum. With that in mind, I have spent entire three days to research and experiment various data collecting tools. As the discussion forum is simple a webpage, I focused my search on autonomous web scraping tool which are free, intuitive but flexible enough to support different customized data queries. Below you can find a list of the tools that I have experimented with.
All of them are developed as an extension/add-on of Chrome browser and mostly free. From this list, I have found Web Robots Scraper the best fit since it allows users to program their own scraping function in Javascript to manipulate and export the data in the format that is required by data analysis software (i.e. UCINET for social network analysis).
The screenshot of the customized scraping function
that I programmed in Web Robots Scraper
After spending quite a great amount of time to develop this automatic data collecting procedure, I think that this procedure would be interesting and helpful for other researchers examining online discussion forum, so I will include it in the appendix of my paper and post it on my blog when I finish this project.
Stay tuned!
Reference:
Lowes, S., Lin, P., & Wang, Y. (2007). Studying the Effectiveness of the Discussion Forum in Online Professional Development Courses. Journal of Interactive Online Learning, 6(3). Retrieved from http://www.ncolr.org/jiol/issues/pdf/6.3.3.pdf
