Sunday, February 25, 2024

Australian Universities Accord Final Report: Mostly Good

Australian Universities Accord
Final Report
, 25 February 2024
The Australian Universities Accord Panel, chaired by Professor Mary O’Kane, has released a 408 page final report (25 February 2024). Some of the suggestions I made in my submission (and no doubt many others made), found their way into the report).

There is also a Summary Report by the Department of Education, which at 34 pages is longer than the executive summary in the actual report, so seems a bit pointless (unless the department just wanted to put its own spin on things).

The report, while mentioning democracy and civic values, emphasizes the economic benefits of research and higher education. It raises issues of low demand from students, casualisation of the university workforce, and financial pressures. Equity also features prominently. The report makes 47 recommendations to address these, and other, issues.

Targets and VET

The recommendations under broad headings, the first of which is the "National Tertiary Education Objective". Interestingly this is proposed for both higher and vocational education, to: 

a. "underpin a strong, equitable and resilient democracy
b. drive national economic and social development and environmental sustainability."

Vocational education is mentioned further in the Attainment targets:

a. "a skilled workforce to meet the changing needs of the economy through a tertiary education attainment target of at least 80% of the working age population with at least one tertiary qualification (Certificate III and above) by 2050 compared with 60% in 2023

b. growth in Commonwealth supported places in higher education to achieve this target, more than doubling the number of students in Commonwealth supported places from around 860,000 in 2022 to 1.8 million in 2050 across all age groups

c. growing numbers of younger Australians with a university education, through an attainment target of 55% of 25 to 34-year-olds with a bachelor degree qualification or above by 2050 compared with 45% in 2023, noting that many will also have a VET qualification

d. a strong and growing contribution to tertiary attainment driven by TAFE and the vocational system, with a planning assumption that 40% of 25 to 34-year-olds will have a tertiary level vocational or technical qualification in 2050, noting that some people have both a VET and higher education qualification ..."

As it happens, in my submission to the accord panel, I suggested an 80% vocational target but at a more ambitions Certificate IV level, rather than Certificate III. The panel's proposed target of 55% of young people with a bachelor degree may seem modest, given this is already at 45%. However, we may have already past peak degree, with enrollments dropping, as young people decide a degree is not their best investment. Government would do well to agree with this assessment, and decide to fund more flexible forms of education, rather than traditional bachelors degrees.

Flexible Courses and RPL

Under "A more flexible and responsive skills system" the panel recommends "working with tertiary education providers, industry and business to adopt a consistent, published, national approach to recognition of prior learning (RPL) and credit recognition":

i."address historical, cultural and institutional barriers to RPL and credit recognition 

ii.make it easier for students to gain maximum credit for previous study and minimise

the time taken and cost to get a new qualification

iii.include recognising appropriate work experience

iv.improve student mobility to enter, exit and return to tertiary education"

While RPL is routine in VET, the difficulty will be in persuading and educating the university sector in how to do it. While universities could learn much from the VET sector, it will be difficult for the upper end of the university sector to accept this. Part of my day job for several years has been processing applications for credit for prior study, & RPL. This is something I have been formally trained to do in TAFE, & university programs, and though being on the accreditation standards board of my profession, but is not a skill many of my colleagues share, or value. Many universities in Australia share course designs across institutions, and international, where external professional recognition imposes standardization. However, this is not formally recognized, so a time consuming process is required to process every application for every course credit. This could be improved by giving incentives to universities to work together, and train staff.

The panel recognize the need for more flexible qualifications, with a recommendation for modular, stackable qualifications:

"5. That to help Australians quickly get the skills they need to fill jobs that are in shortage, the Australian Government establish a comprehensive system of modular, stackable and transferable qualifications, including microcredentials, consistent with a reformed Australian Qualifications Framework." (Page 19)

This one recommendation is worth a whole report in itself. In my submission, I recommended nested programs, rather than the more ambitious stackable. There is a subtle difference here. A university can create a set of qualifications which a student can undertake in sequence (nested). It is much harder to allow the student to assemble a larger qualification from an assortment of smaller ones (stackable). Even the much easier task of providing standalone micro-credentials is one that universities have been slow to achieve. The level of educational design skills  would need to be considerably increased at Australian universities to achieve this aim.

