Pre-service teacher education · Foundations of Teaching and Teacher Education
Teacher Professionalism in Local and International Contexts
This 150-minute critical inquiry seminar positions teacher professionalism as a contested, context-dependent construct rather than a fixed checklist of traits. You will interrogate four competing discourses — trait-based, managerial, democratic and activist — compare how professionalism is framed in your own jurisdiction and in a contrasting international context, and test whether "global best practice" can travel. AI is integrated throughout as an object of critique, not an oracle: you will use it to generate comparative accounts, then systematically test those accounts against authoritative sources, identify bias and hallucination, and document where human judgement overrode the machine.
Professionalism is treated as something people argue about, not something a document settles. Every claim you meet — from a ministry, a union, a code of conduct or an AI assistant — is read as a positioned account with interests behind it. Your job is to surface those interests, weigh the evidence, and be willing to move.
Work top to bottom. Tick tasks as you complete them; your progress, quiz answers, critique verdicts and reflections are saved in this browser only. Use Export portfolio at the end to keep a copy for your summative submission.
Duration 150 minutes · first-year undergraduates · 90 students in 15 groups of six · English · Critical Inquiry Seminar with Social Constructivist and Scaffolded Instruction elements · AI integration in ILOs, activities, assessment and feedback. If your course prefers no AI integration, the AI-annotated elements can be removed without altering the core design.
Section 1
Intended Learning Outcomes
Seven outcomes, each paired with the AI integration engineered into it. The first five build the disciplinary argument; outcomes 6 and 7 make critical AI literacy an assessable outcome in its own right rather than a bolt-on.
| ILO | Description | 🤖 AI Integration |
|---|---|---|
| ILO 1 | Analyse competing conceptualisations of teacher professionalism (trait-based, managerial, democratic, activist) and surface the assumptions and interests underpinning each. (Bloom's: Analyze · Critical Inquiry: assumption-surfacing) | In the introduction, students critique an AI-generated paragraph and identify which discourse the AI implicitly assumes; the AI Critique Lab makes that assumption-surfacing explicit. |
| ILO 2 | Compare and evaluate how teacher professionalism is framed in the local jurisdiction and in at least one contrasting international context, using policy documents, codes of conduct and professional standards as evidence. (Bloom's: Evaluate · Critical Inquiry: evidence appraisal) | AI generates the initial comparison in Segment B; students then verify every claim against a named, human-authored source before it can enter their argument. |
| ILO 3 | Critique the transferability of "international best practice" in teacher professionalism, accounting for historical, cultural, political and economic context. (Bloom's: Evaluate · Critical Inquiry: contextual reasoning) | Students identify the AI's Anglocentric bias, its over-generalisation and its erasure of genuine contextual conflict in the comparative task. |
| ILO 4 | Construct and defend a reasoned, evidence-based position on a contested issue in teacher professionalism within structured seminar dialogue, and revise that position in response to counter-arguments. (Bloom's: Create · Critical Inquiry: dialogic position defence) | The portfolio reflective commentary asks where and why human judgement took precedence over an AI recommendation in building that position. |
| ILO 5 | Collaborate in an inquiry community to interrogate evidence, challenge peers' reasoning constructively, and co-build a shared claim with warrants. (Bloom's: Analyze / Evaluate · Social Constructivist element) | Groups jointly interrogate AI output and must negotiate a shared verdict, so the critique is a collective product rather than a private opinion. |
| ILO 6 | Evaluate AI-generated accounts of teacher professionalism for accuracy, completeness, recency, cultural bias and hallucinated evidence — identifying at least two specific errors, omissions or Anglocentric framings. (AI literacy: critical evaluation of AI outputs) | Core AI outcome Assessed directly through the Critical AI Engagement Log and Portfolio Part B; at least one hallucination, bias or omission must be documented. |
| ILO 7 | Exercise independent judgement in deciding when, whether and how to act on AI-generated content, documenting what was accepted, modified or rejected and articulating where human expertise and contextual knowledge took precedence. (AI literacy: independent judgement and transparency) | Core AI outcome Assessed through the AI transparency note in Segment D, the Feedback Interrogation Sheet and the "Critical AI Engagement" rubric criterion. |
AI output is treated as a hypothesis to be tested, never a source to be cited. Every AI-derived claim must be verified against a named, human-authored source before it enters your argument. The lesson deliberately engineers situations in which the AI is fluent, confident and wrong — because fluency is a stylistic property, not evidence of correctness.
The session moves from analyse (ILO 1) through evaluate (ILOs 2–3, 6) to create (ILO 4) — a deliberate progression from describing discourses to defending a position.
Assumption-surfacing → evidence appraisal → contextual reasoning → dialogic position defence. The same spine recurs in every segment, so the intellectual moves become habitual.
ILO 5 is assessed formatively through the observation protocol and peer checklist, not in the individual portfolio — preserving individual accountability while valuing collaboration.
Section 2
Pre-Class Preparation
Approximately 90 minutes of inquiry priming: core readings, a collaborative annotation task, an adaptive diagnostic pre-test and an AI engagement task. This is framed as priming rather than a full flipped model — the seminar does the heavy lifting.
