Inclusive Education 

Physics Inclusion Award - launch meeting - 27th September 2024

Day begins with presentations reviewing Project Juno, and introducing how the PIA was co-designed.
Following this, a presentation and workshop on designing an inclusive peer assessment process, information on the Award Force platform, and sharing of local good practice.    Always nice to start a meeting with a link to the IoP code of conduct.

Lauren Crawford (independent consultant) talks about the final report on Juno, which will be published soon but currently under final review.
Video presentation from Jane Smith- slides will be made available.  Juno ran 2008-2023, with midpoint evaluations and additions (e.g. addressing bullying & harassment issues).  Challenges: workload, resourcing, uneven impact on female staff. Impact was good at local level but still issues sector-wide. Reciprocal agreement with AthenaSwan was valuable.   Notes the shift in social awareness - #MeToo movement, various aspects of government policy/legislation supporting EDI.  High awareness that Juno exists, but depressingly low awareness of what it actually means... and only 1/3 respondents report that Juno had an impact on them/their academic workplace. This is spun as positive, but clearly still insufficient.
New PIA will be 5yr rather than 3yr, with annual updates.
At least one department does NOT intend continuing with monitoring and evidencing gender data post-Juno...

This is nice: Female acad staff % is inching upwards. 16-22% cf lower/static numbers with high variance for non-engaged departments, but a lot of variation across the board.  In other subjects, female representation is 49%.  Students (UG/PGT/PGR) are also showing increases from 24-28% but doesn't really look appreciably different to non-engaged departments. [for reference, the Sheffield 1st year cohort in 24/25 is 66% mr, 32% ms/miss, 2% mx]

Principle 6 - women are less likely than men to agree that staff/students know how to report B&H, and marginalised groups are more likely to experience it/have less confidence in its handling.  Professional Conduct and Intersectionality will be key foci of the PIA.

Take home messages: good but uneven impact, some negative (workload on marginalised groups), Principle 6 still essential, more work needed.

Good audience question on the impact of schoolification in the sector - the PIA will be MUCH less admin-heavy and can identify and respect overlaps/issues in reporting/evidencing.


Physics Inclusion Award

Shanice Dunk talks about the PIA. Three levels - 0 [expression of interest] and 1-3. Online platform, four themes for evidencing, and a 5yr portfolio approach with annual updates.
Themes are:

Professional conduct and intersectionality embedded throughout.  Levels are [1] establishing structures to support all and develop knowledge of EDI issues  [2] Embedding the learning/strengthening EDI knowledge [3] implementing leading practice and sharing good practice.

11 departments were engaged in the pilot scheme. Really hard to get negative feedback out of them!

Feedback on the pilot:

Current work: mapping onto AthenaSwan and getting reciprocal agreements/add-ons in place.

Nice messaging that PIA's approach to inclusivity is not restricted to the protected chars of the equality act: e.g. Ireland's Cultural Heritage characteristics, Social Class, Neurodiversity. The assessment in PIA will be contextualised to local circumstances/needs.


Peer Assessment Processes

Assessment Timeline for the PIA - first panel in Jan2025.  Process is designed to be more streamlined and less work for depts/assessors. Panel of assessors will have clear remit/requirement for embedded diversity in members and draw on external experts as needed. Panel members will be trained & upskilled, with regular review/feedback from panel and applicants to allow continual improvement of the PIA processes.  

2x Test Panels were created, with a fair amount of collective re-calibration of grading/feedback/intentifying the key descriptors. Still some ongoing work on whether a mix of L1/L2 is at L1, L2, or in between, and if there are red flags that can hold an application back.


Application key dates/Online Platform:

March 2025 panel - expressions of interest Nov 11- Dec 1, notified by Dec 6, submit by Feb 16, feedback in May. Sep 2025 panel has EoIs in June 3-23, notification by June 28, applications in by Jul 28 with feedback in Oct.
Departments can start with the EDI self assessment tool to get ahead of the PIA cycle - priority will be given to Juno expiry departments.

The platform is the same as that used for the IoP's awards applications.  Additional features include good practice sharing gallery, and a noticeboard for comms. It's worth engaging with it at L0 to start on the galleries/evidencing processes. Looks like a really good way to share/explore good practice. Have suggested that they set it up into curated themes to allow useful/easy navigation of resources.