Work Integrated Learning

The panel proposes to have studnts "earn and learn while studying" with a "Jobs Broker" for relevant part-time work, paid by employers. 

"7. That to ensure students develop work relevant skills for employment after their study, the Australian Government increase opportunities for students to both earn and learn while studying by:

a. establishing a national brokerage system (‘Jobs Broker’) to support tertiary education students find part-time work and placements relevant to their fields of study. Delivery should be through a provider that charges paid subscriptions by employers. The service should be free for students, and allow them to earn income while studying and reduce cost of living pressures

b. promoting work-integrated learning (WIL) by working with peak bodies for employers, industry, business and tertiary education providers to deliver more WIL opportunities in curricula across all disciplines, and provide training to industry supervisors

c. improving measures of graduate generic skills as part of the Graduate Outcomes Survey and Employer Satisfaction Survey. The Australian Tertiary Education Commission should showcase best practice as part of its ‘State of the Tertiary Education System’ annual report

d. using models like degree apprenticeships that encourage an employment relationship as part of course design." (Page 19)

Again this is a recommendation which deserves a whole report itself, with implications for how higher education is provided. For students to be engaged in significant part time work requires them to be part time students. Flipping the Australian university system from a focus on full time study, to part time would seem radical, but is just a recognition of what has already happened. Much of the stress for students and staff comes from students who are holding down a job while studying full time. 

For the last eight years I have been helping teach students undertaking Work Integrated Learning, mostly in groups, but also individual interns. Even in the vocationally orientated filed of computing, during a boom in demand for skills in areas such as cyber security, analytics, and AI, this is not easy. Outside vocationally orientated disciplines, such as computing, engineering, and business, this is going to be especially difficult. There is then the problem of training academic staff to design, teach and assess WIL.

Professional Accreditation

The Panel's recommendations on "Professional accreditation bodies", were slightly jarring for someone who has served on accreditation setting committees. As a member of my profession, I am used to telling universities, and governments, what they have to do, not the other way around. 

"9. That to ensure professional accreditation including placement requirements are appropriate for industry and business skill needs, tertiary education providers and the Australian Government, through the Australian Tertiary Education Commission, work with professional accreditation bodies, to agree a code of conduct for these bodies. The code should ensure that any accreditation requirements are evidence-based and proportionate to the gain they provide and that placement requirements ensure that students gain industry relevant skills and experience without imposing onerous placement length and conditions." Page 20

That accreditation requirements should be evidence based is reasonable, and work placements not onerous. However this should be a mutual obligation, with the universities required to show they are teaching & assessing work relevant skills. 

One example of where WIL assessment can go wrong is with e-portfolios. Students are told to reflect on what they learned on their placement, without training in how to write reflectively. The approach I have been using for the last few years is to frame the reflective e-portfolio in the form of an application for a job, so it is relevant to the student, and have the careers staff teach this.

Participation targets for First Nations, Low SES, Regional, Rural, Remote  and Students with a Disability

In recommendation 10, the panel proposes participation targets for undergraduate university students by 2035: 3.3% First Nations, 20.2% low SES, 24.0% regional, rural, and remote, and current rate for students with disability. These are very modest targets, but even so no penalties for non-achievements, or incentives are proposed. Also by emphasizing university this may not be be in the best interests of studnts woH might benifit from attending VET first. 

Building aspiration including through increasing readiness for tertiary education and providing career advice

In Recommendation 11, the panel proposes work by federal, and state governments on "... outreach programs designed to develop familiarity with tertiary education". Harder is to "... ensure post-school pathways are visible and integrated into secondary schooling ...".

Fee-free preparatory courses

The Panel calls for free university preparatory courses. One way to fit this in the current system, I suggest, would be to have the preparatory courses run by the VET sector.

