Some thoughts on computational teaching.

I am a big fan of digital education. In my previous job as a senior lecturer, after several years of repeatedly teaching the same topics for many years, I had honed a bank of vignettes to explain various concepts to students in a way that I believed captured their essence and helped them to stick in people’s brains. But once perfected, the task of having to repeat the same thing over and over again becomes irksome, especially when you know there are technological tools perfectly suited to the job of digitally capturing your delivery and effortlessly scaling it up to an almost unlimited audience. In the department of Multidisciplinary Engineering Education, where teaching at scale is the day job, we routinely rely on an array of digital tools to make our delivery effective and efficient. 

The pandemic has thrust digital education squarely into the limelight and The University of Sheffield is more strategically considering their approach to how it is used. My concern with this newly developed excitement about this area is that almost any activity involving a student and a computer, in pretty much any context, now falls under the umbrella term of “digital education”. The problem with conflating many different facets of digital education, which all have their own objectives and constraints, under the same banner is that it creates barriers to finding commonality in approaches to share best practice and eliminate instances of reinventing the wheel. To my mind, there are (at least, I’d be interested to discuss further) three distinct branches of digital education. 

  1. The first is the use of digital tools to serve teaching to students. These are tools that are on the interface between the student and the teachers, such as virtual learning environments.
  2. The second is the use of tools to administer the teaching process, such as student record systems or attendance monitoring. These tools are on the interface between the students or teachers and the institution.  As these tools have a sufficiently different objective in what they are trying to achieve, I feel they deserve to be separated out as a distinct branch of digital education. 
  3. The third is the teaching of digital skills. Unlike the previous two, this isn’t about the use of digital tools to achieve a job, but the teaching of concepts about the operation of hardware/software and creation of digital content.

This particular order for the branches of digital education is more or less the order in which I would place my level of excitement that I have for each of them. In the first category there are mature technologies to achieve the bulk of what academics need in order to deliver teaching. There are some incremental gains to be had within this space but it is unlikely there will be a revolutionary change any time soon. Some exciting opportunities exist in the world of remote access to laboratory teaching and MEE is part of the vanguard, but this is a niche subject area. 

There is, for me at least, some excitement about the second branch. From an educators point of view the administration of teaching is not particularly interesting and often something that needs to be done for auditing purposes, but does not directly contribute to the learning experience.  That being said, anything that can be done to reduce the time and intellectual effort required to perform administrative duties frees up those resources for the teacher to invest in the more important job of designing and delivering education. So from an efficiency perspective, and as an engineer I am delighted by the elegance of efficient systems that solve problems, I feel there are some ripe opportunities in this second branch of digital education. 

The third branch is about teaching computational concepts and this is the one that excites me the most, for lots of reasons. Pockets of this type of teaching exist in different contexts across our institution providing an opportunity for coalescing delivery with some high quality, centrally supported resources. There is also the delicious circularity of delivering the third branch of digital education using the first and second branch. But it is the universally applicable requirement for all graduates to have a robust foundation in understanding how computers work and a mindset that allows them to be exploited to achieve their creative visions that, for me, makes this the most exciting branch. And the knowledge and experience of computers that students bring with them when they arrive is incredibly varied. 

I grew up at a time when computers were just becoming widespread enough for each school to have a few and the curriculum was aware enough of their potential to be important in the future, but most teachers didn’t have, or want to have, any experience of using them. A few people, including myself, had a simple home PC running a command line based DOS or, if you were rich, RISC OS with a fancy graphical interface running on an Acorn. Within a generation the use of computer hardware and digital systems has changed from the preserve of industry experts and knowledge hobbyists to an indispensable part of virtually every aspect of people’s lives. It is now unthinkable that an employee in almost any capacity wouldn’t be able to use the basic functions on a computer or mobile device. 