If depts. have an existing AS award, they submit that status and the platform will filter out questions that are already covered within the scope of that.


Good Practice in MPS

Matt gave an update on good practice in Sheffield

Faculty of Science EDI&Wellbeing Away Day - 11th September 2024

Welcome from Prof Robert Mokaya [Provost and DVC]

Robert's research group will be moving from Nottingham to Chemistry within the School of MPS - nice to know we have a new DVC who knows the Chemistry discipline and the challenges they face. Keeping a high profile for Chemistry within the new schools structure is something that a lot of us are keen to see. Robert flags up the new staff survey coming up next month - our lived experiences are valued, and I'm sure people will be vocal on that front.  Emphasises the value of diverse research and educational cultures.  Governance review of EDI leadership has led to changes - UEB sub committee, chaired by Robert, looks at delivery of our EDI work and future strategy. Council EDI committee will provide oversight and scrutiny of our work. Will allow tUoS to streamline/build consistency.  Emphasis that the consistency goal is a race to the top, not a race to the bottom. Notes Athena Swan silver and submission for bronze Race Equality Charter, but awards are not as valuable as the practice on the ground and the impact on lived experiences. Working to raise awareness through staff training and other routes.  Institutional driver to address pay gaps.

Updates from Departments/Schools/Teams


Staff Race Equality network updates

With speakers Betty Anyika and Mohammad Rajjaque.

Value of networks: should be led by staff, for staff. Provide a safe space for colleagues and enable critical input into policy design, implementation and supervision. Flagging the networks for parents, race equality, women, LGBT+, disability, belief/no belief, carers. [Aside: the Disability Staff Network has... history. I have social/mental scars from the last 5+ years of how that network has been supported and how its concerns have or have not been addressed. Good work being done there, but also some past issues that need to be learned from.]

Policy Advocacy and Impact: revamping the regrading process for PS staff to limit routes for bias, providing input on estate policy and design for physical accessibility, accommodating spaces, improving accessibility of location for occyhealth and safety centre. Also a lot of work for Athena Swan and REC. Network chairs represented at tUoS EDIC.

Networks: very broadly intersectional, and also represent the majority of staff members

Race Equality at Sheffield:
Racial inequality is recognised as a fact. Diversity is limited in upper management and PS leads. We have a range of formal processes for raising concerns, but their impact is questionable. [Aside: anecdotally, I have a sense that many colleagues feel the same is true of other issues that get Report&Supported.] Challenges: recruitment and progression, where there is data indicating systemic bias in outcomes. [Aside: my survey of SRDS experiences in physics indicated poor prior experience across all marginalised demographics, esp. those multiply-marginalised.]

Recurring themes: 

Quite obviously, when you get intersectional, this becomes a nightmare for individuals.

Discussion questions: (we got Q2)
Your line manager and immediate colleagues should be primary sources of information and guidance/. Elevate, research support team, other research networks et. are secondary sources. What do your experiences suggest? What works?

Table response: immediate colleagues and some line managers are well intentioned but the degree to which they're well informed is uneven across the university. Reliance on networks and ad-hoc groups of peers to access resources/signposting - this has a problem of in/out group access.

The Ethics of AI use in Education

First question - how do we use (or not use) it, and how confident are we in talking about AI to friends/family, colleagues and students. Mixed views, but generally we're less confident discussing it with students.
In practice: AI fuels stereotypes, and is subject to the foibles of its training data. Users should be aware of providing appropriate prompts to dismantle inherent biases.

There are differences between AI and GenAI. Be clear that GenAI is only a subset of AI more broadly.

tUoS uses Google Gemini as an approved tool. The reason for this is to provide equity of access to GenAI as a tool.  Gemini (formerly Bard) is improving as a tool [Aside: let's not think too hard about how]. 

tUoS is taking a positive approach to GenAI in learning and teaching, recognising the limitations of these tools, and able to engage critically with the tools and their outputs. We prioritise fair access and understand the ethical issues surrounding data, privacy, intellectual property. We are providing clarity in expectations for use to avoid issues of failings in academic integrity. Universities will work collaboratively to share best practice as the technology/educational applications evolve.  Using an acknowledge/describe/evidence approach to GenAI use.