Support to participate and succeed in learning

In recommendation 13 the Panel proposes funding for under-represented groups, to meet the targets set in Recommendation 10. It would have been a bit easier to follow the recommendations, if these had been together. No specific amounts are mentioned, or if this additional funding is to come out out of that already provided to universities. Also there is no mention of developing curricular, materials, delivery techniques to suit these students. As an example, it is likely many of the studnts would benifit from blended programs, delivered in part remotely at home, and on country. Also the content of courses may need to be different. Such course content and delivery techniques would also greatly enrich the leanring experience of other students.

Financial support for placements

The panel proposes in recommendation 14 government funding for work placements in nursing, care and teaching professions. The argument presented, that this would reduce financial hardship, makes no sense, unless the government support is extended to all disciplines. 

Student income support

In recommendation 15 the panel asks the federal government increasing allowances and loosening rules. Hard to argue with that, except perhaps to suggest abolishing some of the rules.

Reducing student contributions and reforming HELP repayment arrangements

Of most interest to students will be the panels recommendation 16, to  reduce the burden of HELP loans, and in particular the effect of the Job-ready Graduates (JRG) package. At the time it was introduced, many warned charging humanities students as a way to direct them to what the then government saw as more vocationally relevant programs would not work. It did not work, and the studnts should not be further punished for past bad government policy. 

While previous recommendations were not specific about money, recommendation 17 proposes universities charging "high fees" (over $40,000) for graduate coursework (I assume a Masters), "be required to re-invest a proportion of income earned back into scholarships and bursaries to support students from under-represented backgrounds to access these courses". This Robinhood approach doesn't appear to be proposed for undergraduate courses, or for international student fees. Also it is a relatively mild levy, being retained within the institution.

Ensuring student safety and experience

The Panel calls for a National Student Ombudsman to handle complaints in recommendation 18. The Australian Government recently announced such an office would be established. It is usual for an industry ombudsman to be funded by the industry members, but there is no mention of funding in the announcement. 

More controversial is recommendation 19, calling for student unions funded from student fees. This would go counter to the previous government's approach of cutting off funding to student unions. As an online student for seven years, I was slightly annoyed by services and amenities charges, when they went to the upgrading of facilities I had never seen, and would never use, because they were a thousand kilometers away.

Early at-school offers

Recommendation 20 reads like a cartel wanting to maintain the market share of its members. The panel recommends "early at-school offers for 2025 and 2026 should not be issued before September" to "maintain the integrity of senior secondary certificates". However, if universities are able to determine which students are likely to succeed at studies without certificates, they do not have integrity. If the certificates are no longer of value to universities, they should be modified to suit other users, such as employers, and VET. This would also certificates to emphasize work relevant skills.

Quality learning and teaching

Recommendation 21 shows the Panel placing hope over available evidence, by suggesting measures to "prepare the higher education sector for growth in student numbers". I suggest a further decrease in domestic student numbers is more likely, and would be a better outcome in terms of public policy. However, in any case I agree with the panel about "using proven innovative learning approaches which embrace online and hybrid teaching modalities". However, what is lacking is any proposal for university teachers being required to be qualified to teach. Paradoxically, while universities teach the highest level teachers they have the lowest requirements for teaching qualifications. The panel is not proposing to change that.

International education

At this point, it is time for a break ...


Thursday, February 22, 2024

What country does Generative AI use as much power and water as?

Generative AI is the new bitcoin. Pundits used to compete for which country they could compare Bitcoin's power use with. Some nominated were: Greece, Ireland, * Netherlands, and New Zealand. I put a section on this in my Sustainable Computing course.

I tried asking Copilot:

"What country does Microsoft Copilot use as much power and water as?"

It replied:
"Microsoft Copilot, like all datacenters, consumes energy and water. Let's delve into the specifics:

1. **Power Usage Effectiveness (PUE)**: This metric gauges how efficiently a datacenter utilizes energy. The closer the PUE number is to "1," the more efficient the energy use. Microsoft tracks PUE globally and across operating geographies. While newer datacenters and those operating at peak utilization tend to be more efficient, the goal is to get as close to "1" as feasible¹.