- Hargreaves, A. (2000). Four ages of professionalism and professional learning. Teachers and Teaching, 6(2), 151–182. read the article
- Sachs, J. (2001). Teacher professional identity: Competing discourses, competing outcomes. Journal of Education Policy, 16(2), 149–161. read the article
- UNESCO/ILO (1966). Recommendation concerning the Status of Teachers. read the Recommendation
- Local source (select one): Hong Kong EDB Guidelines on Teachers' Professional Conduct EDB guidelines · UK DfE Teachers' Standards DfE standards · AITSL Australian Professional Standards for Teachers AITSL standards
- International source (select one): OECD TALIS results TALIS indicators · Ontario College of Teachers Ethical Standards OCT standards · Education International Education International
- Collaborative annotation: annotate Sachs (2001) with the four tags below and reply to one peer — see the Hypothesis instructions.
- Diagnostic pre-test: eight adaptive items, completed before class.
- AI engagement task: run the prompt below and complete the Pre-Class AI Observation Log.
Your pre-test result, your two Hypothesis annotations, and your completed AI Observation Log — the seminar begins from what the pre-test reveals, not from the readings alone.
Annotate the Sachs (2001) chapter with four tags. Each student posts at least two annotations and replies to one peer's annotation before class.
[ASSUMPTION]— an unstated premise[EVIDENCE]— an empirical or documentary warrant[CONTESTED]— a claim you believe is arguable[CONTEXT]— a claim that may not travel across contexts
Open Hypothesis collaborative annotation and follow the numbered steps below.
- Who gets to define what counts as "professional" behaviour for teachers — the state, the profession, the market, or the community?
- Whose interests are served by each definition?
- What evidence would change your mind about whether teaching is a profession?
- What is lost when a definition of professionalism travels from one context to another?
Keep these questions open as you read — the tribunal in Segment C will return to them.
Prompt an AI assistant (ChatGPT, Claude, Copilot or Gemini) with the following request, then complete the Pre-Class AI Observation Log and submit it to the LMS before class.
Complete the observation log:
- Paste the AI's response.
- Identify one claim you believe is accurate and explain how you would verify it.
- Identify one claim that is inaccurate, oversimplified or culturally biased, and explain why.
- Identify one thing the AI omitted that your readings suggest matters.
- Check any citations the AI provided: do they exist? Do they say what the AI claims? Hallucinated citations are common — this is a deliberate teaching moment.
- Reflection prompt: "Did the AI's answer shape what you think professionalism is? If so, how will you guard against that?"
Tool Run the adaptive pre-test in your LMS (Blackboard / Moodle / Canvas)
Purpose: to surface the three target misconceptions before teaching, so the session can be aimed at them rather than at assumed gaps. Learning objective: activates prior knowledge for ILO 1 and generates the aggregate data used in the pre-test review.
- In your LMS course, open Quizzes (Blackboard) / Quiz activity (Moodle) / Quizzes (Canvas) and create a new quiz titled "Diagnostic: Discourses of Teacher Professionalism".
- Add the eight items, tagging each with a misconception category so the report can be read by category: professionalism = compliance, best practice is universally transferable, and professionalism is a fixed, ahistorical trait list.
- Enable branching / adaptive feedback: for a wrong answer on a discourse-recognition item, show a short remediation note pointing to the relevant section of Sachs (2001).
- Set the attempt limit to one, allow a 15-minute window, and turn off the timer display for students with registered accommodations.
- Publish the quiz to the cohort and confirm the item statistics are visible to instructors only.
- Before class, export the item analysis report and note which categories scored lowest — you will display these aggregates in the pre-test review.
- Do not release correct answers to students; the same categories are re-tested in the post-test.
Tool Set up the Hypothesis collaborative annotation on Sachs (2001)
Purpose: to make assumption-surfacing a visible, social act before the seminar. Learning objective: ILO 1 (assumption-surfacing) and ILO 5 (collaborative inquiry), by requiring students to tag and respond to each other's reading of the same text.
- Create a free account at Hypothesis and install the browser extension, or use your LMS's built-in Hypothesis integration.
- In your LMS, create a Hypothesis assignment and attach the Sachs (2001) reading as the target PDF or web page.
- Set the assignment group to "course group" so all 90 annotations are visible to the cohort, and set the deadline to 24 hours before class.
- Publish the four tag conventions to the class:
[ASSUMPTION],[EVIDENCE],[CONTESTED],[CONTEXT]. - Post a worked model annotation yourself — for example, tag a sentence from Sachs's abstract as
[CONTEXT]and explain in two sentences why it may not travel. - Require each student to post at least two annotations and to reply to one peer; use the Hypothesis activity dashboard to check participation.
- Before class, filter the annotations by tag and pull three
[CONTESTED]examples to open the seminar with.
Tool Run the pre-class AI task with ChatGPT, Claude, Copilot or Gemini
Purpose: to generate a concrete, checkable AI account for students to interrogate. Learning objective: ILO 6 — evaluating AI output for accuracy, bias and hallucinated evidence.
- Choose one assistant — ChatGPT, Claude, Copilot or Gemini.
- Paste the prompt exactly as written in the task above, replacing the bracketed jurisdiction with your own.
