Skip to content
 

How can teachers of (large) online classes use text data from online learners?

Dustin Tingley sends along a recent paper (coauthored with Justin Reich, Jetson Leder-Luis, Margaret Roberts, and Brandon Stewart), which begins:

Dealing with the vast quantities of text that students generate in a Massive Open Online Course (MOOC) is a daunting challenge. Computational tools are needed to help instructional teams uncover themes and patterns as MOOC students write in forums, assignments, and surveys. This paper introduces to the learning analytics community the Structural Topic Model, an approach to language processing that can (1) find syntactic patterns with semantic meaning in unstructured text, (2) identify variation in those patterns across covariates, and (3) uncover archetypal texts that exemplify the documents within a topical pattern. We show examples of computationally-aided discovery and reading in three MOOC settings: mapping students’ self-reported motivations, identifying themes in discussion forums, and uncovering patterns of feedback in course evaluations.

This sounds like it could be useful, especially if the data collection and analysis is all automatic. I’m sure the model will have a lot of problems—all models do—but that’s ok. The instructor could run this program, look at the results, see what makes sense, and see what doesn’t make sense. Ideally the program would come with some feedback options so that Reich et al., as developers of the software, can improve the model and make it more useful. Thus, a system with its own built-in mechanism for improvement. Perhaps my posting here can start that process going.

2 Comments

  1. Rahul says:

    Most online surveys have a comments / sugesstions / “any other option” to be entered in a freeform text box.

    Do they typically pipe that to /dev/null? What’s the status quo in the survey / market research industry?

  2. Keith O'Rourke says:

    Might also facilitate more testing of concepts rather than primarily calculations (puzzle solving) in any statistics course …

Leave a Reply