One detrimental aspect to this rapid advancement in the utilization of digital systems is that they have been optimised to the point that users do not necessarily need to know how they work in order to achieve tasks. For my generation and those before me, the barrier to entry for early adopters was significant, as without at least a modicum of understanding about how computers worked you were unable to achieve even simple tasks. There was a distinction between people who knew how to use computers and those who didn’t, and the former were often called for tech support by the latter.  Nowadays, with almost no comprehension of the complexity required to realize such a technically demanding challenge, even tiny children are able to operate an iPad (other tablet computers are available) to navigate to their favourite TV shows on demand. There is still a spectrum of knowledge of the underpinning concepts of how computers operate within the current generation, but now there is also a much more widespread confidence held by people to simply use the computer as a familiar tool. My worry is that this confidence is conflated with expertise, especially amongst an older generation that may not be able to tell the difference.

I’ve noticed this trend during my time in higher education. Students are increasingly able to use a computer to perform tasks with limited understanding about fundamentally what it is doing to achieve the outcome. Why is that a problem? I’m a terrible cook, but I’m able to feed myself and my family by blindly following a recipe. Likewise, learning to use a piece of software can get you to a predetermined outcome. Given the ubiquity of computers previously discussed and the pivotal role they do and will play in people's careers, my ambition is to raise the expectation for all students to possess a computational mindset. Graduates should be able to go beyond following a recipe or reaching predetermined outcomes, but aspire to create and innovate, identify opportunities and constraints and understand how and why things don’t work so that problems can be solved. When it comes to digital, students should be the master chefs, not the recipe followers. And I don’t think there is any field of study where these ideas and skills would not be relevant. 

A journey for teaching excellence: from teamwork to artificial intelligence

I was only a few months into my current post when I was asked to support the Global Engineering Challenge (GEC) and Engineering You're Hired (EYH) weeks. I was mesmerized. What struck me was how engineering background diversities were removed by having students from different departments working together to emphasize the learning of another fundamental skill: teamwork.

However, I felt that those two events were too isolated in the students’ undergraduate programme, and I thought that my teaching settings (i.e., laboratory activity in small groups throughout the academic year) would be ideal to have an ongoing and explicit engagement with teamwork. After attending a Belbin Team Role theory workshop [1] and gaining further knowledge during my PGCert, I developed some content about teamwork theory to integrate into my teaching. In particular, I embedded Tuckman’s Team Development Model [2] into my introductory speech before a laboratory for first-year students. Based on my interpretation of the students’ body language (e.g., no eye contact, looking unengaged), my attempt was a failure. Acknowledging my inexperience, I cherished the feedback from some colleagues. A senior colleague pragmatically spotted the issue, describing it as a demand-offer problem: first-year students enrolled to learn engineering and not pedagogical theory. Another colleague, based on his successful experience, suggested a gaming component to introduce this topic to students. Therefore, I considered removing the teamwork element from my introductory speech and adding it to the practical part of the session. However, this would have required a coordinated effort of the whole teaching team including the GTAs. Unfortunately, at that time, GTA performance in MEE was referred to as inconsistent and often poor in the yearly students' survey.

This became a pressing matter to address both for MEE, striving to meet the highest standards of TEF [3], and for myself. I realized that engaging students in non-engineering topics required a more professionalised teaching team. Thus, given the wide use I make of GTAs, the success of the teaching in the Structures Lab hinged not only on my ongoing training but also on the one of the GTAs.

After critically reviewing the literature about GTA training and contextualising the findings with the reality at the University, Faculty and Departmental levels, I decided to introduce a GTA training that is continuous and subject-specific and that supports the development of their teaching profile. My aim with this training was not only to support the institutional commitment to staff development [4] but also to allow GTAs to gain pedagogical skills essential for their future careers.

The new methodology consisted of an ongoing cycle of training-teaching-feedback. The training, based on microteaching [5], was run in two stages. The first one, individual, where, in turn, each GTA familiarised with pedagogical concepts contextualised to engineering practical teaching; and the second one, designed to create a “safe” environment for rehearsing among peers, where the trained GTA would lead the group session. The feedback, gathered as an online form, represented a critical element of the aforementioned cycle, as it provided GTAs with an opportunity for self-reflection but also allowed me to assess the training based on GTA self-efficacy [6].