2. **Water Usage Effectiveness (WUE)**: WUE measures the efficiency of water consumption for humidification and cooling in datacenters. It's calculated by dividing the liters of water used by the total annual power needed to operate datacenter IT equipment (measured in kWh)¹.

Now, let's address your question. Microsoft's water consumption for its cloud data centers increased by **34%** in 2022, totaling **6.4 million cubic meters** of water. Generative AI workloads are partly responsible for this surge³. However, it's essential to note that this data pertains to Microsoft's overall operations, not specifically to Microsoft Copilot.

In summary, Microsoft's datacenters, including those supporting Copilot, have a significant environmental impact. However, pinpointing a specific country for comparison would require more detailed data. 🌐💧🔌

Source: Conversation with Bing, 22/02/2024
(1) How Microsoft measures datacenter water and energy use to improve Azure .... https://azure.microsoft.com/en-us/blog/how-microsoft-measures-datacenter-water-and-energy-use-to-improve-azure-cloud-sustainability/.
(2) Microsoft's water consumption jumps 34 percent amid AI boom. https://www.datacenterdynamics.com/en/news/microsofts-water-consumption-jumps-34-percent-amid-ai-boom/.
(3) Microsoft 365 admin center Microsoft 365 Copilot usage - Microsoft 365 .... https://learn.microsoft.com/en-us/microsoft-365/admin/activity-reports/microsoft-365-copilot-usage?view=o365-worldwide.
(4) Measuring usage, adoption, and the impact of Microsoft Copilot. https://techcommunity.microsoft.com/t5/copilot-for-microsoft-365/measuring-usage-adoption-and-the-impact-of-microsoft-copilot/ev-p/4051091."

Tuesday, February 20, 2024

Learning to Reflect in the Age of AI

 


This semester I am aiming to take a tentative step into AI for teaching. This will be in three workshops planned at ANU, for the computer project student's capstone e-portfolio for the ANU Techlauncher program. The students have to prepare a portfolio in the form of an application for a real job. The question was: do we try to ban students from using generative AI to help them with this, or do we show them how to use it effectively and ethically? I will attempt to do the latter. 

The ANU now provides Microsoft Copilot, as part of the Office suite. This provides the opportunity to take studnts through exercises to use it, to help with their writing. I have not used Copilot, but have explored the technology it is based on since 2022. The idea is to get the students in workshop groups to ask a career related question of Copilot, then critique & improve the answer.

A couple of weeks ago I attended a two-day Symposium at USyd on using AI this way, with team-based learning. One tip given at the symposium to stop students simply relying on the answers given by the AI. The idea is to prompt students with a very localized question, which the AI model can only answer with generalities: 

Last week I attended an ANU AI Assessment Question Drop In Session. The impression I got was that the ANU would not be averse to using AI this way. One of the other people who dropped in for advice was already proposing to run some team-based AI sessions along these lines.


Sunday, February 18, 2024

Australia Can Offer Low Cost Online International Education and Premium On-campus at the Same Time

In a recent preprint, Heller and Leede (2024) find that the third of Australian university students who are on-campus international tend to come from high GDP countries. The researchers propose that instead Australia provide low cost online education for the benifit of developing nations of the region. While well meaning, I suggest Heller and Leede, need to consider how to make such a strategy palatable to the Australian Government, which would need to fund the strategy, and the Australian universities, which would need to implement it. The Australian Government has a brief to aid its own citizens, not those of other countries. Why would Australia voluntarily give up billions in export dollars from one of its major industries, and forego cherry picking the best and brightest minds of the region? Why would Australian universities give up a large part of their funding? Is there any reason why Australia should cut the price of this export commodity, and not others, such as wheat? 

I suggest, as hard as it is, it is possible to make a case for low cost online courses. Part of that case is geopolitical. Australia is competing for influence in its region. Since the days of the cold war, one way to compete has been with education. Under the Colombo Plan, Australia offered education to make influential friends in the region. A second argument is economic. One way to maximize revenue for sale of a commodity is to sell the same thing at different prices to different customers. This has been achieved by Australian supermarkets selling goods at higher prices in metropolitan areas through small boutique stores. It is also achieved by online retailers, who can set a different price for each retailer. Those offering services al often offer discounts for pensioners, ostensibly as an act of charity, but also because it is better to sell the product at a lower price that not at all to someone who can't afford a higher price. 