- Copy the full response into the "paste the AI's response" field of the Pre-Class AI Observation Log.
- For each citation the AI gives, search the title and author in your library catalogue or a search engine. Record whether it exists and whether it says what the AI claims.
- Highlight one accurate claim, one biased or oversimplified claim, and one omission.
- Answer the reflection prompt in two or three sentences.
- Submit the completed log to the LMS before class; bring a copy to the seminar for Segment B.
Section 3
Teaching & Learning Activities
150 minutes total — Introduction 20 min (13%) · Development 105 min (70%) · Synthesis and closure 25 min (17%). AI touchpoints are marked throughout.
Hook — "The Professional or the Bureaucrat?" 5 min
Two contrasting artefacts are displayed side by side.
An extract affirming that teachers should enjoy academic freedom and that teaching should be recognised as a profession exercising professional judgement in the interests of students and society.
An extract from a contemporary policy specifying lesson-plan templates, scripted curricula and test-score targets tied to appraisal.
Ask: "Both claim to be about teacher professionalism. Can both be right?" Students vote via Mentimeter before any discussion. The split is recorded and revisited at closure.
Pre-test review with misconception analysis 6 min
Display the aggregate pre-test results. Explicitly name the three target misconceptions and ask students to locate where in their readings each is challenged. Use a think-pair-share (60 seconds each) before a whole-class debrief.
- (a) professionalism = compliance with standards
- (b) "best practice" is universally transferable
- (c) professionalism is a fixed, ahistorical trait list
AI-generated claim critique — modelled critical questioning 6 min
The instructor displays a pre-screened AI-generated paragraph on teacher professionalism containing two deliberate flaws: a hallucinated citation and an Anglocentric generalisation (for example, "all high-performing systems now require teachers to hold a master's degree"). The instructor then models the critical questioning ladder aloud:
| Rung | Question |
|---|---|
| 1 | What is the claim? |
| 2 | What evidence is offered — and does that evidence exist? |
| 3 | Whose context is being generalised from? |
| 4 | What is missing? |
| 5 | What would I need to check before using this? |
Students then apply the ladder in pairs to a second AI-generated sentence. This models the exact protocol used in Segment B.
Real-world connection and roadmap 3 min
Frame the stakes: teacher shortages, migration of teachers across borders, debates over professional registration, and the global circulation of "what works" in education. Preview the session arc: map the terrain → compare contexts with AI → tribunal → position statement.
Interactive mini-lecture with think-pair-share 10 min
Present the four discourses using an animated concept map. Pause after each discourse for a 60-second think-pair-share: "Which discourse does your own schooling experience most resemble?"
| Discourse | Core emphasis | Key proponents / sources | Primary risk |
|---|---|---|---|
| Trait-based / classical | Knowledge base, autonomy, ethical code, service ideal | Flexnerian tradition; UNESCO/ILO (1966) | Idealised; ignores context |
| Managerial | Standards, accountability, measurable outcomes | Sachs (2001); performativity literature | Deprofessionalisation |
| Democratic | Collegiality, professional learning communities, trust | Hargreaves (2000); Sachs (2001) | Can be co-opted by management |
| Activist / transformative | Social justice, equity, advocacy, community engagement | Critical pedagogy traditions | Marginalised in policy |
Group card-sort task 10 min
Each of the 15 groups receives a set of 12 statement cards drawn verbatim from real codes, standards and policy documents (anonymised). Groups classify each card into one of the four discourses and mark any card that is contested — that is, a card that could plausibly belong to two discourses. Each group must justify one contested classification to a neighbouring group.
Assumption archaeology within Segment A
Each group receives one policy extract and answers four questions in writing, then exchanges extracts with a neighbouring group and compares assumptions.
- What does this document assume a teacher is?
- What does it assume a teacher should be trusted to do?
- Whose voice is absent from this document?
- What would a teacher in a very different context find strange about it?
Poll: "Which discourse best describes teacher professionalism in your local context today?" Live results are displayed. The instructor facilitates a brief discussion of divergence — particularly where students from different school backgrounds disagree. This divergence is itself the lesson: professionalism is experienced differently across settings.
Structured AI interaction task 20 min
Groups of six work in three sub-pairs, each with a distinct AI prompt.
Each sub-pair then completes the Critical AI Engagement Log using this template (available in Google Docs or Sheets):
| Column | Prompt |
|---|---|
| AI claim | Quote the specific claim |
| Verification source | Which reading, policy document or database did you check it against? |
| Verdict | Accepted Modified Rejected |
| Rationale | Why? What was wrong, missing or contextually inapplicable? |
| Human override | What would a knowledgeable human say instead? |
Each group must document at least three verdicts, including at least one rejection and at least one hallucinated or unverifiable citation.
AI verdict negotiation within Segment B
The full group compares the three AI outputs and discovers that the AI produces inconsistent claims depending on how the question is framed — a powerful demonstration that AI output is prompt-dependent and not authoritative. Groups document the inconsistencies in their log and discuss: "If the AI says different things depending on how we ask, what does that tell us about using it as a source?"
Instructor mini-briefing on AI limitations 5 min
- Hallucination: AI generates plausible but non-existent citations and statistics. Always verify.