Data imported into Gemini within the institution is not used as training data for Google.

Key considerations:

AI is not neutral. Hoffman et al (2024) identify overt and covert racism. UNWomen has a report of interest: "Who creates AI and what biases are built into AI data (or not) can perpetuate, widen or reduce gender equality gaps"

Users of GenAI in last 12 months skew male (50%) cf female (37%). Based on differences in awareness? Also some indication of ethnicity differences in GenAI use in chemistry (Potts, UCL) [Aside: I hesitate to refer to this as a 'gap'] However, student takes on GenAI are much more nuanced and aware of the ethical considerations.

Nice quote from the Algorithmic Justice League:

 
“Equitable AI requires that people have agency and control over how they interact with an AI system. To have agency, people must first be aware of how these systems are used all around them — for example, at airports, stadiums, schools, hospitals and in hiring and housing — who is involved in creating the system — from business, government and academia — and the risks and potential harms.” 

Disability and AI - there are some interesting aspects here. I'm taking the text below from the slides from Jenny, Rob and Sophie.


Do take a look at some of the responses in social media to NaNoWriMo (below - Sarah Gailey's article is great) regarding intersection of disability with GenAI use - it's worth hearing.

We're discussing limitations and opportunities. Our table is tremendously negative... It's out the bag and important to engage with it as a tool, but we are very conscious of the energy/climate change implications and ethics of theft/ownership. Individually, I'm more comfortable with code training sets than language training sets given that open access code repositories are a big thing, but what this misses is the transfer/recognition of credit of creators. I'm individually uncomfortable with the GenAI art dropdowns in Blackboard Ultra - IMO this is normalising the abuse of artists. We're talking a lot about how best to support students and maintain academic integrity, and how to avoid disadvantaging particular subsets of students.  I like that (locally) we have a lot of writing support for physics students, which hopefully builds confidence.  Side-rant from me on the hubris of scientists when it comes to valuing the work of creatives (artists/writers)

The discussion has prompted me to contact the VLE team to flag up concerns with pushing GenAI art tools on academics...

Here's an interesting article: The Backwardness of Generative AI: Making Exploitation Cool Again

Below are some social media posts from Twitter and Bluesky, on the use and abuse of AI in written works.   First is reporting that T&F journals use/provide all articles to LLM training sets.  Second and third are some tweets from the fiction writing community regarding the [formerly] much-loved NaNoWriMo  novel writing community adopting GenAI sponsorship, followed by pop culture pundit, author, and voice artist Alasdair Stuart drawing on the community to identify and highlight positive usage of GenAI (if any). The voice acting  community is under similar pressures to the writing and art communities in terms of their works being stolen for training data.  The last two tweets are related to the use of AI in medical care, with a specific slant towards the impact on radiologists.

I've also come back to this to add a link to the following article on Australian news outlet Crikey: AI worse than humans in every way at summarising information, government trial finds

Good Practice Talks:

First talk: Mindset awareness workshops in biological services. Workshop began with exploring 'how are you feeling', followed by an eye-contact blame/shame exercise. [no-one was forced, but I'm getting mental hives hearing about it...] Next exercise was emotion mapping, on axes of activity-level of emotions and negativity/positivity of emotions.  Sectors were then re-labelled as Survival // Active // Burn Out // Recharge. Knowing where you are can help your mindset.

[Aside: as someone with chronic severe depression, I exist in the Burn Out sector, emotionally, with the briefest in-the-moment forays into the others.  Additionally, I feel that this type of intervention (esp eye contact, mapping of behaviours) if not sensitively handled has the potential to reinforce neurotypical modes of emotional expression, and the associated systemic disadvantage.] 