Australia could adopt a differential pricing strategy for students based on an ability to pay. This could be done first of all by offering low cost online courses. While online courses offer the same educational outcomes as face to face courses, they are perceived to be inferior. Also while the same course offered with the same staffing level costs about the same to deliver, they are perceived to be cheaper. As a result, Australia could offer low cost online courses, without significantly cannibalizing its on-campus enrollments. 

Australia did run the "Virtual Colombo Plan" (VCP), a $230M Australian Government/World Bank initiative from 2001 to 2006, for on-line education in developing nations (Curtain, 2004 and McCawley, Henry and Zurstrassen, 2002). The VCP ended in 2006 with little ceremony, but it did help fund the African Virtual University.

Tom Worthington addressing the
Computer Society of Sri Lanka
National IT Conference 2018 
In 2018, in a presentation in Colombo,  I suggested countries of the Indo-Pacific could jointly educate professionals using mobile devices, in to counter the influence of China's Belt and Road Education Plan. I provided more details in a formal paper (Worthington, 2018). Rather than creating free courses, dependent on intermittent foreign aid programs. I suggested that allied educational institutions, and entrepreneurs, could be assisted to set up not-for-profit, and for-profit programs.

Reference

Heller, R. F., & Leeder, S. R. (2024, January 2). A population perspective on international students in Australian universities. https://doi.org/10.35542/osf.io/hsnke

Worthington, Tom. Blended Learning for the Indo-Pacific. In Teaching, Assessment, and Learning for Engineering (TALE), 2018 IEEE 7th International Conference on. IEEE. url https://doi.org/10.1109/TALE.2018.8615183

Friday, February 9, 2024

Effective Learning Techniques Versus Feeling in Education Policy Development

In the last short talk at Team-Based Learning Collaborative Asia Pacific Community Symposium was by Prof Judy Currey at Deakin University. Professor Currey cited Deslauriers (2019), on the well known phenomenon that students prefer passive learning. It occurred to me that this might also apply to policy makers and legislators, who think that students learn best when listening to a teacher. Fortunately the head of the universities accord review, Professor O’Kane, gets it: "... stressed better tutorial-style teaching to provide for experiential learning and experimentation was needed ...' (Uni graduates lack crucial skills and online study is making it worse, Gus McCubbing and Tom Burton, AFR Aug 22, 2023).

References

Deslauriers L, McCarty LS, Miller K, Callaghan K, Kestin G. Measuring actual learning versus feeling of learning in response to being actively engaged in the classroom. Proc Natl Acad Sci U S A. 2019 Sep 24;116(39):19251-19257. doi: 10.1073/pnas.1821936116. Epub 2019 Sep 4. PMID: 31484770; PMCID: PMC6765278  DOI: 10.1073/pnas.1821936116.

Meet my AI alter ego

Phillip Parannik, and Finn Johnson
At day 2 of the Team-Based Learning Collaborative Asia Pacific Community Symposium we have been learning how to create AI characters to be used as tutors, & simulated clients for students. I was very skeptical of the idea, but in a few minutes I created "Tom", a simulation of me, for teaching computer students. I can now retire. ;-)

This is using Inworld, which was originally created for use in computer games, but is now being used for teaching. The workshop is being run by Phillip Parannik, and Finn Johnson from University of Sydney. The process was very much simpler than the last workshop on AI tutors I took part in, 2018.

If anyone in Canberra would ;like a system built with this, the ANU computer project students could help. 

Team Based Learning Helped with Remote Learning

Larry K. Michaelsen
Greetings from day 2 of the Team-Based Learning Collaborative Asia Pacific Community Symposium. The keynote today is Larry K. Michaelsen, pioneer of Team Based Learning (TBL). He said that teams formed slower online than face to face, also that technology had provided more options for prework. Also Dr Michaelsen reminisced about the marked sense card readers which were used for scanning student responses before laptops and smart phones were used. He pointed out that part of the team bonding process was filling out the one joint response. 