The analysis of the feedback confirmed the effectiveness of the training method and was also instrumental to identify some hidden problems that I will need to address moving forward. For example, some GTAs underwent a “shock phase” in their second semester, from which the literature suggests they can exit only through continuous training [6]. Also, only a few GTAs seemed eager to develop their teaching profile. Making GTAs value their role as teachers may require a broader strategy possibly involving their supervisors to find the right balance between their teaching profile and the contrasting but primary research one. I believe that both issues stemmed from neither the cognitive nor the psychomotor domains of GTAs, but, conversely, from a not sufficiently high level in the affective one. They seek knowledge (as inferred from the feedback analysis), but what they probably need is a purpose.

Nonetheless, what alarmed me the most, was the poor engagement that GTAs had with feedback. While for the proposed training this was probably due to the unsuitable format, I also believe that this represents a symptom of a deeper and wider issue: some GTAs did not know how to provide feedback. As I was thinking about the root of the issue, a chat with a PGCert fellow, made me realized that there is a general expectation that everyone knows how to provide/receive feedback. Alas, this is far from the status quo. For example, I was indoctrinated to provide/receive feedback but I was never trained. I suspect neither are the GTAs or our students.

I have come full circle and I now understand that to teach teamwork to students I will need to focus not only inside my classroom but also tackle wider issues. On one hand, I will need a team of professional GTAs “practically” supported with ongoing training and “emotionally” empowered to recognise their value. On the other hand, more broadly, I will need to work towards a more effective strategy for teaching and deploying feedback that takes into account both the students need of almost real-time feedback and the current limited time that staff can dedicate to provide it. I believe that only technology can overcome this deadlock and specifically I look with interest at artificial intelligence (AI). The use of AI in education has grown over the past two decades [7], and, while challenges still exist, AI also offers a wealth of opportunities like the potential for providing verbal and oral feedback to students that is both personalised and just-in-time. While this will help students to customize their engagement with feedback, I also plan to explore peer feedback as a strategy to teach students (as well as GTAs) to provide feedback and act on it.


  1. Belbin, R.M., (2011). Management teams: Why they succeed or fail. Human Resource Management International Digest.
  2. Nestor, R., (2013). Bruce Tuckman’s Team Development Model. [online]. [Viewed 15 June 2020]. Available at:
  3. The University of Sheffield, (2016). Learning and Teaching at the University of Sheffield 2016- 2021 [online]. [Viewed 27 01 2020]. Available at:!/file/FinalStrategy.pdf
  4. Bagshaw, A. (2017). A Beginner’s Guide to the Teaching Excellence Framework [online]. [Viewed 27 01 2020]. Available at:
  5. Calonge, D. S., Mark, K. P., Chiu, P. P., Thadani, D. R., & Pun, C. F. (2013). Extreme-Teaching-2 (XT²): Evaluation of an innovative semester-long intensive GTA training program based on microteaching. International Journal of Teaching and Learning in Higher Education, 25(1), pp. 129-143.
  6. Chiu, P. H. P., & Corrigan, P. (2019). A study of graduate teaching assistants’ self-efficacy in teaching: Fits and starts in the first triennium of teaching. Cogent Education, 6(1), p. 1579964.
  7. Roll, I., & Wylie, R. (2016). Evolution and revolution in artificial intelligence in education. International Journal of Artificial Intelligence in Education, 26(2), pp. 582-599.

Reusable Teaching Blocks


This blog post is, in equal parts, personal musings, an idea and a call to arms. Reusable teaching blocks are an idea that could be very powerful?

The department of Multidisciplinary Engineering Education is charged with a singular mission: to make the delivery of practical teaching efficient through scale. We identify the content and resources that can be shared by more than one engineering discipline and, when there is broad appeal, it justifies investing a significant amount of effort in developing a high quality, reusable resource.  Chemical, Aerospace and Civil engineers may apply their knowledge to different contexts, but they all require a fundamental understanding of, for example, fluid mechanics. Once we have developed a scalable fluid mechanics lab activity for one degree programme, it is really easy to offer it another with a few tweaks to account for a particular context. 

Because our activities are designed to be implemented using equipment in our laboratories, the extent to which they can be shared is limited by the geographical proximity of our students. Additionally, the equipment requires maintenance, space and physical infrastructure, all of which makes it difficult to replicate the teaching we deliver outside the University of Sheffield. But what if we could replicate the model of sharing teaching resources with the didactic teaching of theoretical engineering concepts rather than practical ones?