- Anglocentric and WEIRD bias: training data over-represents English-language, high-income, Western sources. "International" often means "US, UK, Australia".
- Averaging: AI tends to produce a smooth consensus that erases genuine contextual conflict.
- Recency: policy changes fast; AI knowledge has a cutoff.
- Authority laundering: fluent AI prose can feel authoritative even when unsupported.
AI output is a hypothesis to be tested, never a source to be cited. Every AI-derived claim must be verified against a named, human-authored source before it enters your argument.
Cross-check against the Source Pack 10 min
Groups verify their AI claims against the provided Source Pack (printed and digital): the UNESCO/ILO (1966) Recommendation, OECD TALIS indicators, the local code of conduct, and one contrasting national standards document. Groups annotate the pack in the margin: "AI said X; source says Y."
"This house believes that teacher professionalism is best secured through external accountability standards rather than professional self-regulation."
Stakeholder role allocation
One role per group, with two groups sharing the largest roles.
State regulator: defends external standards and public accountability.
Defends self-regulation, professional judgement and collective standards.
Caught between compliance and the conditions for good teaching.
Asks what transparency and trust mean for families and students.
Brings evidence on standards, learning and policy transfer.
Must defend an AI-generated position on the motion, then publicly concede where the AI's reasoning fails under cross-examination — a deliberately designed critical AI literacy role.
Format
Instructor facilitation moves
- "What is your evidence, and how do you know it?"
- "Whose interests does that definition serve?"
- "What would have to be true for the opposing view to be correct?"
- "Is that claim context-specific or general?"
Students may, as an accessibility accommodation, submit a written tribunal statement in place of an oral opening statement.
Position-shift vote
Students re-vote on the motion. The instructor highlights shifts and asks students to name the argument that moved them — making position revision an explicit, valued outcome rather than a failure of consistency.
Individually, draft a 150-word position statement on the tribunal motion. The instructor first models one aloud, thinking through the choice of evidence and the AI transparency note; you then draft your own using the structured template.
Requirements
- At least one piece of evidence from a local source and one from an international source.
- At least one counter-argument acknowledged and rebutted.
- An explicit AI transparency note: "One AI-generated claim I [accepted / modified / rejected] was ___, because ___."
This draft feeds directly into the summative portfolio (Part C).
Sentence starters
"My claim is ___ because the evidence from ___ suggests ___" · "The strongest objection is ___, but this fails to account for ___" · "One AI-generated claim I rejected was ___, because the source said ___" · "I want to challenge that because ___"
Peers exchange drafts for a two-minute "evidence check" before submission.
Check each claim against your readings, the Source Pack and your own reasoning. A confident tone is not evidence of correctness. Use Accept for claims that survive checking, Modify where a claim is partly right but over-generalised, and Reject where it is false, hallucinated or contextually inapplicable.
Pick a prompt to see a simulated AI response — each one is fluent, confident and flawed in a specific way. Then write your critique using the five-rung ladder.
Post-test 5 min
Re-administer the eight pre-test items plus two transfer items (9 — apply; 10 — evaluate). Complete it in Section 4.
Peer-teaching summaries — claim-and-warrant wall 8 min
Each group posts a single "claim + warrant" sentence to a Padlet wall: "We claim that ___, because ___." The class then reads the wall and identifies (a) the best-warranted claim and (b) the most contested claim. The instructor synthesises the pattern, highlighting where warrants were strong and where claims rested on assertion.
This makes the evidentiary standard of the summative portfolio visible before it is assessed. Learning objective: ILO 5 — co-constructing a warranted claim in an inquiry community.
AI reflection discussion 7 min
- "How did AI support or hinder your inquiry today?"
- "What did the AI get wrong that you caught? How did you catch it?"
- "When did you need to override an AI suggestion, and what told you to?"
- "Did anyone find themselves agreeing with the AI before checking? What does that tell us?"
The instructor records recurring AI failure modes on the board — these become revision targets for the next iteration of the lesson.
Preview of next session 5 min
Next session: Teacher Identity, Professional Learning Communities, and Career-Long Professional Development. Bring one example of a professional learning community from your local context and one from an international context, and note one AI claim about professional learning communities that you intend to verify.
- Frame AI as evidence to examine, not an authority to copy. State the session rule aloud: AI output is a hypothesis to be tested, never a source to be cited.
- Model the questioning ladder before you ask students to use it. Students imitate the intellectual moves they see, not the ones they are told about.
- Require at least one rejection. A log in which every AI claim is "Accepted" is treated as evidence of unreflective reliance, not of a good AI.
- Name the failure modes explicitly — hallucination, Anglocentric bias, averaging, recency, authority laundering — so students have vocabulary for what they notice.
- Reward position revision. Publicly value students who change their mind with reasons; otherwise the tribunal rewards performance rather than inquiry.
- Keep human judgement sovereign. Every AI-derived claim must be traced to a named, human-authored source before it enters an argument.
Tool Run the opening, discourse and position-shift polls in Mentimeter
Purpose: to make the distribution of opinion visible before, during and after argument, so that movement becomes measurable. Learning objective: ILO 1 and ILO 4 — surfacing assumptions and evidencing position revision.