A key point for me was that this team building process could occur in asynchronous mode. That is Dr Michaelsen said that as long as there was a deadline for the team response to be submitted, the participants could be anywhere, arranging their own work. 

Dr Michaelsen suggested AI would have more difficulty answering TBL questions, as they are very specific. He gave the example of "Where would you locate a dry cleaning business in Norman Oklahoma, and why?". So I asked ChatGPT, and it gavce a very generic response, which an assor would not be impressed with. When I asked for more specific locations, it gave answers Dr Michaelsen said that anyone who knows Norman would know are not good locations:

"Determining the ideal location for a dry cleaning business in Norman, Oklahoma would depend on various factors such as target market, competition, accessibility, and zoning regulations. Here are some considerations to help you decide on a suitable location:

  1. Commercial Areas: Look for commercial areas with a mix of retail and service-oriented businesses. Areas with a high concentration of offices, hotels, or shopping centers can attract potential customers who may need dry cleaning services.

  2. Residential Neighborhoods: Consider placing the dry cleaning business in or near residential neighborhoods, as this would make it convenient for local residents to access your services. Proximity to residential areas can generate regular and loyal customers.

  3. Proximity to Offices or Corporate Parks: If you are targeting the professional working population, setting up near offices, corporate parks, or business districts may be advantageous. Many professionals require dry cleaning services for their work attire.

  4. Accessibility and Parking: Choose a location with easy access and sufficient parking space. Customers prefer convenience, and a location with ample parking makes it more convenient for them to drop off and pick up their items.

  5. Competitor Analysis: Assess the competition in different areas. If there are already established dry cleaning businesses in a particular location, evaluate whether there is room for another competitor or if there's an underserved market elsewhere.

  6. Demographics: Understand the demographics of the area. Consider factors such as income levels, age groups, and lifestyles of the residents to ensure that your services align with the needs of the local community.

  7. Zoning Regulations: Check local zoning regulations to ensure that the chosen location complies with any restrictions related to operating a dry cleaning business. Some areas may have specific zoning requirements for businesses that use certain chemicals.

  8. Visibility and Signage: Choose a location with good visibility from the road, and invest in clear and attractive signage. This will help attract passing traffic and make your business easily identifiable.

  9. Cost Considerations: Evaluate the cost of renting or purchasing space in different areas. Balance the costs with the potential customer base and revenue generation of each location.

  10. Future Growth: Consider the potential for future growth and development in the chosen location. An area that is experiencing growth may attract more customers over time.

Ultimately, a comprehensive analysis of the local market and careful consideration of these factors will help you determine the most suitable location for your dry cleaning business in Norman, Oklahoma."

One of the delegates suggested occasionally tell the students to turn off their gadgets. Dr Michaelsen suggested then asking studnts to reflect on the differences in the experience. However, I suggest the instructors need to also reflect on what technology they are dependent on: would the be willing to turn off the lights and air-conditioning in the room, turn off their hearing aid, take off their glasses? One generations new tech is just routine for the next. While is is tempting to think of the good old days, some technology is so useful it is worth becoming dependent on it.  



Thursday, February 8, 2024

Online Team Based Learning

Greetings from the first workshop at Team-Based Learning Collaborative Asia Pacific Community SymposiumRosa Howard, University of Sydney, is talking on "Adapting TBL to Changing Times". This highlighted the benefits of providing text base input, as opposed from verbal. This raises an interesting issue as to if graduates need to be able to confidently talk in front of large groups. If it becomes routine to use computer mediated communication in the workplace talking to a group becomes an obsolete skill, like handwriting.