Second Talk: Jupyter Books for accessible course notes - Experiences in physics and astronomy. (Matt Mears - also prior practice from people like Alex Best in maths and many people in physics)

Some examples: PHY11006 PHY383

[and then my laptop battery ran out of juice]


Decolonising the Curriculum: Another University is Possible. 10th September 2024


Talks:


Dr Meesha Warmington: ‘Best Practice, Opportunities and Challenges to Developing Inclusive Curriculum Practices’

Meesha discussed her team's recent survey of practices across the university. Departments were surveyed with questions related to curricula content, assessments, learning environments and training/development/visibility. 11 departments completed the survey , and they took a qualitative approach to analysis of the data.

The survey responses indicated several different types of work carried out in this area: specific initiatives/resources for individual modules or learning activities; departmental strategy has a clear focus on encouraging more work in this area; looking to student engagement via focus groups/student voice/co-production to develop the curriculum; a broader focus on inclusivity/diversity throughout the curriculum.

Focusing on assessments, the survey revealed  a range of approaches to support decolonisation and greater inclusivity of the curriculum: diversification of assessment methods; constructive alignment of assessment; student centered approaches to assessment development; support workshops for assessment skills; ongoing review/redevelopment of assessments.

What's good/useful in this area?  The survey identified: time/funding/toolkits to support staff; EDI/inclusivity strategically integrated into core activities; knowledge sharing; active engagement with student voice (focus groups, co creation, student led initiatives); transparent and regular communication.

There remain a large number of challenges and opportunities:

The key recommendations from this work are to address the following: Staff workloads - knowledge gaps - inconsistencies in student experience


Tomás Rocha Lawrence, the SU Liberation Officer, spoke on behalf of the SU Education Officer on "Another University is Possible", focusing on decolonisation, decarbonisation, demilitarisation, democratisation. The Dakota Access Pipeline (DAPL) was used as an example of how these 4Ds are intrinsically linked, and to highlight systemic racism/colonialism in academia. 


Dr Emma Hughes: Decolonising the Alfred Denny Museum

The Alfred Denny Museum is a zoological museum. It is fundamentally a teaching collection, still used for practical sessions on anatomy and taxonomy, and managed by a volunteer community of staff and students who are trained in curatorial skills.   The artefacts provide a physical link between biological science and colonialism. However, there are some strong opinions in the general public related to the degree of impact/interrelation between science/colonialism. Other artefacts in the museum include human remains, which have in the past fed into racist science such as craniology. Some colonially-acquired remains have been subsequently sold on, but others have been traced and are now in the hands of experts with an eye towards potential repatriation or other appropriate disposal. Aboriginal names have been added to Australian animal remains ahead of the linnaean scientific taxonomic names. A lot of different forms of work being done here!


Dr Alex Rajinder Mason: Reimagining university partnership work with racially marginalised communities

Alex works at the Centre for Equity and Inclusion, which primarily supports PGR students of colour. The Centre is developing communities and networks of researchers, students, artists, educators and community groups experienced with racial justice. It aims to confront systemic racism in academia, and to develop and embed anti-racist policies.

Early piece of work was a collaborative visual contract project between Sheffield PGRs and Centre partners - partnership contracts are often used, but contracts have their own flaws, and the artistic lens aimed to address the systemic inequalities of a colonialistic legal framework. Aim was to apply native/indigenous values and different cosmologies of partnership.  The English language is inherently colonialistic in some sense, and not always adequate to communicate radical, non-western ideas.  The University's standard partnership agreement was reconsidered and repositioned visually, clause-by-clause.  The resulting art cards come with a text page on the obverse: description of the image, interpretation of the image, and the legal text.  The legal text provides legal rigour, but in some sense undermines the core concept of the cards as it positions the work within a wider social framework which retains systemic racial-colonial inequality.

Current use: the visual contract is applied in steering group meetings and knowledge exchange projects. It's an adaptable and versatile example of work which is decolonial, inclusive, and uplifting of the global majority. It encourages a relational and fluid approach to partnerships, and helps sidestep problematic hierarchies.

Challenges: encouraging users outside the Centre to adopt this approach; embedding it into everyday operations; ensuring users adopt the more radical/decolonial aspects of the project.

Feedback from Bluesky:

Ideas:

Decolonising Astrophysics - 21.4.2022

Slides from my talk in 2022, focusing on a continually redeveloped piece of decolonisation work in PHY127

Inskip-astro.pptx