- Sign in at Mentimeter and create a new presentation called "Teacher Professionalism — Tribunal".
- Add slide 1 as a Multiple choice poll: "Teacher professionalism is best secured by external accountability standards rather than professional self-regulation" with Agree / Disagree / Unsure.
- Add slide 2 as a Multiple choice poll for the Segment A discourse question: "Which discourse best describes teacher professionalism in your local context today?" with the four discourses as options.
- Duplicate slide 1 and rename it "Position-shift vote" so the closing vote can be compared directly with the opening vote.
- Set voting to audience pace and enable "show results live" so the split appears on screen as students vote.
- Display the QR code and six-digit code; students vote on their phones without creating an account.
- Run the three polls at the points marked in the run sheet, and screenshot the opening and closing results so the shift can be shown and discussed.
- Ask students who changed position to name the argument that moved them — record these on the board.
Tool Build the claim-and-warrant wall in Padlet
Purpose: to turn each group's inquiry into one public, checkable claim so the whole cohort can compare warrant quality. Learning objective: ILO 5 and ILO 4 — co-building a claim with warrants and identifying the most contested claim.
- Sign in at Padlet and create a new Wall board titled "We claim that… because…".
- Set the posting permission to "Anyone with the link can write" so all 15 groups can post without accounts, or use the Padlet backchannel in your LMS.
- Enable comments and likes/reactions so the class can signal the best-warranted and most contested claims.
- Post an example card yourself: "We claim that managerial professionalism is expanding, because appraisal policies in our Source Pack tie progression to test-score targets."
- Share the QR code with the class and ask each group to post exactly one sentence of the form "We claim that ___, because ___."
- Allow five minutes of silent reading, then ask students to react to the strongest warrant and to the claim they most dispute.
- Project the wall and synthesise the pattern: where warrants cited sources, and where claims rested on assertion.
Tool Capture the 3-2-1 exit ticket with Google Forms
Purpose: to collect an immediate, private read on what students can now defend, what tensions remain, and what they want to pursue. Learning objective: consolidates ILOs 1–4 and generates the evidence base for the reflection section.
- Open Google Forms and create a form titled "Exit ticket — Teacher Professionalism".
- Add three Paragraph questions: "3 claims you can now defend with evidence"; "2 tensions you still see"; "1 question you want to pursue".
- Add a fourth question as Multiple choice: "Which AI task today was most valuable?" with the Segment B comparison, the critique lab, and the simulated chat as options.
- Turn on Collect email addresses only if you need to follow up with individuals; otherwise leave it off to encourage candour.
- Under Settings → Responses, enable "Show summary charts and text responses" so students can compare their answers with the cohort.
- Link the form in your LMS and keep it open for the final five minutes of the session.
- After class, read the "2 tensions" column first — it tells you what to revisit next session.
Section 4
Assessment Methods
Formative assessment runs throughout the seminar; summative assessment is a single comparative inquiry portfolio with an explicit critical AI engagement criterion.
The post-test mirrors the pre-test misconception categories — discourses of professionalism, comparative and contextual reasoning, and critical AI literacy — and adds two transfer items: item 9 (apply) asks you to analyse a scripted-lesson-plan policy through the four discourses, and item 10 (evaluate) asks why Finnish approaches might not transfer to your local context.
Critical Inquiry Observation Protocol
The instructor and two teaching assistants circulate with a structured observation sheet capturing:
- Quality of evidence use — does the student cite sources or assert opinions?
- Quality of questioning — does the student probe assumptions?
- Willingness to revise a position in light of counter-evidence.
- Distribution of participation across group members.
Real-time polling and exit ticket
- Mentimeter polls at three points: opening vote, discourse poll, position-shift vote.
- Exit ticket (3-2-1): 3 claims you can now defend with evidence · 2 tensions you still see · 1 question you want to pursue.
Peer-assessment checklist with calibration
Before the tribunal, the class calibrates on a 30-second anonymised video clip of a seminar exchange, scoring it against a four-item checklist — evidence, reasoning, responsiveness, clarity. This calibration ensures inter-rater consistency before peer scoring begins.
AI-supported formative checks
- Students submit their AI-generated comparison alongside their annotated critique and Critical AI Engagement Log.
- The instructor reviews the logs specifically for evidence of critical engagement versus unreflective reliance — a log where every AI claim is "Accepted" with no verification triggers a targeted check-in.
- Spot-check: the instructor selects three AI claims from across the cohort and asks the class to verify them live in the following session.