The Fear of Failure & TBL

Greetings from the first workshop at Team-Based Learning Collaborative Asia Pacific Community Symposium. This is using Intedashboard software and ChatGPT. Something which slowly dawned on me during the morning session was that the medical educators were using Team Based Learning (TBL) not as a general term for learning in teams, but a very specific methodology. I felt a bit like the student who was not told there was to be a test.  At the start of the workshop we were asked to lo into Intedashboard, and was immediately asked a question I didn't understand, at which point I had the unprepared student panic. Having recovered from this I struggled with some of the acronyms MCQ (multiple choice questions), ILO (no idea). 

The workshop was modeled on TBL, and this was a familiar format for me. What I found fascinating were examples of giving ChatGPT very complex requests to create course material. Doing this never occurred to me, no more so that I would get a random person off the street and ask them to do so. Habitually I would consider the qualifications and experience of any person, company or organisation. But ChatGPT is essentially a black box. 

As a team we then used ChatGPT to create objectives for a learning exercise. It was interesting to see the range of experience with ChatGPT, with some participants having never used it, while others routinely used it to create course content. My team got a bit naughty, asking ChatGPT to write the evaluation of what we thought of ChatGPT. Another team accidentally ended up with a rubric for assessing learning objectives. One team found what ChatGPT too wordy, but another went further and told it to be less wordy, which it then was.

In terms of a TBL exercise I found this like the previous development exercises. The first is the worry of getting the software used to work. The next is how rushed everything seems to be. There is then the worry of getting the exercise done in time. My approach would always be to produce a draft answer, then refine it. However, that is an approach difficult to get a group to do. 

ps: One use for ChatGPT I hadn't thought of was suggested b y one of our team: planning an itinerary for a trip. 

Team Based Learning & AI

Professor Simon Buckingham Shum
Greetings from the University of Sydney where I am taking part in the Team-Based Learning Collaborative Asia Pacific Community Symposium. Most of the staff are from USyd medical schools, but I fell at home as we are talking about experiential learning. The keynote is Professor Simon Buckingham Shum, Director of the UTS  Connected Intelligence Centre, being provocative about AI. He suggested universities need to have their own generative AI tools, as it is not acceptable for staff and  students to have to enter their sensitive information into offshore commercial ones. One practical use is in debriefing students after they participate in a simulation. What is most useful about the Professor's analysis is that it treats AI as a tool, not a problem, and asks us how we might use it.

Professor Buckingham Shum pointed to a paper he wrote about how one such tool carried out a very good analysis of a complex policy paper (2024). There was one flaw in the Professor's analysis: he claimed the AI was in error by referring to an "Argument from ignorance" which did not exist. It turns out that this is known to philosophers such as John Locke, and politicians such as Donald Rumsfeld. When I asked about this the Professor suggested that students need to be gently introduced to the limitations of the AI, so they are not awed by it.

I am looking forward to workshops, and short talks later today and tomorrow, with people not only from Australia, but around the region.

ps: One area where TBL AI might be used is in helping students meet diffuse learning outcomes.

Reference

Buckingham Shum, S. (2024). Generative AI for Critical Analysis: Practical Tools, Cognitive Offloading and Human Agency. 1st International Workshop on Generative AI for Learning Analytics: 14th International Learning Analytics and Knowledge Conference (LAK’24), March 18-22, 2024, Kyoto, Japan [PDF]

Tuesday, February 6, 2024

Learning & Teaching This Year

Professor Abby Cathcart
Greetings from the ANU Learning & Teaching Day: Connect & Inspire. Professor Abby Cathcart, QUT Director Student Success & Teaching Advancement, addressed putting joy back into university study. She pointed out the difficulties with online learning during the pandemic. As she pointed out this was made worse by it being used as an emergency measure, and provided a vision for a path out of this conundrum. One off the wall idea was to shut down the learning management system for six months, but a more practical one was to require senior academics to teach. 

What resonated with me in Professor  Cathcart's presentation was communicating to new academics that teaching is a scholarly endeavor. This doesn't require every academic to become an educational specialist, but they need to learn a little. Another point I liked was that we need to do less as teachers. Teachers, and students can become overwhelmed by the online engagement required. When learning course design one of the approaches I took was to estimate the time required by the teacher, and the student, for each activity, and ensuring this fitted in a time budget for the course.