We propose the idea of discrete, reusable teaching blocks, which are stand alone (or with a small amount of defined pre-requisite knowledge) digital, scalable objects that are used to deliver compartmentalised theoretical content. We feel that the time for this idea is now. After the covid-19 pandemic engineering educators have been kickstarted into developing lots of online resources, and gained experience, confidence and competence in doing so they may not have had before.  The sector has had an opportunity to test en-masse what works well and what doesn't. And students have been forced to embrace new ways of being served content and have identified what aspects they find valuable and appropriate to be delivered in this way. 

The zeitgeist in engineering education is for learning through authentic or problem based approach. This is an excellent approach to engage students with real world applications and demonstrate the value of their knowledge, understanding and skills.  But it is important that we find mechanisms to actually teach the large amount of knowledge, understanding and skills that are required to engineer solutions to problems. Any tools used to make the teaching of engineering concepts more efficient, for both staff and students, allows more effort that can be expended on staff facilitating and students engaging with more open ended project type learning. 

Should all the teaching of didactic content be delivered online and with homogenous blocks of content? Of course not. Individual building blocks used in construction have no form in their own right. It takes the application of creativity and vision to assemble a collection of building blocks into a recognizable and coherent structure. And the same can be true for reusable teaching blocks. We can separate the functions of producing teaching material blocks and the function of selecting, arranging and contextualising them  If the job of producing the contents is done once, done well and used by many different courses, the intellectual effort of the educator, previously used to build the content, can be focused on the curation placed in the context of the disciple. 

The idea of reusable teaching blocks aligns perfectly with the method of flipped classrooms. When learning is flipped, didactic learning is delivered while staff are away from the students,allowing activities that are more valuable to be conducted during the face to face teaching.  Reusable teaching blocks could be used while the students are self-studying and the contact time is used to inspire and enrich the curriculum by articulating how the blocks relate to one another and using the content in the context of the specific disciple.

CDIO is ostensibly a framework to codify teaching through problem based, design/build/test learning to allow quick deployment without the need to reinvent the wheel. Educators can dip into the framework and pull out the resources and teaching methodology to directly embed into their curricula. While problem based learning and design/build/test are engaging ways of demonstrating the application of engineering theory to real world problems, it can’t be used without students having basic engineering knowledge and understanding. I mean, it’s right there at the bottom of Bloom, indicating that you can’t do the application and analysis without the lower levels of the pyramid being first being there to rest on: 

What if a collaboration of like minded educators could get together to set up a framework with similar reach and value as CDIO, but for the reusable teaching blocks of didactic engineering concepts? Wouldn’t that be valuable for all engineering educators? We would be very interested in joining forces with anyone that could help make this idea come to life. 

The first step would be to define a framework. Here is my starter for ten: 

In the metaphor, the type of blocks we are imagining about are more like wooden building blocks than lego (other variously coloured interlocking plastic bricks are available).  Lego blocks have a pre-defined orientation and limited ways in which they can interconnect. Reusable teaching blocks should be more like wood teaching blocks that can be oriented and arranged to build any conceivable structure. The concepts are the blocks, the educator is the architect designing the structure and the disciple context is the glue that holds the blocks together in the chosen form. To allow facilitate this, reusable blocks should ensure:

  • They are stand alone or with as little prerequisite knowledge as possible. 
  • Where prerequisite knowledge is unavoidable, blocks should be clearly signposted (preferably to other reusable teaching blocks). However complex learning environments can be created by stacking a number of blocks into ‘Elements’. These can be reused in part or whole as needed.
  • They are atomistics: blocks should aim to introduce as few concepts as possible 
  • Specific applications or examples of concepts should be separated from the description of the concept. Context and examples can be created as additional blocks to form part of the menu.
  • Terminology or nomenclature should be fully explained to ensure it is stand alone.
  • They are free of decorative clutter, subject or context information or corporate branding.
  • They may have associated micro assessments to ensure engagement.

Popular Posts