Individual, 1,500 words plus appendices, due Week 6.
| Part | Length | What it requires |
|---|---|---|
| Part A — Comparative analysis | 500 words | Analyse how teacher professionalism is framed in the local jurisdiction and one contrasting international context, drawing on at least four sources including at least one policy document and one peer-reviewed source. |
| Part B — Critical evaluation of AI-generated content | 400 words + appendix | Present the Critical AI Engagement Log with at least three documented instances of accepting, modifying or rejecting AI-generated content, each with verification source and rationale. Include at least one identified hallucination, bias or omission. |
| Part C — Position statement | 400 words | Defend a position on a contested issue in teacher professionalism, with counter-argument and rebuttal. |
| Part D — Reflective commentary | 200 words | Articulate where and why human judgement took precedence over AI recommendations in your work. |
| Part E — Appendix | — | Full AI interaction transcripts, prompts used, and the Pre-Class AI Observation Log. |
| Criterion (weight) | Exemplary (A) | Proficient (B) | Developing (C) | Inadequate (D/F) |
|---|---|---|---|---|
| Conceptual understanding of teacher professionalism (25%) | Accurately distinguishes and critically deploys all four discourses; identifies tensions between them. | Distinguishes discourses accurately with minor imprecision. | Describes discourses but conflates some; limited critical framing. | Treats professionalism as a fixed trait list; major conceptual errors. |
| Comparative and contextual analysis (20%) | Comparison is genuinely contextual — historical, political and cultural factors are analysed, not listed. | Solid comparison with some contextual reasoning. | Comparison is largely descriptive; context treated superficially. | Comparison is a list of surface features; no contextual reasoning. |
| Evidence and source quality (15%) | All claims warranted by named, verified, high-quality sources; source selection is judicious. | Most claims well-warranted; sources appropriate. | Some claims unsupported; sources of uneven quality. | Claims unsupported or sourced only from AI output. |
| Critical AI engagement (20%) | Evaluates AI outputs rigorously; identifies specific inaccuracies, biases and omissions; documents accept/modify/reject decisions with clear, defensible rationale; articulates where human judgement took precedence. | Evaluates AI outputs with specific examples; documents decisions with adequate rationale. | Evaluation is general ("the AI was a bit wrong"); rationale is thin or post hoc. | Accepts AI output uncritically, or rejects it without reasoned evaluation; no documentation. |
| Argumentation and position defence (10%) | Position is clear, nuanced, and defended against the strongest counter-argument. | Position is clear and defended against a counter-argument. | Position stated but weakly defended; counter-argument tokenistic. | No clear position; no engagement with counter-arguments. |
| Transparency and academic integrity (10%) | Full, precise disclosure of AI use with prompts and transcripts; all AI-derived claims verified and attributed to human sources. | Clear disclosure; verification evident. | Disclosure incomplete or vague. | Undisclosed AI use, or AI output presented as a source. |
Collaboration quality (ILO 5) is assessed formatively through the observation protocol and peer checklist, not in the individual portfolio, to preserve individual accountability.
Tool Configure a rubric-loaded AI feedback assistant (Stage 1)
Purpose: to provide fast criterion-referenced formative feedback that students then interrogate, rather than to grade. Learning objective: ILO 7 — exercising independent judgement about when and how to act on AI output.
- Create a custom GPT (or an LMS-integrated assistant) and paste the six summative rubric criteria from the table above into its instructions.
- Add an instruction that the assistant must give feedback only against the criteria, must quote the student's own words, and must never invent sources.
- Add a second instruction requiring the assistant to state, for each comment, whether it is about argument structure, evidence use or clarity.
- Test the assistant on three sample drafts and note where it produces generic or inaccurate comments — these become the examples you use in the Week 5 calibration session.
- Set up an instructor review queue so that no AI comment reaches a student before you have annotated or removed it (Stage 2).
- Distribute the Feedback Interrogation Sheet alongside the released feedback and require students to complete it before acting on any comment.
- Record every student override, with rationale, and reward defensible overrides under the "Critical AI Engagement" criterion.
Tool Use Kahoot as a low-stakes misconceptions check
Purpose: to re-test the three target misconceptions competitively and quickly, and to show the class its own error pattern in real time. Learning objective: consolidates ILO 1 by making the misconceptions visible a second time, after teaching.
- Sign in at Kahoot and create a new quiz titled "Professionalism: misconception check".
- Add six True/False or Multiple choice questions built directly from the pre-test misconception targets.
- Include one deliberately tricky item about policy transfer to prompt debate, for example "If a policy works in Finland it will work here."
- Set the timer to 20 seconds per question and enable the "show the class answer distribution" option.
- Host in Classic mode or assign Challenge mode if you want it completed asynchronously.
- After each question, pause and ask one student to justify the correct answer with a source — the discussion, not the score, is the point.
- Share the report with the class and compare the error pattern with the pre-test aggregate.
Section 5
Constructive Alignment Matrix
Every outcome is matched to the activity that develops it, the assessment that evidences it, the pedagogy that structures it and the AI integration that supports or complicates it.