However, I was not convinced by Professor Cathcart call for joy in teaching. Rather I suggest academics should act professionally. They need to ensure they are competent to teach, and in doing so be able to do it efficiently, without overwhelming themselves, or their students. The average academic should not innovate in their teaching, instead they should learn researched, proven methods. I was bemused and frustrated during COVID-19 by academics proudly reporting how they were inventing ways to teach online, which actually had been known and used for at least a decade.

Professor Cathcart then addressed the issue of unngaged studnts. She [pointed out that just under half of studnts said they preferred blended learning, but most did not collaborate online. As an international online student for three years I experienced a deep sense of loneliness. What helped were group projects, where I was required to collaborate with other students. 

Professor Cathcart mentioned clubs and societies at university which help with engagement. However, a student has only a limited amount of time for study, squeezed in between work, family, and other responsibilities. Any time spent in extra curricular actives will therefore come from study time. As a graduate student I did not participate in any social activities, as I was there to get through the program as quickly as possible. My view was that social activity risked distraction from study, and failure. This is a reasonable, and ration approach for students. As a result, I suggest collaborative activities have to be built into the formal compulsory, assessed program. 

Professor Cathcart cited Taylor et al 2022 relating emergency online teaching to the stages of grieving. She pointed out that learning how to teach online can make the teacher's view of teaching online more positive. As someone how spent seven years being trained to teach, and particularly online, I can understand this point of view. However, for those for whom research is the priority, it is very difficult to convince them to do a whole lot of study and await enlightenment. 

A lighted point came in the presentation when Professor Cathcart discussed how teaching could bring joy for the teacher. She quipped that discussion of joy was always illustrated with a middle aged woman doing yoga, but that was not her. 

In her introduction, Grady Venviolle mentioned hat it was great to see parents with the new students on campus. I found this a little odd, as I was escorted to primary school for the first day, but the idea a parent would accompany me to university seems weird. Also I did not set foot on the campus of the last two universities I studied at. One was more than 1,000 km away, the on the other side of the planet. Having a campus is useful for a university, but not something every student can access. The challenge is how do we provide students with options, wherever they are.


Monday, February 5, 2024

Climate Update 2024: Extraordinarily Hot Globally

Greetings from the Climate Update 2024 at the Australian National University (ANU) where Genevieve Bell, the new Vice-Chancellor reflected on Nugget Coombs, who took her ten pin bowling. Dr Coombs is better know as one of the founders and early VCs of the ANU. Dr Bell pointed out that her predecessor had envisaged the University not just carrying out scholarship for its own sake, but addressing social issues.

Professor Mark Howden, Director of ANU Institute for Climate, Energy & Disaster Solutions and Vice-Chair of the UN Intergovernmental Panel on Climate Change then gave us the bad news on how the planet is heating faster than expected "Extraordinarily Hot Globally". There is much for those working on mitigation and adaptation.

Thursday, February 1, 2024

Designing Social Isolation Out of University Programs

Carmen Vallis, University of Sydney
In "Helping Students Overcome Social Isolation at University", Carmen Vallis, University of Sydney, points out the Higher Education Support Amendment Bill 2023 requires universities to provide students with the support they need. They suggest one-off or ongoing intervention may be needed for individual students suffering social isolation, and not just those from underrepresented groups.

Rather than treating social isolation as a problem to be diagnosed and cured, I suggest designing university programs and courses to incorporate teaching of social skills and group activities. Rather than treating students from underrepresented groups as an exception to what a normal student is, incorporate into the program what the students need to learn.  This is not about putting tennis on the curriculum, but recognizing that students learn better in a group, and will be required to work with other people when they graduate. The skills of teamwork are central to any profession.

As a graduate student I experienced social isolation. This was because I was an online international student, studying a different discipline with people who were a different gender from a different culture. What helped were group exercises, and peer feedback exercises, where I was reluctantly forced to interact with other students. I now teach students who are required to work in cross cultural teams. The students complain about having to do this, but they do know that after they graduate this is how they will be working.