Week 7

Analytics Methods

Tools and methods share a heritage in statistics and established quantitative and qualitative research methods. This week will introduce three analytics methods: social network analytics, discourse analysis, and influence analysis. Additional methods and tools to support learning analytics methods will be discussed in the next several weeks.

We won't be going into great detail about the data-mining process: acquiring, preparing (cleaning), and exploring data. The AU School of Computing and Information Science has a course devoted to data mining that will go into greater technical detail. For now, it's important to note that the quality of your data will dramatically impact the time spent on analysis. Good quality data that doesn't require much cleaning or normalizing is more useful for analysis. If data is poor quality, additional time and effort will be required to get the data into a usable format for analysis. If you're interested in a quick refresher (or introduction) to data mining, review the following resources:

Introduction to Data Mining. Retrieved from https://youtu.be/f7NfO16l04U

Caroline Haythornwaite: Learning networks, crowds, and communities.

Erik Duval: Attention please! Attention metadata and interaction.

Readings

Arnold, K.E. (2010). Signals: Applying academic analytics. EDUCAUSE Quarterly, 33(1). Retrieved from https://er.educause.edu/articles/2010/3/signals-applying-academic-analytics

Bakharia, A., & Dawson, S. (2011, March). SNAPP: A bird’s-eye view of temporal participant interactions. Paper presented at the first International Conference on Learning Analytics and Knowledge, Banff, Alberta. https://dl.acm.org/doi/10.1145/2090116.2090144]

Duval, E. (2011, March). Attention please! Attention metadata and interaction. Keynote presented at the first International Conference on Learning Analytics and Knowledge, Banff, Alberta.

Haythornthwaite, C. (2011, March). Learning networks, crowds, and communities. Keynote presented at the first International Conference on Learning Analytics and Knowledge, Banff, Alberta.

Kirkpatrick, J. (2010, November 30). Awesome: DIY data tool Needlebase now available to everyone. Retrieved from https://archive.is/w4J6C

Romero, C., & Ventura, S. (2007). Educational data mining: A survey from 1995 to 2005. Expert Systems with Applications, 33(1), 135–146. Retrieved from http://www.ecst.csuchico.edu/~bjuliano/csci693/Presentations/2008w/Materials/Lobban/DOCS/educational_data_mining.pdf

Suthers, D., & Rosen, D. (2011, March). A unified framework for multi-level analysis of distributed learning. Paper presented at the first International Conference on Learning Analytics and Knowledge, Banff, Alberta. [Video at https://dl.acm.org/doi/10.1145/2090116.2090124]

Activities

1. Continue working on your concept map as well as your analytics model project (due next week).
2. Tools like Needlebase are part of the next generation of relatively simple-to-use tools for interacting with data. Google Correlate is another example. These tools are defined by ease of use – individuals can often begin using them without an understanding of statistical or analytics methods. In your forum discussion posting this week, detail your view of this disconnect between "easy-to-use tools" and "limited understanding of methodology". What are the negatives and positives of this trend?