| ILO | Learning activity | Assessment | Pedagogy link | AI integration link |
|---|---|---|---|---|
| 1 | Segment A: mini-lecture, think-pair-share, card sort, assumption archaeology, Mentimeter poll | Pre/post-test items 1–4 and 9; exit ticket; observation protocol | Critical Inquiry: surfacing assumptions behind each discourse; Constructivist card-sort knowledge-building | Introduction AI-generated paragraph critique — students identify which discourse the AI implicitly assumes |
| 2 | Segment B: AI comparative inquiry and Source Pack cross-check | Portfolio Part A; Critical AI Engagement Log | Critical Inquiry: evidence appraisal across contexts | AI generates the initial comparison; students verify every claim against named human-authored sources |
| 3 | Segments B and C: Source Pack analysis, tribunal debate | Portfolio Parts A and C; post-test item 10 | Critical Inquiry: contextual reasoning; challenging policy borrowing | Students identify the AI's Anglocentric bias, over-generalisation and erasure of contextual conflict |
| 4 | Segment C tribunal; Segment D position statement drafting | Portfolio Part C; position-shift vote; peer-assessment checklist | Critical Inquiry Seminar: structured dialogic position defence and revision | Portfolio Part D: articulating where human judgement overrode AI recommendations |
| 5 | All group segments; Padlet claim-and-warrant synthesis | Observation protocol; peer-assessment checklist; Padlet wall | Social Constructivism: co-construction of warranted claims | Collaborative AI critique — groups jointly interrogate AI output and negotiate verdicts |
| 6 | Pre-class AI task; Segment B structured AI interaction; instructor mini-briefing | Portfolio Part B; Critical AI Engagement Log; Pre-Class AI Observation Log | Critical Inquiry: treating AI output as a hypothesis requiring verification | Core AI literacy outcome — at least one hallucination, bias or omission must be documented |
| 7 | Segment D AI transparency note; Feedback Interrogation Sheet | Portfolio Part D; Feedback Interrogation Sheet; rubric criterion "Critical AI Engagement" | Critical Inquiry: independent reasoning and epistemic responsibility | Core AI literacy outcome — documented accept/modify/reject decisions with rationale |
Section 6
Required Resources & Technology
Reference cards for the twelve key terms, an interactive flashcard deck, a gamified scenario quiz, a concept-map builder, and the platforms and materials the session needs.
Reference cards — the twelve key terms
Each card gives the term, its key sources, what it claims and its most common limitation. These are the concepts the flashcard deck and scenario quiz draw on.
🃏 Flashcard deck — drag-and-drop matching
Drag an item chip (or tap it, then tap a slot) onto the matching description. Correct matches lock in green. The deck covers all twelve glossary terms — the same terms you will meet in the tribunal and the portfolio.
Item chips
Descriptions
🎮 Gamified scenario quiz
Identify which concept best explains each scenario, then answer the peer-review prompt that appears after you choose. These scenarios are drawn from real policy and classroom situations.
🧠 Concept-map builder Summative artefact
Choose a scenario, add at least three nodes, and record how each concept applies. If any node came from AI-generated content, tick the disclosure box — the map feeds the portfolio's critical AI engagement evidence.
Add at least three nodes.
🔧 Platforms and technology
Hosts the readings, the adaptive pre-test, submission of the Critical AI Engagement Log and the final portfolio. See the LMS instructions in Section 2.
Used for the four-tag annotation of Sachs (2001) before class. Full steps are in Section 2.
Runs the opening vote, the discourse poll and the closing position-shift vote. Full steps are in Section 3.
Collects one warranted claim per group at closure. Full steps are in Section 3.
Shared Critical AI Engagement Logs and the 3-2-1 exit ticket. Exit-ticket steps are in Section 3.
Used for the pre-class task and the three Segment B prompts. Prompt steps are in Section 2.
- UNESCO AI Competency Framework for Teachers — read the framework
- AI Incident Database — real-world examples of AI failure — browse the incident database
- Instructor-prepared one-pager: "AI Failure Modes in Education Policy"
- Education International — teacher-union perspectives on professionalism
- Source Pack (printed, 15 copies): UNESCO/ILO (1966) Recommendation extract; OECD TALIS indicator summary; local code of conduct; one contrasting national standards document.
- Card-sort sets (15 sets of 12 laminated statement cards, colour-coded for accessibility).
- Critical AI Engagement Log template (printed and digital).
- Feedback Interrogation Sheet template.
- Role cards for the tribunal (six stakeholder roles).
- Bilingual glossary: professionalism, professionality, performativity, deprofessionalisation, autonomy, accountability, collegiality, professional learning community, policy borrowing.
- Timer, flipchart paper, markers.
- Hargreaves, A. (2000). Four ages of professionalism and professional learning. Teachers and Teaching, 6(2), 151–182. https://doi.org/10.1080/713698714
- Sachs, J. (2001). Teacher professional identity: Competing discourses, competing outcomes. Journal of Education Policy, 16(2), 149–161. https://doi.org/10.1080/02680930116819
- UNESCO/ILO (1966). Recommendation concerning the Status of Teachers. unesco.org
- OECD. TALIS — Teaching and Learning International Survey. oecd.org/education/talis
- AITSL. Australian Professional Standards for Teachers. aitsl.edu.au
- UK Department for Education. Teachers' Standards. gov.uk
- Hong Kong Education Bureau. Guidelines on Teachers' Professional Conduct. edb.gov.hk
- Ontario College of Teachers. Ethical Standards for the Teaching Profession. oct.ca
- Hypothesis — collaborative annotation platform. web.hypothes.is
Verify all links before distribution; institutional access may be required for journal articles.
Section 7
Differentiation & Inclusivity
Multiple entry points, tiered inquiry tasks, sentence starters and an AI Use Ladder so that a cohort of 90 with very different confidence levels can all meet the same critical standard.
