Intro

In 2022, Bodong Chen and Stephanie D. Teasley (Chen & Teasley, 2022) gave an interesting report on the state of learning analytics (LA) for collaboration in both academic and professional spaces. They discussed some key elements of the overlap of the use of the LA technology and identified how with different needs and use cases for the same technology, there needs to be coordination between the education and professional communities to help streamline the evolution and maturity of LA. The conflation of the terminology, the scale of group sample sizes, and the reason or purpose for using LA are elements that popped out at me during my reading.

Terminology

In any field of study or work, there is always jargon that is very specific to that field. Often, a field even has its own slang. Just as often, different fields can use common terms that are assumed that everyone knows because they are not jargon. Chen & Teasley touch on this early in the paper and passively refer to this point throughout the paper. They identified that often the terms collaboration and cooperation are used interchangeably which may be fine in some cases, but in other cases, such as the very specific application of LA to provide insight into either group or individual learning, these terms mean something entirely different. Collaboration means working together for a purpose common to everyone working together, while cooperation means working together for a purpose exclusively for one of those in the group (Moseley, 2020). From an LA standpoint, this is a huge differentiator depending on what is the scope of the measurement. If we are looking to evaluate how a group works together to achieve a stated goal the evaluation of each group member’s contribution to the greater collaboration needs to be identified. Likewise, if we are looking to evaluate how an individual learner is doing on their goals and they get help along the way, that needs to be identified and addressed that someone cooperated with the learner to achieve their goals.

Scale

While the tools and the technology of LA are the same in both the academic and professional spaces, the scale often is not. And “scale” carries different meanings between these use cases as well. Scale in the academic space can often refer to “small-scale” instances where scale is measured by a handful of students, a closed-ended assignment, or a short term. Whereas scale in the professional space can include large teams, research and development with no upfront definition of project completion, or long-term projects. If one is to transpose findings between the academic and professional implementations, this scale needs to be identified early in the adaptation and determine whether or not the findings would be particularly useful.

Purpose

Chen and Teasley also discuss the reason behind why one would use LA in either the academic or professional space. The reasons they suggest are to have LA be the engine behind the artificial intelligence (AI) (i.e., the actual intelligence) that is working either as a learning partner or a learning regulator. If working as a partner, then the LA-supported AI takes the role of tutor (in academic use) or virtual assistant (in professional use). If working as a regulator, then the AI can monitor classrooms in academic use to expand teacher oversight (i.e., virtual teacher’s aide) or it can monitor software interactions in business and mission-critical operations.

Surprise

The surprise I found in reading this is that Chen and Teasley mention several areas of technical curiosity I have in the same paper and paint the picture in my head of a Venn-like diagram of the potential overlap and integration between them – LA, AI, Human-Computer Interaction (HCI), and Human-AI Interaction. Each of these technologies can feed into each other with the proper systems engineering and integration if the system is built with the big picture of possibilities in mind from the beginning.

Eureka!

The Aha, or eureka, moment I had while reading this chapter is also, I hate to admit, something of a “well duh” moment. I have been searching for the how or where within the sphere of AI I would find the means to create adaptive, real-time feedback into a learner’s workflow. I knew that somewhere there had to be something that generated the properly timed, formatted, and stated information that would drive the next best action to take from the software-defined tutor and I now know its name – learning analytics. Trying to look at this notion of an AI-supported learning platform from the mind of an industrial automation engineer, I was half-expecting some sort of auto-dynamic lookup table (LUT) that just “existed.” Gee… what a rookie. Now, I know that the “auto-dynamic LUT” is the database populated by LA. Eureka!

References

Chen, B., & Teasley, S. D. (2022). Learning Analytics for Understanding and Supporting Collaboration. In G. S. Charles Lang (Ed.), Handbook of Learning Analytics (Second ed., pp. 86-95). Society for Learning Analytics Research. https://doi.org/10.18608/hla22

Moseley, C. (2020). Collaboration vs cooperation: what’s the difference? Jostle.me: https://blog.jostle.me/blog/collaboration-vs-cooperation