Draft Preliminary note on identifiability and PII review · click to expand

This appendix is being circulated to the committee as a draft. It contains material that has not yet completed a final review for personally identifiable information. Before the appendix is finalized for submission, every mention of a named individual will be reviewed against the taxonomy in §B.6.5: students named in instructor-produced materials will be anonymized unless explicit written consent for educational use is documented; named guest speakers will be retained as public professional identities with their professional context attached; Luma-platform workshop feedback will be reviewed for anonymization; and external journalism is retained as already published and consented.

Index
§B.2

How analytic autoethnography supplements design-based research

This section develops the supplementation relationship between the main document's primary methodology (design-based research with constructivist and self-determination-theory grounding) and this appendix's supplementary autoethnographic posture. The two methodologies operate at different levels and produce different kinds of contribution. Reading them together gives a fuller account of what my four iterations made possible to understand than either methodology alone.

B.2.1 What design-based research contributes

Design-based research, as McKenney and Reeves (2018) articulate it, is a methodology for designing educational interventions through iterative cycles in which the researcher specifies design conjectures, enacts them in real instructional settings, observes outcomes, and refines the conjectures across iterations. The main document of the dissertation operates under this methodology. The four course iterations are the iterative-design cycles; the curriculum-design principles (modularity, learner choice, continuous feedback) are what the DBR analysis surfaces; the comparison across the four iterations supports the principles' generalizability claims.

DBR is appropriate to this work. It licenses the iterative-refinement structure the iterations actually took; it surfaces principles that are practically useful to other curriculum designers; and it integrates the constructivist learning theory and self-determination theory that inform the curriculum's pedagogical design. The main document develops the DBR analysis at length.

B.2.2 What analytic autoethnography adds

What the DBR analysis developed in the main document does not surface is theoretical insight drawn from the instructor's reflexive analytic position on her own practice. Curriculum-design principles are practical outputs; they tell other curriculum designers what to do. They do not tell other researchers what the iterations reveal about how generative-AI pedagogy emerged as a field, how a pioneer instructor's practice operated under tool turnover, or what conceptual phenomena (like hallucination-as-pedagogy) the iterations exposed.

Analytic autoethnography (Anderson 2006) is the methodological framework that licenses these additional analytic claims. By treating my insider position as an analytic resource rather than as a confound to be controlled, autoethnography surfaces:

The supplementation is additive. The DBR analysis of the main document is unaffected by what this appendix does; what this appendix does is sit alongside it and extend the analytic reach.

B.2.3 What the iterations produced for both methodologies

The four iterations between Spring 2024 and September 2025 produced data of two kinds simultaneously. Curricular and operational data fed the DBR analysis: weekly module schedules, slide decks, assignments, the Luma feedback survey for Iteration 3, attendance and engagement metrics, syllabi. Reflective and contextual data fed (and feeds) the autoethnographic analysis: the Weekly Updates Prelim Document for Iteration 1, the cross-iteration newsletter and podcast, the CU RMACC webinar, the K-12 outreach corpus, external media coverage.

The two streams of data are not separate corpora; they overlap. The Canvas LMS exports (CV-1, CV-2) feed both analyses: they document curriculum design (DBR-relevant) and they document the iteration in the form a complete-member researcher can read autoethnographically. The slide decks across iterations (DK-1.FG, DK-1.W01, DK-2 series, DK-3, DK-4) are similar. What changes between the two methodologies is the analytic question asked of the same artifacts: "what curriculum-design principle does this iteration surface?" (DBR) versus "what theoretical claim does my reflexive position on this iteration license?" (analytic autoethnography).

B.2.4 Why analytic autoethnography is the right supplementary methodology

I considered three candidate supplementary methodologies before choosing analytic autoethnography. Each of the others would have offered something but did not fit the configuration as well.

Thematic coding of student interviews was the supplementary path the original proposal specified. It was not enacted because the iterations did not produce IRB-governed structured interview data. Other learner-facing data (the Luma feedback corpus, the student teach-out presentations, the UW KidsTeam co-design data) was generated, and the main document's DBR analysis reads it; but the data does not match what thematic coding presumes.

Evocative autoethnography (Ellis, Adams, and Bochner 2011) would have produced an affectively resonant first-person narrative of the teaching practice. The genre is established and well-developed. It would, however, sit awkwardly alongside the main document's DBR analysis; evocative narrative and DBR analysis operate from different epistemological commitments. The combined dissertation would read as two unrelated works rather than as a coherent supplementation.

Practitioner inquiry (Cochran-Smith and Lytle 2009) is closer in spirit to what this appendix does, and the appendix draws on it as background. As a primary supplementary methodology, however, practitioner inquiry typically operates at the same level as DBR (surfacing practical knowledge for teachers) and would therefore not add an additional analytic layer.

Analytic autoethnography offers what the supplementation requires: a methodology with its own rigor structure (Anderson's five criteria), its own established place in engineering education and HCI literatures, its own genre of contribution (theoretical findings from insider analytic position), and full compatibility with the DBR primary framing of the main document. Chapter B §B.3 below applies Anderson's criteria to the corpus.

B.2.5 What this supplementation does not claim

The supplementation does not claim that analytic autoethnography is necessary for the dissertation. The main document stands on its DBR analysis; the curriculum-design principles do not require autoethnographic support to be valid. What the supplementation claims is that the autoethnographic posture surfaces an additional layer of theoretical findings the DBR analysis developed in the main document does not surface, and that those findings are themselves valuable scholarly contributions.

The supplementation also does not claim methodological neutrality. Section B.7 names my investment in the work. The supplementation's purpose is to enrich the analytic record, not to provide a second independent test of the main document's claims.

With this supplementation framing established, §B.3 below applies Anderson's five criteria to my artifact corpus in detail.

Index