- Multiple entry points: text-based (readings), visual (concept maps, card sorts), oral (tribunal), written (position statement) — students may demonstrate inquiry competence through any combination.
- Structured sentence starters for seminar dialogue: "My claim is ___ because the evidence from ___ suggests ___"; "I want to challenge that because ___".
- Quiet preparation space available for students who need to draft before speaking.
- Alternative to public speaking: students may submit a written tribunal statement in place of an oral opening statement.
- All PDFs screen-reader accessible and tagged; large-print and high-contrast versions of card-sort materials; captioned video materials.
- Extended time on the pre-test and post-test for students with registered accommodations.
- Lead a group's AI critique and present the group's verdicts to the class.
- Investigate teacher migration and the international recognition of qualifications as a professionalism issue.
- Explore decolonising critiques of "global" professionalism frameworks.
- Design a comparative rubric for evaluating national professional standards.
- Tier 3 inquiry task: critique the epistemological assumptions of the comparative method itself; examine teacher professionalism in a low-income or post-colonial context.
| Tier | Task |
|---|---|
| Tier 1 — foundational | Identify and describe the four discourses; locate one example of each in the Source Pack. |
| Tier 2 — core | Compare two contexts and evaluate one AI claim against a source. |
| Tier 3 — extension | Critique the epistemological assumptions of the comparative method itself; examine teacher professionalism in a low-income or post-colonial context. |
Students self-select a level; the minimum requirement for the portfolio is Level 2. This ensures students with low AI confidence are not disadvantaged while maintaining the critical engagement standard.
Use the Source Pack only. The critical reading task is identical; only the object of critique changes.
Use AI to generate ideas, but verify everything against named sources. Full log required for AI-derived claims.
Use AI to draft the comparison, with a full critical log documenting every accept, modify and reject decision.
- Scaffolding for struggling learners: AI generates simplified explanations of Sachs's four discourses; students must check the simplification against the original text and note what was lost.
- Extension for advanced learners: AI generates comparative accounts of teacher professionalism in under-represented contexts (for example Rwanda, Vietnam, Chile), which students then verify and critique.
- Accessibility accommodations via AI: AI-generated captions for video materials; text-to-speech for readings; translation support for multilingual students; AI-assisted summarisation of dense policy documents (with verification required).
Section 8
Reflection & Improvement
Success indicators tied to the Critical Inquiry Seminar, student feedback instruments, AI-specific reflection data, and the modification strategies that follow from each pattern of evidence.
- Proportion of students who revise their position between the opening and closing votes (target: ≥40%).
- Quality of questioning in the tribunal — the frequency of assumption-probing questions versus opinion assertions.
- Distribution of participation across all 15 groups: no group silent, no individual dominating.
- Quality of the Padlet claim-and-warrant statements — are warrants evident and sourced?
- Critical Inquiry Self-Evaluation Form (post-session, 5 min): "How well did I use evidence today? How well did I question assumptions — my own and others'? Did I change my mind, and why?"
- Mid-semester focus group with 8–10 volunteers on the seminar format.
Student feedback on AI integration
"Did AI tools enhance or distract from your learning? Were you able to maintain your own thinking? Which AI task was most valuable, and which felt like a distraction?"
Instructor reflection on AI effectiveness
"Did the AI integration achieve its intended goals? Where did students struggle with AI criticality? Did any students over-rely on AI despite the log requirement? Did the 'AI' tribunal role work as intended?"
Data on student critical engagement
- Percentage of AI claims correctly identified as inaccurate, biased or hallucinated.
- Distribution of accept / modify / reject verdicts across the cohort.
- Number of hallucinated citations caught versus missed.
- Correlation between Critical AI Engagement Log quality and overall portfolio performance.
- Instances where students overrode AI feedback, and the quality of their rationale.
| If you observe… | Then… |
|---|---|
| Position revision rates are low | Strengthen the cross-examination structure and introduce a mid-tribunal "evidence challenge" pause, drawing on Critical Inquiry principles of structured disputation. |
| AI critique is superficial | Replace the generic AI comparison prompt with a deliberately flawed, instructor-authored AI response containing known errors, so students practise detection before generating their own. |
| Group participation is uneven | Introduce assigned dialogic roles (questioner, evidence-checker, synthesiser, devil's advocate) rotating each segment, consistent with Social Constructivist role structuring. |
| The Source Pack is too dense for first-year students | Tier the Source Pack into "essential" and "extension" extracts, consistent with Scaffolded Instruction. |
| AI access is inequitable | Provide a curated set of pre-generated AI transcripts for students without reliable access, preserving the critical evaluation task while removing the access barrier. |
| The tribunal runs over time | Move Segment D drafting to a post-class asynchronous task with a same-day submission deadline, preserving the reflective momentum. |
Tick each task as you complete it. Your progress is saved in this browser and feeds the portfolio summary below.
Portfolio
Critical Reflection Portfolio
Everything you record here is stored locally on your device. Review it, then export it as evidence for your summative submission.
Record the specific decisions you made about AI output today. These entries become Portfolio Part D and the raw material for the "Critical AI Engagement" rubric criterion.
Nothing is uploaded anywhere — all data stays in your browser's local storage.