Great teaching cannot be captured by ‘dangerously soulless’ algorithms

Beware reliance on teaching excellence framework metrics, says Claire Taylor

五月 31, 2016
teaching and learning
Source: iStock

Recently, I was listening to a BBC radio feature exploring how new artists are identified for the Radio 1 playlist.

Now, in my naivety I thought that this was a straightforward enough process, at the heart of which was the simple act of listening to the artist in question (ideally both recorded and live) followed by inviting others to also listen to the same artist and deciding if they warranted further exposure through national radio. Well, to some extent this is true; listening, experiencing and responding to the artist in question are components of the process, but there is much, much more.

I was not aware, for example, that alongside a basic listening exercise (although often of an excerpt of a track, not the whole thing) there is also a process that looks at data analytics related to the artist’s presence online. Metrics such as YouTube views, SoundCloud hits, Shazam ratings, Twitter followers and Facebook likes comprise the dataset that assesses what is or is not "good" music, and influences what is and what is not played on the radio.


Also by this author: Never mind the teaching, where's the learning?


In other words, hearing good music is not sufficient. The assessment of what is good has to be supported by data that measures online popularity through views, hits, followers and likes.

I like to think that I know a good tune when I hear it. I would go further and say that I know what great music sounds like and feels like (to me). And that is the nub of the matter – "greatness" is exemplified by some sort of connection to me as an individual person. The music transports me somewhere or connects with an emotion or unleashes a memory, or quite simply makes me feel good.

This impact is felt through a connection to an individual, not through data analytics. And the same is true of great teaching.

Great teaching has to be seen, heard, felt and experienced in order to achieve validation as "great". It has to impact on individual learners in the moment of delivery, speaking to the heart and the soul as well as to the mind. Great teaching goes beyond the mere transaction of sending and receiving information to the very heart of challenging, questioning, disassembling and reassembling the essence of knowledge itself and that is a tricky thing to identify, capture and measure through data analysis.

Now, don’t get me wrong. I like data. I love accessing information that helps me understand things better. I enjoy poring over figures and interpreting what they may mean for the student experience at my university.

I see data as a helpful tool in ensuring that our students are supported to make the most of the opportunities we provide for them. Data has the potential to help us identify patterns of engagement and intervene for the benefit of student learning. Overall, data can be a good thing.

But when data becomes the only thing on which we base decisions and even judgements, then that is when we are entering dangerous territory. In relation to great teaching, data – for example in the form of learning analytics, student achievement measures or employment statistics – provide just one tool from which measures of greatness or excellence may be derived. Furthermore, it is a flawed tool, because if we are truly honest we know that labelling something as excellent is essentially a value judgement – an assessment of something set within a very individualised context and operating with particular standards and priorities.

In other words, what one individual experiences as great or excellent may be another’s worst nightmare.

Proposed teaching excellence framework (TEF) metrics are only proxy measures for capturing the complicated, mysterious and essentially experiential art of teaching; these metrics are mere shadow representations of excellence and not actual true measures.

As university leaders and managers we ignore this fact at our peril and so in addition to scrutiny of relevant data sets I challenge us all to actively seek out great teaching within our own universities – look for it, identify it, enjoy it and celebrate it. Great teaching itself can only be measured in terms of how it is received by the learner through direct, meaningful experience and interaction between the learner and teacher herself.

Surely these softer, qualitative, experiential measures should be at least of equal and probably of more importance than a neatly packaged form of management information, just as with the Radio 1 playlist selection mechanism?

We operate in an environment where statistics related to an artist’s popularity have become key decision-makers in relation to who is given radio airtime and who does not make the cut. But a reliance on data and algorithms can feel dangerously soulless; I am far more interested in connecting with a really good song and I hope that the analogy for great teaching is clear.

Great teaching touches the heart and the mind; it makes a difference that goes beyond metrics; it acts as a catalyst for deep and lasting transformation for the learner. Data has its place when it comes to developing an evidence base in relation to what may be construed as good; but great teaching has to be experienced to be believed. 

Claire Taylor is pro vice-chancellor (academic strategy) at St Mary’s University, Twickenham.


Write for us

If you are interested in blogging for us, please email chris.parr@tesglobal.com

请先注册再继续

为何要注册?

  • 注册是免费的,而且十分便捷
  • 注册成功后,您每月可免费阅读3篇文章
  • 订阅我们的邮件
注册
Please 登录 or 注册 to read this article.

Reader's comments (2)

My experience of this is that it became a little repetitive.
And Learning Analytics (with those "scary" algorithms) requires a feedback loop to learners - otherwise it's just educational data mining.